From joey at joescan.com Tue Apr 1 16:21:00 2003 From: joey at joescan.com (Joey Nelson) Date: Tue Apr 1 16:21:00 2003 Subject: Need help on supporting Fabiatech FB2510 mainboard. In-Reply-To: <20030331073609.GA19198@lea.ulyssis.org> Message-ID: <001d01c2f898$c3ecaae0$6501a8c0@pink> I have linuxbios partly working on this GX1/CS5530 based mainboard. I created a mainboard/fabiatech/fb2510 directory and copied the files from mainboard/advantech/pcm-5823. In Config, I changed all references to FDC37B72x to FDC37B78x (By the way superio/SMC/FDC37B78x/superio.c reports itself as FDC37B72x). Everything else I left the same. I based the build config off config.example, changing the payload from eepro100.ebi to rtl8139.elf (there was no make target for bin/rtl8139.ebi in Etherboot 5.1.7). I tried to run getpir on the FB2510 under the existing Award Bios, but it failed to find a valid PCI IRQ Routing table. I include the output below. I went ahead and built linuxbios (using the irq_tables.c from pcm-5823) and programmed the FB2510's flash with the resulting romimage. Booting with this romimage was quite successful up until configuring the com port. Booting freezes right after this point. ------ begin of linuxbios serial console output ------- Call init Enabling com device: 04 iobase = 0x03f8 irq=4 ------ end of linuxbios serial console output ------- Actually originally it would only print "io" from the last line. Looking in superio.c I found that the comments said initialization should be happening before enabling, but in the code it was happening the other way around. Looking in FDC7B72x it was done in the init then enable order with a delay(1) between the init and enable. So I went ahead and made that change in FDC7B72x/superio.c. After that I got the above output. I'm guessing this problem is because my irq routing table is not being properly configured. Is this a good guess? How else can I get the information to put in irq_table.c? ------ begin of getpir output ------- Opening memory... Found PCI IRQ Routing table signature at c0229d4c bytes from top of the memory Validating../ Version is:13429 Table size:33126 Invalid version Invalid table size (not a multiple of 16) Found PCI IRQ Routing table signature at c16744e0 bytes from top of the memory Validating../ Version is:13429 Table size:33126 Invalid version Invalid table size (not a multiple of 16) Found PCI IRQ Routing table signature at c1688bcc bytes from top of the memory Validating../ Version is:57116 Table size:16387 Invalid version Invalid table size (not a multiple of 16) Found PCI IRQ Routing table signature at c168a39c bytes from top of the memory Validating../ Version is:17863 Table size:476 Invalid version Invalid table size (not a multiple of 16) Found PCI IRQ Routing table signature at c1690e70 bytes from top of the memory Validating../ Version is:17863 Table size:476 Invalid version Invalid table size (not a multiple of 16) Found PCI IRQ Routing table signature at c1a62e3c bytes from top of the memory Validating../ Version is:17863 Table size:57552 Invalid version Closing memory No table for you... ------ end of getpir output ------- Joey Nelson From yenneo at oreka.com Wed Apr 2 08:26:01 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Wed Apr 2 08:26:01 2003 Subject: Flashing In-Reply-To: <20030331050854.11360.87288.Mailman@nwn.definitive.org> Message-ID: <200304021355.PAA95460@mailhub1.isdnet.net> What is the best way to re-flash an Epia? -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From engard.ferenc at innomed.hu Wed Apr 2 11:06:01 2003 From: engard.ferenc at innomed.hu (Fery) Date: Wed Apr 2 11:06:01 2003 Subject: newbie -- general question Message-ID: <3E8B1241.247B5C06@innomed.hu> Hi all! I am sorry if I am asking stupid, I have just falled into an embedded software project. So far I have only developed desktop applications, but I have a strong affection to embedded systems. :) So, the task I want to get through is as follows: We have an existing product (a medical bedside monitor) working on a x86-based system, the "kernel" is dos/dpmi, which, in fact, is used only to switch the processor to protected mode (no task management, (almost) no memory management, etc.). Now, we have to rewrite the program, as the hardware is changing, the code is unmanageable (it was not me, I am a fresh employee at the company :), and anyway. The new hardware's core will be the geode gx1 processor with the CS5530 companion chip, and we have an Advantech PCM-5820 mainboard with these chips to start developing and play with, and possibly use in the final system (if the self-made mainboard developing is too slow / too much money). I proposed to use a kernel under our new program, as we do not need to implement many low- and middle-level protocols (such as the tcp/ip stack, ethernet driver, etc.), maybe(?) task management, etc. As nobody used this approach at the company yet (me neither), there are no experiences how hard to port a kernel to a custom hardware, and what constraints place the kernel upon our program. I have a few years linux experience (but not hardware or kernel module programming), so naturally my first idea was linux or rtlinux. I have also found the linux4.tv project, which use quite similar hardware, downloaded the source, but I didn't get the point except that possibly we can get out the fb video driver source code for gx1. As currently this motherboard has quite slow bios boot-up, and my idea to use linux, I have found the linuxbios project. And, because there are many smart people here, I have some questions to you: 1. Can I use linuxbios to boot a single executable, which is analog to the current use of dpmi? E.g., I use linuxbios (beside that it boots up the hardware) to switch into protected mode, and I have a single binary what somehow gets some parameters (e.g., memory size), and runs in privilege level 0 (no kernel)? Can I use standard tools (like gcc) to compile and link such a beast? 2. Can I use linuxbios to boot RTLinux? 3. Can I use linux/rtlinux only as a "big lib" of device drivers and other stuff? I.e. the scheduler is at my hand, and rtlinux is only a process which runs when I let him run, and not too long either. :) 3. Do you know a good mailing list about rtlinux and embedded systems programming for a newbie? 4. Maybe are there other (possibly non-free) kernels which come into question? Sorry about the long and slightly offtopic post. Thanks: Circum From ts1 at cma.co.jp Thu Apr 3 00:59:00 2003 From: ts1 at cma.co.jp (SONE Takeshi) Date: Thu Apr 3 00:59:00 2003 Subject: Flashing In-Reply-To: <200304021355.PAA95460@mailhub1.isdnet.net> References: <20030331050854.11360.87288.Mailman@nwn.definitive.org> <200304021355.PAA95460@mailhub1.isdnet.net> Message-ID: <20030403062716.GA925@cma.co.jp> On Wed, Apr 02, 2003 at 03:47:29PM +0100, Philippe CABANNES wrote: > What is the best way to re-flash an Epia? I'm using BIOS Savior (RD1-PL) and flash_rom in the current LinuxBIOS tree to flash my EPIA 5000. -- Takeshi From steve at nexpath.com Thu Apr 3 01:03:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 3 01:03:00 2003 Subject: newbie -- general question In-Reply-To: <3E8B1241.247B5C06@innomed.hu> References: <3E8B1241.247B5C06@innomed.hu> Message-ID: <3E8BD933.9070003@nexpath.com> > 3. Do you know a good mailing list about rtlinux and embedded systems > programming for a newbie? > You might check this site for articles and information: http://www.linuxdevices.com/ I don't see any reason why you couldn't use linuxbios to start your own binary, compiled with gcc. I would recommend using linux though, and something like busybox (http://www.busybox.net/) and uclibc (http://www.uclibc.org/) will keep it small. I believe rtlinux is now a commercial product (?); you may not need it for a tightly controlled embedded system; maybe be a custom driver or two will do it. Just some ideas. -Steve From Luca.Pellanda at italtel.it Thu Apr 3 02:06:01 2003 From: Luca.Pellanda at italtel.it (Pellanda Luca) Date: Thu Apr 3 02:06:01 2003 Subject: Problems with etherboot + AA patch IDE Message-ID: I'm running as 2.11.90.0.29 and compiling etherboot 5.0.6 + the AA IDE boot patch found on the list causes problems: gcc -E -Wp,-Wall -DASK_BOOT=3 -DANS_DEFAULT=ANS_NETWORK -DMOTD -DIMAGE_MENU -DBA CKOFF_LIMIT=7 -DCONGESTED -DLINUXBIOS -DCONFIG_TSC_CURRTICKS -DCONSOLE_SERIAL - DCOMCONSOLE=0x3f8 -DCOMPRESERVE -DCONFIG_PCI_DIRECT -DELF_IMAGE -DPOLLED_IDE -O s -ffreestanding -fstrength-reduce -fomit-frame-pointer -mcpu=i386 -malign-jumps =1 -malign-loops=1 -malign-functions=1 -Wall -W -Wno-format -Wno-unused -DVERSIO N_MAJOR=5 -DVERSION_MINOR=0 -DVERSION=\"5.0.6\" -DRELOC=0x400000 start32.S | as -o bin32/start32.o {standard input}: Assembler messages: {standard input}:119: Warning: indirect ljmp without `*' {standard input}:235: Error: suffix or operands invalid for `ljmp' make: *** [bin32/start32.o] Error 1 I think that RELOC is not good value. Why ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From agnew at cs.umd.edu Thu Apr 3 02:51:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 3 02:51:00 2003 Subject: Problems with etherboot + AA patch IDE In-Reply-To: Message-ID: <20030403031742.J22502-100000@www.missl.cs.umd.edu> Instead of running "make", run "make bin32/sis900.ebi" or possibly "make bin32/mynetcard.ebi" if your board doesn't use a sis900 (not that it should matter, the network booting is left as a fall back if the polled ide booting doesn't work). - Adam Agnew On Thu, 3 Apr 2003, Pellanda Luca wrote: > > > I'm running as 2.11.90.0.29 and compiling etherboot 5.0.6 + the AA IDE boot > patch found on the list causes problems: > > gcc -E -Wp,-Wall -DASK_BOOT=3 -DANS_DEFAULT=ANS_NETWORK -DMOTD -DIMAGE_MENU > -DBA > CKOFF_LIMIT=7 -DCONGESTED -DLINUXBIOS -DCONFIG_TSC_CURRTICKS > -DCONSOLE_SERIAL - > DCOMCONSOLE=0x3f8 -DCOMPRESERVE -DCONFIG_PCI_DIRECT -DELF_IMAGE > -DPOLLED_IDE -O > s -ffreestanding -fstrength-reduce -fomit-frame-pointer -mcpu=i386 > -malign-jumps > =1 -malign-loops=1 -malign-functions=1 -Wall -W -Wno-format -Wno-unused > -DVERSIO > N_MAJOR=5 -DVERSION_MINOR=0 -DVERSION=\"5.0.6\" -DRELOC=0x400000 start32.S > | as > -o bin32/start32.o > {standard input}: Assembler messages: > {standard input}:119: Warning: indirect ljmp without `*' > {standard input}:235: Error: suffix or operands invalid for `ljmp' > make: *** [bin32/start32.o] Error 1 > I think that RELOC is not good value. Why ? > > From Luca.Pellanda at italtel.it Thu Apr 3 03:31:01 2003 From: Luca.Pellanda at italtel.it (Pellanda Luca) Date: Thu Apr 3 03:31:01 2003 Subject: R: Problems with etherboot + AA patch IDE Message-ID: When I run " make bin32/3c900-tpo.ebi" it's ok but if I try to run "make bin32/3c900-tpo.fd0" I see the same error. Regards, Luca -----Messaggio originale----- Da: Adam Agnew [mailto:agnew at cs.umd.edu] Inviato: gioved? 3 aprile 2003 10.21 A: Pellanda Luca Cc: 'linuxbios at clustermatic.org' Oggetto: Re: Problems with etherboot + AA patch IDE Instead of running "make", run "make bin32/sis900.ebi" or possibly "make bin32/mynetcard.ebi" if your board doesn't use a sis900 (not that it should matter, the network booting is left as a fall back if the polled ide booting doesn't work). - Adam Agnew On Thu, 3 Apr 2003, Pellanda Luca wrote: > > > I'm running as 2.11.90.0.29 and compiling etherboot 5.0.6 + the AA IDE boot > patch found on the list causes problems: > > gcc -E -Wp,-Wall -DASK_BOOT=3 -DANS_DEFAULT=ANS_NETWORK -DMOTD -DIMAGE_MENU > -DBA > CKOFF_LIMIT=7 -DCONGESTED -DLINUXBIOS -DCONFIG_TSC_CURRTICKS > -DCONSOLE_SERIAL - > DCOMCONSOLE=0x3f8 -DCOMPRESERVE -DCONFIG_PCI_DIRECT -DELF_IMAGE > -DPOLLED_IDE -O > s -ffreestanding -fstrength-reduce -fomit-frame-pointer -mcpu=i386 > -malign-jumps > =1 -malign-loops=1 -malign-functions=1 -Wall -W -Wno-format -Wno-unused > -DVERSIO > N_MAJOR=5 -DVERSION_MINOR=0 -DVERSION=\"5.0.6\" -DRELOC=0x400000 start32.S > | as > -o bin32/start32.o > {standard input}: Assembler messages: > {standard input}:119: Warning: indirect ljmp without `*' > {standard input}:235: Error: suffix or operands invalid for `ljmp' > make: *** [bin32/start32.o] Error 1 > I think that RELOC is not good value. Why ? > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From pyro at linuxlabs.com Thu Apr 3 08:16:01 2003 From: pyro at linuxlabs.com (steven james) Date: Thu Apr 3 08:16:01 2003 Subject: newbie -- general question In-Reply-To: <3E8B1241.247B5C06@innomed.hu> Message-ID: Greetings, Depending on the requirements of the device, you could consider running your app as init from the linux kernel. Possibly setting SCHED_FIFO or SCHED_RR will be needed for more responsiveness. LinuxBIOS will certainly improve the boot time and reliability. G'day, sjames On Wed, 2 Apr 2003, Fery wrote: > Hi all! > > I am sorry if I am asking stupid, I have just falled into an embedded > software project. So far I have only developed desktop applications, but > I have a strong affection to embedded systems. :) > > So, the task I want to get through is as follows: > > We have an existing product (a medical bedside monitor) working on a > x86-based system, the "kernel" is dos/dpmi, which, in fact, is used only > to switch the processor to protected mode (no task management, (almost) > no memory management, etc.). Now, we have to rewrite the program, as the > hardware is changing, the code is unmanageable (it was not me, I am a > fresh employee at the company :), and anyway. > > The new hardware's core will be the geode gx1 processor with the CS5530 > companion chip, and we have an Advantech PCM-5820 mainboard with these > chips to start developing and play with, and possibly use in the final > system (if the self-made mainboard developing is too slow / too much > money). > > I proposed to use a kernel under our new program, as we do not need to > implement many low- and middle-level protocols (such as the tcp/ip > stack, ethernet driver, etc.), maybe(?) task management, etc. As nobody > used this approach at the company yet (me neither), there are no > experiences how hard to port a kernel to a custom hardware, and what > constraints place the kernel upon our program. > > I have a few years linux experience (but not hardware or kernel module > programming), so naturally my first idea was linux or rtlinux. I have > also found the linux4.tv project, which use quite similar hardware, > downloaded the source, but I didn't get the point except that possibly > we can get out the fb video driver source code for gx1. > > As currently this motherboard has quite slow bios boot-up, and my idea > to use linux, I have found the linuxbios project. And, because there are > many smart people here, I have some questions to you: > > 1. Can I use linuxbios to boot a single executable, which is analog to > the current use of dpmi? E.g., I use linuxbios (beside that it boots up > the hardware) to switch into protected mode, and I have a single binary > what somehow gets some parameters (e.g., memory size), and runs in > privilege level 0 (no kernel)? Can I use standard tools (like gcc) to > compile and link such a beast? > > 2. Can I use linuxbios to boot RTLinux? > > 3. Can I use linux/rtlinux only as a "big lib" of device drivers and > other stuff? I.e. the scheduler is at my hand, and rtlinux is only a > process which runs when I let him run, and not too long either. :) > > 3. Do you know a good mailing list about rtlinux and embedded systems > programming for a newbie? > > 4. Maybe are there other (possibly non-free) kernels which come into > question? > > Sorry about the long and slightly offtopic post. > > Thanks: > Circum > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From ebiederman at lnxi.com Thu Apr 3 16:33:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 3 16:33:01 2003 Subject: Quick Question, If you Please In-Reply-To: <1049324256.24412.2056.camel@yost> References: <1049324256.24412.2056.camel@yost> Message-ID: I am swamped at the moment so I will redirect this to the LinuxBIOS mailing list and see you get some answers. Eric Kevin Takacs writes: > Hello, > > I came across your informative article at: > http://www.linuxjournal.com/article.php?sid=4888 . > > I read with some interest as a potential client has approached us with > installing a flavor of Linux on an embedded device (sis550). We're > primarily a Java web app shop and don't know where to begin looking to > see if this is outside the realm of our skills. I've done countless > installs of various distros, but have never recompiled kernels and the > like. > > Would you mind suggesting a few links for a beginner like me? Or > possibly sketching what's involved in taking the piece of hardware and > install Linux on it. For example, do you create a file system on a > flash card? How? Are there any vanilla (pre-built) solutions > particular chip sets? > > Thank you in advance for any information you could supply. > > Regards, > Kevin > -- > Kevin Takacs > Vice President, Technical Services > The Berndt Group, Ltd. > www.berndtgroup.com > > Always Thinking. > > 2850 N Charles St, 2nd Fl > Baltimore, MD 21218 > Email: kevin at berndtgroup.net > Voice: (410) 889-5854 > Fax: (410) 889-5904 From ijpriya at hotmail.com Thu Apr 3 21:24:01 2003 From: ijpriya at hotmail.com (Devi Priya) Date: Thu Apr 3 21:24:01 2003 Subject: BIOS Message-ID: Hello, I am working on a project on personal video recorder based on National's geode sc1200 chip. I have been assigned the task of writing BIOS for it. I have the xpress loader from NSC. This has the mainboard initialization part. My device has linux OS. It has other external peripherals like Audio codec, audio video decoder, IR, TV tuner etc. Should I have to initialize these external peripherals too in my BIOS program? _________________________________________________________________ Vrroooom?? Fasten your seatbelts. http://server1.msn.co.in/msnSpecials/formula2003/index.asp Get set for F1 2003 From robb at 20tauntongrove.net Sun Apr 6 16:17:01 2003 From: robb at 20tauntongrove.net (Robert Boardman) Date: Sun Apr 6 16:17:01 2003 Subject: Cyrix 5520 Chipset Message-ID: <20030406214540.63f7cc33.robb@20tauntongrove.net> Hi List This is my first post this looks like exactly what I'm looking for (hopefully) just one question Does the linuxbios project support the cyrix 5520 chipset? reguards robb From rminnich at lanl.gov Sun Apr 6 21:27:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sun Apr 6 21:27:01 2003 Subject: Cyrix 5520 Chipset In-Reply-To: <20030406214540.63f7cc33.robb@20tauntongrove.net> Message-ID: On Sun, 6 Apr 2003, Robert Boardman wrote: > Does the linuxbios project support the cyrix 5520 chipset? does cyrix exist any more? kron From rminnich at lanl.gov Sun Apr 6 22:11:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sun Apr 6 22:11:01 2003 Subject: IRQ assignment on my IBM X24 Message-ID: this seems really poor bios design: @mini rminnich]# lspci -v -v | grep Interrupt Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin C routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Pretty much everyone goes to IRQ 11. And a bunch of Interrupts go unused: CPU0 0: 3381694 XT-PIC timer 1: 13873 XT-PIC keyboard 2: 0 XT-PIC cascade 8: 1 XT-PIC rtc 11: 77842 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, wlan0, Ricoh Co Ltd RL5c476 II, Ricoh Co Ltd RL5c476 II (#2) 12: 360259 XT-PIC PS/2 Mouse 14: 85559 XT-PIC ide0 NMI: 0 ERR: 0 Is there any reason a BIOS would do this? ron From bari at onelabs.com Sun Apr 6 22:13:01 2003 From: bari at onelabs.com (Bari Ari) Date: Sun Apr 6 22:13:01 2003 Subject: Cyrix 5520 Chipset References: Message-ID: <3E90E608.5010409@onelabs.com> The Cyrix 5520 was the chipset before the NSC 5530. The registers are similar but not compatible with the 5530 code. IIRC the 5520 came out back in 1997 about when NSC took over Cyrix. Bari ron minnich wrote: >On Sun, 6 Apr 2003, Robert Boardman wrote: > > > >>Does the linuxbios project support the cyrix 5520 chipset? >> >> > >does cyrix exist any more? > >kron > >_______________________________________________ >Linuxbios mailing list >Linuxbios at clustermatic.org >http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > From ebiederman at lnxi.com Mon Apr 7 00:53:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 7 00:53:00 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: References: Message-ID: ron minnich writes: > this seems really poor bios design: > @mini rminnich]# lspci -v -v | grep Interrupt > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin C routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > > Pretty much everyone goes to IRQ 11. > > And a bunch of Interrupts go unused: > > CPU0 > 0: 3381694 XT-PIC timer > 1: 13873 XT-PIC keyboard > 2: 0 XT-PIC cascade > 8: 1 XT-PIC rtc > 11: 77842 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, > wlan0, Ricoh Co Ltd RL5c476 II, Ricoh Co Ltd RL5c476 II (#2) > 12: 360259 XT-PIC PS/2 Mouse > 14: 85559 XT-PIC ide0 > NMI: 0 > ERR: 0 > > > Is there any reason a BIOS would do this? The motherboard is a lousy design. Only vary rarely have I not seen the basic irq assignments come down to traces on the motherboard. Eric From ollie at sis.com.tw Mon Apr 7 03:27:01 2003 From: ollie at sis.com.tw (ollie lho) Date: Mon Apr 7 03:27:01 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: References: Message-ID: <1049701789.7768.1.camel@ollie> On Mon, 2003-04-07 at 10:40, ron minnich wrote: > this seems really poor bios design: > @mini rminnich]# lspci -v -v | grep Interrupt > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin C routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > > Pretty much everyone goes to IRQ 11. > > And a bunch of Interrupts go unused: > > CPU0 > 0: 3381694 XT-PIC timer > 1: 13873 XT-PIC keyboard > 2: 0 XT-PIC cascade > 8: 1 XT-PIC rtc > 11: 77842 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, > wlan0, Ricoh Co Ltd RL5c476 II, Ricoh Co Ltd RL5c476 II (#2) > 12: 360259 XT-PIC PS/2 Mouse > 14: 85559 XT-PIC ide0 > NMI: 0 > ERR: 0 > > Wow !!! Is it the common case for IBM notebooks ?? I though they were built for quality !! -- ollie lho From stuge-linuxbios at cdy.org Mon Apr 7 03:43:00 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Mon Apr 7 03:43:00 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: <1049701789.7768.1.camel@ollie> References: <1049701789.7768.1.camel@ollie> Message-ID: <20030407080442.GB27884@foo.birdnet.se> On Mon, Apr 07, 2003 at 03:49:50PM +0800, ollie lho wrote: > On Mon, 2003-04-07 at 10:40, ron minnich wrote: > > this seems really poor bios design: > > @mini rminnich]# lspci -v -v | grep Interrupt [..snip..] > > > > Pretty much everyone goes to IRQ 11. > > Wow !!! Is it the common case for IBM notebooks ?? I though they were > built for quality !! My T20 looks just the same: $ lspci -vvv|grep Interr Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin D routed to IRQ 11 Interrupt: pin ? routed to IRQ 9 /* ACPI */ Interrupt: pin A routed to IRQ 11 $ cat /proc/interrupts CPU0 0: 5081475 XT-PIC timer 1: 188937 XT-PIC keyboard 2: 0 XT-PIC cascade 8: 425349 XT-PIC rtc 11: 1550126 XT-PIC cs46xx, usb-uhci, i82365, eth0 12: 751324 XT-PIC PS/2 Mouse 14: 138703 XT-PIC ide0 15: 5 XT-PIC ide1 //Peter From karl at kalle.csb.ki.se Mon Apr 7 04:16:00 2003 From: karl at kalle.csb.ki.se (Karl Hammar) Date: Mon Apr 7 04:16:00 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: Message from ron minnich of "Sun, 06 Apr 2003 20:40:19 MDT." References: Message-ID: <200304070842.h378g6S3001127@opal.aspo.lcl> > > this seems really poor bios design: > @mini rminnich]# lspci -v -v | grep Interrupt > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin C routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin B routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > Interrupt: pin A routed to IRQ 11 > > Pretty much everyone goes to IRQ 11. > > And a bunch of Interrupts go unused: > > CPU0 > 0: 3381694 XT-PIC timer > 1: 13873 XT-PIC keyboard > 2: 0 XT-PIC cascade > 8: 1 XT-PIC rtc > 11: 77842 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, > wlan0, Ricoh Co Ltd RL5c476 II, Ricoh Co Ltd RL5c476 II (#2) > 12: 360259 XT-PIC PS/2 Mouse > 14: 85559 XT-PIC ide0 > NMI: 0 > ERR: 0 > > > Is there any reason a BIOS would do this? > > ron ... You can go into your laptops bios and reassign them if you don't like that setup, I had the same setting in my tpx20. Is there anyone who knows if that setup is bad or not? Does it do any difference? The things I can think of is: . drivers don't like to share interrupts, possible resulting in errors I have seen this previously, but not tested it resently . I have heard something like: "to use the irq 0-15 you have to access the interrupt controller (8259) and to that end you have to access the ISA bus, which means slow access" Is that true? The only hint I can find is: $ grep -A3 'Interrupt ReQuest' /usr/src/kernel-source-2.4.20/Documentation/Configure.help (there is a small number of Interrupt ReQuest lines in your computer that are used by the attached devices to gain the CPU's attention -- often a source of trouble if two devices are mistakenly configured to use the same IRQ). The program procinfo to display some $ grep -A2 distributing /usr/src/kernel-source-2.4.20/Documentation/i386/IO-APIC.txt between the PIRQ lines. (distributing IRQ sources properly is not a necessity, PCI IRQs can be shared at will, but it's a good for performance to have non shared interrupts). Slot5 should be used for videocards, they $ grep -A16 'Level-Triggered Interrupts' /usr/src/kernel-source-2.4.20/Documentation/mca.txt Level-Triggered Interrupts ========================== Because MCA uses level-triggered interrupts, a few problems arise with what might best be described as the ISA mindset and its effects on drivers. These sorts of problems are expected to become less common as more people use shared IRQs on PCI machines. In general, an interrupt must be acknowledged not only at the ICU (which is done automagically by the kernel), but at the device level. In particular, IRQ 0 must be reset after a timer interrupt (now done in arch/i386/kernel/time.c) or the first timer interrupt hangs the system. There were also problems with the 1.3.x floppy drivers, but that seems to have been fixed. IRQs are also shareable, and most MCA-specific devices should be coded with shared IRQs in mind. $ grep -A1 'SA_SHIRQ' /usr/src/kernel-source-2.4.20/Documentation/pci.txt All interrupt handlers should be registered with SA_SHIRQ and use the devid to map IRQs to devices (remember that all PCI interrupts are shared). $ grep -A1 'Switching between' /usr/src/kernel-source-2.4.20/arch/i386/kernel/i8259.c * This function assumes to be called rarely. Switching between * 8259A registers is slow. $ grep -A1 'interr' /usr/src/kernel-source-2.4.20/arch/i386/kernel/io_apic.c | grep pola /* EISA interrupts are always polarity zero and can be edge or level /* ISA interrupts are always polarity zero edge triggered, /* PCI interrupts are always polarity one level triggered, /* MCA interrupts are always polarity zero level triggered, Are there any other issues abut irq assignments? BTW, does things crash on you, do you get worse performance, can't different drivers function with the same irq, or what? Regards, /Karl ----------------------------------------------------------------------- Karl Hammar Asp? Data karl at kalle.csb.ki.se Lilla Asp? 2340 Networks S-742 94 ?sthammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting ----------------------------------------------------------------------- From rminnich at lanl.gov Mon Apr 7 09:52:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 7 09:52:00 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: Message-ID: On 6 Apr 2003, Eric W. Biederman wrote: > ron minnich writes: > > > this seems really poor bios design: > > @mini rminnich]# lspci -v -v | grep Interrupt > > Interrupt: pin A routed to IRQ 11 > > Interrupt: pin B routed to IRQ 11 > > Interrupt: pin C routed to IRQ 11 > > Interrupt: pin A routed to IRQ 11 > > Interrupt: pin B routed to IRQ 11 > > Interrupt: pin B routed to IRQ 11 > > Interrupt: pin B routed to IRQ 11 > > Interrupt: pin A routed to IRQ 11 > > Interrupt: pin A routed to IRQ 11 > > Interrupt: pin B routed to IRQ 11 > > Interrupt: pin A routed to IRQ 11 > > Interrupt: pin A routed to IRQ 11 > > > > Pretty much everyone goes to IRQ 11. > > > > And a bunch of Interrupts go unused: > > > > CPU0 > > 0: 3381694 XT-PIC timer > > 1: 13873 XT-PIC keyboard > > 2: 0 XT-PIC cascade > > 8: 1 XT-PIC rtc > > 11: 77842 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, > > wlan0, Ricoh Co Ltd RL5c476 II, Ricoh Co Ltd RL5c476 II (#2) > > 12: 360259 XT-PIC PS/2 Mouse > > 14: 85559 XT-PIC ide0 > > NMI: 0 > > ERR: 0 > > > > > > Is there any reason a BIOS would do this? > > The motherboard is a lousy design. Only vary rarely > have I not seen the basic irq assignments come down to traces > on the motherboard. what's worse is the PIR table claims that lots more interrupts are available. Sad, really. ron From rminnich at lanl.gov Mon Apr 7 09:55:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 7 09:55:01 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: <200304070842.h378g6S3001127@opal.aspo.lcl> Message-ID: On Mon, 7 Apr 2003, Karl Hammar wrote: > BTW, does things crash on you, do you get worse performance, can't > different drivers function with the same irq, or what? > I am looking at the interrupt-assignment problem again and was checking what my x24 did. ron From ebiederman at lnxi.com Mon Apr 7 14:23:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 7 14:23:01 2003 Subject: romcc Message-ID: After a lot of pound the code generator seems stable, and I can take a break to release romcc. Currently it appears that any time it completes without error the code generated is correct. And most importantly with a few minor tweaks it compiles the code written for l440gx memory initialization, before the compiler was written. Currently ``constant propagation'', ineffectual code elimination (code that does not affect the output), a good register allocator and always inlining are the only optimizations implemented. On the wishlist are: - using the mmx & sse registers on x86. - x86-64 and Itanium ports. - struct support - non string constant initializers. - A bibliography. But it should be good enough to start playing with. The version gets to be 1.0 when the compiler is useful enough to compile a LinuxBIOS port. ftp://ftp.lnxi.com/pub/src/linuxbios/romcc-0.21.tar.gz Eric From ebiederman at lnxi.com Mon Apr 7 14:24:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 7 14:24:00 2003 Subject: romcc Message-ID: After a lot of pound the code generator seems stable, and I can take a break to release romcc. Currently it appears that any time it completes without error the code generated is correct. And most importantly with a few minor tweaks it compiles the code written for l440gx memory initialization, before the compiler was written. Currently ``constant propagation'', ineffectual code elimination (code that does not affect the output), a good register allocator and always inlining are the only optimizations implemented. On the wishlist are: - using the mmx & sse registers on x86. - x86-64 and Itanium ports. - struct support - non string constant initializers. - A bibliography. But it should be good enough to start playing with. The version gets to be 1.0 when the compiler is useful enough to compile a LinuxBIOS port. ftp://ftp.lnxi.com/pub/src/linuxbios/romcc-0.21.tar.gz Eric From spirit at reactor.ru Mon Apr 7 18:58:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Mon Apr 7 18:58:00 2003 Subject: romcc In-Reply-To: References: Message-ID: <187101614744.20030408032758@reactor.ru> Hello Eric, Monday, April 7, 2003, 10:54:54 PM, you wrote: EWB> After a lot of pound the code generator seems stable, and I can take EWB> a break to release romcc. May I dare to ask what romcc is? :) With best regards, Alexander mailto:spirit at reactor.ru P.S. To all the developers: I noticed "David Woodhouse" in the list of developers at linuxbios.org. Is that the guy from Tyan or just someone with the same name? I knew one guy from Tyan by the name of Dave Woodhouse. So I wonder if that's him... From ebiederman at lnxi.com Mon Apr 7 19:45:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 7 19:45:01 2003 Subject: romcc In-Reply-To: <187101614744.20030408032758@reactor.ru> References: <187101614744.20030408032758@reactor.ru> Message-ID: Alexander Amelkin writes: > Hello Eric, > > Monday, April 7, 2003, 10:54:54 PM, you wrote: > > EWB> After a lot of pound the code generator seems stable, and I can take ^^^^^== pounding ( I should learn english :) > EWB> a break to release romcc. > > May I dare to ask what romcc is? :) Sorry I guess I should have been a little more clear. There has been a long standing issue with LinuxBIOS that much of the code needs to be written in assembly because it is run before memory is initialized, and so there is not a stack available. romcc is a small c compiler that generates code that runs without a stack. The full feature set of C is not available but as much as is easy and is possible is. The goal is to make maintenance easier, and code reviews of the memory setup easier. Eric From ollie at sis.com.tw Mon Apr 7 20:17:00 2003 From: ollie at sis.com.tw (ollie lho) Date: Mon Apr 7 20:17:00 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: <200304070842.h378g6S3001127@opal.aspo.lcl> References: <200304070842.h378g6S3001127@opal.aspo.lcl> Message-ID: <1049761997.7768.10.camel@ollie> On Mon, 2003-04-07 at 16:42, Karl Hammar wrote: > > You can go into your laptops bios and reassign them if you don't like > that setup, I had the same setting in my tpx20. > > Is there anyone who knows if that setup is bad or not? Does it do any > difference? > Theoretically, this will give you larger interrupt latency. When the kernel is handling some specific IRQ line, it disable further interrupts on that line (but enable all other lines). If there are other devices sharing the same line, their interrupts can not be served by the kernel until the first one is completed. On the other hand, if a device is on the other IRQ line, the kernel can service the interrupt by entering another level of interrupt context. (BTW, this is the case in 2.4.x kernel, I am not sure if it is true for 2.5.x.) -- ollie lho From rminnich at lanl.gov Mon Apr 7 23:21:01 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 7 23:21:01 2003 Subject: your mail (fwd) Message-ID: interesting ... from Chuck Cranor: BTW, not direcly related to the paper, but i found another open source BIOS if you are interested: ftp://bakfull.campus.luth.se/bios/nbsd-bios.tar.gz written by the guy who does NetBSD/vax support :) chuck From stuge-linuxbios at cdy.org Mon Apr 7 23:38:01 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Mon Apr 7 23:38:01 2003 Subject: your mail (fwd) In-Reply-To: References: Message-ID: <20030408035906.GD22610@foo.birdnet.se> On Mon, Apr 07, 2003 at 09:50:45PM -0600, Ronald G. Minnich wrote: > BTW, not direcly related to the paper, but i found another open > source BIOS if you are interested: The structure looks quite similar to LinuxBIOS', the hardware support doesn't come close however. (nbsd-bios only has support for the Asus P5A.) It does simple PCI configuration but nothing that will handle a bridge as it seems. Pretty nice code base though! I say we should try to include it, NetBSD will be another feather in the LinuxBIOS cap. > ftp://bakfull.campus.luth.se/bios/nbsd-bios.tar.gz FYI, bakfull means hung over in swedish. None of that in the code though. //Peter From kewalbk at hotmail.com Tue Apr 8 00:45:00 2003 From: kewalbk at hotmail.com (kewal karavinkoppa) Date: Tue Apr 8 00:45:00 2003 Subject: help!!! Message-ID: hi, i need some help reguarding in choosing the motherboard for the linuxbios project.. please help me...... kewal b k, contact no: 91 0831 455116 _________________________________________________________________ Find a partner. For life. http://www.shaadi.com/ptnr.php?ptnr=hmlql Meet at Shaadi.com From terryc at tyanchina.com Tue Apr 8 02:22:01 2003 From: terryc at tyanchina.com (Terry B. Chen) Date: Tue Apr 8 02:22:01 2003 Subject: 533Mhz FSB of E7501 Message-ID: <441383BA85E81D48964152E537886C9D2BF7ED@mail.sh.corp.tyan.com> Dear all; I am trying the Tyan 2723(E7501), but I find it will report spot_check error if the board run in 533 FSB. But if the CPU or the memory is 400 FSB it will be ok. Does anyone else meet the same problem? Best regards Terry chen From Luca.Pellanda at italtel.it Tue Apr 8 04:32:01 2003 From: Luca.Pellanda at italtel.it (Pellanda Luca) Date: Tue Apr 8 04:32:01 2003 Subject: rombios with linuxbios Message-ID: Hi list, I would like to use rombios.c with linuxbios. I have seen that rombios.c includes all bios services. I have compiled rombios.c an I have produced rombios.bin. Questions: 1) How do I install rommbios.bin ? My romimage is 64k, rombios is 64k and both of them are pure binary files. 2)Can linuxbios call rombios.c services even if they are 2 binary files ? regards, Luca -------------- next part -------------- An HTML attachment was scrubbed... URL: From robos at muon.de Tue Apr 8 09:15:01 2003 From: robos at muon.de (Robos) Date: Tue Apr 8 09:15:01 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: <1049701789.7768.1.camel@ollie> References: <1049701789.7768.1.camel@ollie> Message-ID: <20030408134450.GF22925@muon.de> > Wow !!! Is it the common case for IBM notebooks ?? I though they were > built for quality !! You wish! This is on my R31 lspci -v -v|grep Interrupt Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin C routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 and cat /proc/interrupts CPU0 0: 6331560 XT-PIC timer 1: 71762 XT-PIC keyboard 2: 0 XT-PIC cascade 3: 1251424 XT-PIC prism2_cs 11: 428891 XT-PIC Intel ICH3, usb-uhci, usb-uhci, usb-uhci, i82365, eth0 12: 503886 XT-PIC PS/2 Mouse 14: 211297 XT-PIC ide0 15: 447 XT-PIC ide1 and attached that of my girlfriends r32 BTW Ron, if you (by chance) figure out the base address of the i2c bus in the X24 please mail me, I'm trying to get s-video out working with my graphics chip (i830) and need both the i2c and i830 base address for that (must be in the range of 0x05014) Cheers Robos > -- > ollie lho > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- Robos - gpg --recv-keys --keyserver blackhole.pca.dfn.de 6EEADA09 It's GNU/Linux dammit! F U M $ ! -------------- next part -------------- Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin C routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin B routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 Interrupt: pin A routed to IRQ 11 CPU0 0: 30697 XT-PIC timer 1: 471 XT-PIC keyboard 2: 0 XT-PIC cascade 8: 1 XT-PIC rtc 11: 38666 XT-PIC usb-uhci, usb-uhci, usb-uhci, eth0, eth1, Texas Instruments PCI1410 PC card Cardbus Controller, Intel ICH3, radeon at PCI:1:0:0 12: 2424 XT-PIC PS/2 Mouse 14: 9264 XT-PIC ide0 15: 32 XT-PIC ide1 NMI: 0 LOC: 0 ERR: 0 MIS: 0 From ebenard at free.fr Tue Apr 8 09:44:01 2003 From: ebenard at free.fr (Eric BENARD / Free) Date: Tue Apr 8 09:44:01 2003 Subject: IRQ assignment on my IBM X24 In-Reply-To: <20030408134450.GF22925@muon.de> References: <1049701789.7768.1.camel@ollie> <20030408134450.GF22925@muon.de> Message-ID: <200304081613.05594.ebenard@free.fr> Le Mardi 08 Avril 2003 15:44, Robos a ?crit : > > Wow !!! Is it the common case for IBM notebooks ?? I though they were > > built for quality !! > > You wish! This is on my R31 quite the same on a Dell Inspiron 8000 ;-) cat /proc/interrupts CPU0 0: 2502535 XT-PIC timer 1: 17909 XT-PIC keyboard 2: 0 XT-PIC cascade 5: 3512 XT-PIC Maestro3(i) 8: 1 XT-PIC rtc 9: 0 XT-PIC acpi 10: 70036 XT-PIC Texas Instruments PCI4451 PC card Cardbus Controller, Texas Instruments PCI4451 PC card Cardbus Controller (#2), ohci1394, usb-uhci, eth0 11: 1502174 XT-PIC nvidia 12: 134574 XT-PIC PS/2 Mouse 14: 50689 XT-PIC ide0 NMI: 0 LOC: 0 ERR: 0 MIS: 0 From adam at cfar.umd.edu Tue Apr 8 09:47:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Tue Apr 8 09:47:01 2003 Subject: rombios with linuxbios In-Reply-To: Message-ID: <20030408101705.M47872-100000@www.missl.cs.umd.edu> are you talking about Bochs BIOS? If so have a look at the ADLO project which does just that. It in the LinuxBIOS sources. > I would like to use rombios.c with linuxbios. > I have seen that rombios.c includes all bios services. > I have compiled rombios.c an I have produced rombios.bin. > Questions: > 1) How do I install rommbios.bin ? My romimage is 64k, > rombios is 64k and both of them are pure binary files. > 2)Can linuxbios call rombios.c services even if they are 2 binary files ? no, nor it has any need to. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rminnich at lanl.gov Tue Apr 8 10:36:00 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Tue Apr 8 10:36:00 2003 Subject: rombios with linuxbios In-Reply-To: Message-ID: sorry, what is rombios.c? is that my old thing or something new? ron From Luca.Pellanda at italtel.it Tue Apr 8 10:40:00 2003 From: Luca.Pellanda at italtel.it (Pellanda Luca) Date: Tue Apr 8 10:40:00 2003 Subject: R: rombios with linuxbios Message-ID: Sorry, I'm not very precise. rombios.c is Bochs rombios.c. Regards, Luca -----Messaggio originale----- Da: Ronald G. Minnich [mailto:rminnich at lanl.gov] Inviato: marted? 8 aprile 2003 17.06 A: Pellanda Luca Cc: 'linuxbios at clustermatic.org' Oggetto: Re: rombios with linuxbios sorry, what is rombios.c? is that my old thing or something new? ron -------------- next part -------------- An HTML attachment was scrubbed... URL: From pyro at linuxlabs.com Tue Apr 8 10:46:01 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 8 10:46:01 2003 Subject: 533Mhz FSB of E7501 In-Reply-To: <441383BA85E81D48964152E537886C9D2BF7ED@mail.sh.corp.tyan.com> Message-ID: Greetings, Most likely, there is an undocumented register issue in the northbridge. With the OEM BIOS, can you send me an lspci with a 400 FSB and a 533 FSB. That should narrow things down. G'day, sjames On Tue, 8 Apr 2003, Terry B. Chen wrote: > Dear all; > I am trying the Tyan 2723(E7501), but I find it will report > spot_check error if the board run in 533 FSB. But if the CPU or the > memory is 400 FSB it will be ok. Does anyone else meet the same problem? > > Best regards > Terry chen > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From pyro at linuxlabs.com Tue Apr 8 10:55:00 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 8 10:55:00 2003 Subject: 533Mhz FSB of E7501 In-Reply-To: <441383BA85E81D48964152E537886C9D2BF7ED@mail.sh.corp.tyan.com> Message-ID: Greetings, oops, I should have specified, lspci -xxx G'day, sjames On Tue, 8 Apr 2003, Terry B. Chen wrote: > Dear all; > I am trying the Tyan 2723(E7501), but I find it will report > spot_check error if the board run in 533 FSB. But if the CPU or the > memory is 400 FSB it will be ok. Does anyone else meet the same problem? > > Best regards > Terry chen > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From steve at nexpath.com Tue Apr 8 10:59:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Tue Apr 8 10:59:00 2003 Subject: romcc In-Reply-To: References: Message-ID: <3E92EC9F.5090902@nexpath.com> Eric W. Biederman wrote: > After a lot of pound the code generator seems stable, and I can take > a break to release romcc. > I gave it a quick test drive, looks pretty good. I want to put together a complete rom so I can actually execute it, but I have other things on the plate right now, will do so in a week or two I hope. Is the plan to re-code the assy that exists now, or only added code? If yes, then we should find a place to put the code, so all of us are not re-coding the same thing and wasting time. Meanwhile, any immediate plans to add macros with arguments? Or can I use cpp somehow? Otherwise I will have to re-code my use of that or should I wait a week or two. Also, I noticed it doesn't like functions with no args, "int func()" fails, wants "int func(void)". Very minor but thought I would mention. Great work. -Steve From rminnich at lanl.gov Tue Apr 8 11:18:01 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Tue Apr 8 11:18:01 2003 Subject: romcc In-Reply-To: <3E92EC9F.5090902@nexpath.com> Message-ID: On Tue, 8 Apr 2003, Steve Gehlbach wrote: > I gave it a quick test drive, looks pretty good. I want to put together > a complete rom so I can actually execute it, but I have other things on > the plate right now, will do so in a week or two I hope. Is the plan to > re-code the assy that exists now, or only added code? If yes, then we > should find a place to put the code, so all of us are not re-coding the > same thing and wasting time. I want to preserve the option of falling back to assy for those boards that have working assy. That way we can verify problems as they come up (and the will ...) > Also, I noticed it doesn't like functions with no args, "int func()" > fails, wants "int func(void)". Very minor but thought I would mention. This is good. ron From ebiederman at lnxi.com Tue Apr 8 11:18:10 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 8 11:18:10 2003 Subject: romcc In-Reply-To: <3E92EC9F.5090902@nexpath.com> References: <3E92EC9F.5090902@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > After a lot of pound the code generator seems stable, and I can take > > a break to release romcc. > > > > I gave it a quick test drive, looks pretty good. I want to put together a > complete rom so I can actually execute it, but I have other things on the plate > right now, will do so in a week or two I hope. If you have a working etherboot setup you can load hello_world.elf with etherboot. > Is the plan to re-code the assy > that exists now, or only added code? If yes, then we should find a place to put > the code, so all of us are not re-coding the same thing and wasting time. The plan is to open a development branch in the next couple of days and anything that survives on that branch should use it as much as possible. > Meanwhile, any immediate plans to add macros with arguments? Or can I use cpp > somehow? Otherwise I will have to re-code my use of that or should I wait a > week or two. With inline functions being cheap it should not be a problem. You might need to specify -O but otherwise it should be ok. But I may have missed something. > Also, I noticed it doesn't like functions with no args, "int func()" fails, > wants "int func(void)". Very minor but thought I would mention. That is not a function with no args that is a function with a variable number of arguments. And functions with a variable argument count are not something I can see supporting. > Great work. Thanks, Eric From ebiederman at lnxi.com Tue Apr 8 11:24:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 8 11:24:00 2003 Subject: romcc In-Reply-To: References: Message-ID: "Ronald G. Minnich" writes: > On Tue, 8 Apr 2003, Steve Gehlbach wrote: > > > I gave it a quick test drive, looks pretty good. I want to put together > > a complete rom so I can actually execute it, but I have other things on > > the plate right now, will do so in a week or two I hope. Is the plan to > > re-code the assy that exists now, or only added code? If yes, then we > > should find a place to put the code, so all of us are not re-coding the > > same thing and wasting time. > > I want to preserve the option of falling back to assy for those boards > that have working assy. That way we can verify problems as they come up > (and the will ...) For those that have not seen romcc. It outputs assembly. So basically it just creates another .inc file and the whole infrastructure does not need to change. > > Also, I noticed it doesn't like functions with no args, "int func()" > > fails, wants "int func(void)". Very minor but thought I would mention. > > This is good. The weird thing romcc does is that it takes your very last defined function as your main function (no forward references allowed). And that function it gets very upset if it is not void func(void); Because romcc considers it an error to use an uninitialized variable. It always fails though there are some cases where it is does not fail obviously. Eric From mikeh at netxinc.com Tue Apr 8 12:41:01 2003 From: mikeh at netxinc.com (mike hjorleifsson) Date: Tue Apr 8 12:41:01 2003 Subject: Newbie Question In-Reply-To: <20030408160001.32322.11449.Mailman@nwn.definitive.org> Message-ID: <00bc01c2fdf2$6e56f220$83fea8c0@NETXINC.NET> Could linuxbios be used to boot a "thin client" so that load times and bios issues are reduced ? Mike From rminnich at lanl.gov Tue Apr 8 12:47:01 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Tue Apr 8 12:47:01 2003 Subject: Newbie Question In-Reply-To: <00bc01c2fdf2$6e56f220$83fea8c0@NETXINC.NET> Message-ID: On Tue, 8 Apr 2003, mike hjorleifsson wrote: > Could linuxbios be used to boot a "thin client" so that load times and > bios issues are reduced ? yes, indeed. Ollie Lho had linuxbios booting an sis 630 that immediately went into microwindows and then started up VNC -- in a few seconds. He demo'd this at ALS 2000. We are following on this work to build thin clients, with Plan 9 thin clients as one possibility, although X11 will be first. ron From agnew at cs.umd.edu Tue Apr 8 13:06:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Tue Apr 8 13:06:00 2003 Subject: Newbie Question In-Reply-To: Message-ID: <20030408133518.N48819-100000@www.missl.cs.umd.edu> In the interest of being really fast, and saving some binary size to boot, i just want to point out DirectVNC http://www.adam-lilienthal.de/directvnc/ On Tue, 8 Apr 2003, Ronald G. Minnich wrote: > On Tue, 8 Apr 2003, mike hjorleifsson wrote: > > > Could linuxbios be used to boot a "thin client" so that load times and > > bios issues are reduced ? > > yes, indeed. Ollie Lho had linuxbios booting an sis 630 that immediately > went into microwindows and then started up VNC -- in a few seconds. He > demo'd this at ALS 2000. > > We are following on this work to build thin clients, with Plan 9 thin > clients as one possibility, although X11 will be first. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From mikeh at netxinc.com Tue Apr 8 13:20:01 2003 From: mikeh at netxinc.com (mike hjorleifsson) Date: Tue Apr 8 13:20:01 2003 Subject: Newbie Question In-Reply-To: <20030408133518.N48819-100000@www.missl.cs.umd.edu> Message-ID: <00d101c2fdf7$dcba1fd0$83fea8c0@NETXINC.NET> Actually I need to run a citrix client... Or could I setup a linux server as a "proxy" then use directvnc to that server and run the citrix client from the server ? This sounds like it could be hell though doing screen scrape after screen scrape. -----Original Message----- From: Adam Agnew [mailto:agnew at cs.umd.edu] Sent: Tuesday, April 08, 2003 1:38 PM To: Ronald G. Minnich Cc: mike hjorleifsson; linuxbios at clustermatic.org Subject: Re: Newbie Question In the interest of being really fast, and saving some binary size to boot, i just want to point out DirectVNC http://www.adam-lilienthal.de/directvnc/ On Tue, 8 Apr 2003, Ronald G. Minnich wrote: > On Tue, 8 Apr 2003, mike hjorleifsson wrote: > > > Could linuxbios be used to boot a "thin client" so that load times > > and bios issues are reduced ? > > yes, indeed. Ollie Lho had linuxbios booting an sis 630 that > immediately went into microwindows and then started up VNC -- in a few > seconds. He demo'd this at ALS 2000. > > We are following on this work to build thin clients, with Plan 9 thin > clients as one possibility, although X11 will be first. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From ebiederman at lnxi.com Tue Apr 8 19:29:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 8 19:29:00 2003 Subject: [COMMIT] Tree sync up, pass 1 Message-ID: Once again it is time for me to sync up my tree with the public LinuxBIOS tree. I should do this more often. But anyway... Each chunk of the differences that I digest and sync up with I will commit so this is the first of many tonight. - Cleaner #defines in pirq_routing.h - Remove duplicate include of c_start.S in arch/i386/lib/Config - Infrastructure to handle cpu startup bugs - irq routing table size check - split up the i786 microcode updates into their own files - Enable cpu ECC checking on the Athlons - More pci_ids - #defines for the location of the LinuxBIOS checksum - Correct handling of 64bit pci resources Eric From hansolofalcon at worldnet.att.net Tue Apr 8 20:18:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Tue Apr 8 20:18:01 2003 Subject: Newbie Question In-Reply-To: <20030408133518.N48819-100000@www.missl.cs.umd.edu> Message-ID: <000301c2fe31$3c98e1a0$1fa0580c@who5> Hello again from Gregg C Levine Adam, have you used the DirectVNC arrangement in your work? And how well does it work? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of Adam Agnew > Sent: Tuesday, April 08, 2003 1:38 PM > To: Ronald G. Minnich > Cc: mike hjorleifsson; linuxbios at clustermatic.org > Subject: Re: Newbie Question > > In the interest of being really fast, and saving some binary size to boot, > i just want to point out DirectVNC > http://www.adam-lilienthal.de/directvnc/ > > On Tue, 8 Apr 2003, Ronald G. Minnich wrote: > > > On Tue, 8 Apr 2003, mike hjorleifsson wrote: > > > > > Could linuxbios be used to boot a "thin client" so that load times and > > > bios issues are reduced ? > > > > yes, indeed. Ollie Lho had linuxbios booting an sis 630 that immediately > > went into microwindows and then started up VNC -- in a few seconds. He > > demo'd this at ALS 2000. > > > > We are following on this work to build thin clients, with Plan 9 thin > > clients as one possibility, although X11 will be first. > > > > ron > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From agnew at cs.umd.edu Tue Apr 8 21:21:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Tue Apr 8 21:21:01 2003 Subject: Newbie Question In-Reply-To: <000301c2fe31$3c98e1a0$1fa0580c@who5> Message-ID: <20030408215143.C50799-100000@www.missl.cs.umd.edu> Yeah, its uhm.. well, it could use some bug fixing. But its close. More importantly, it has a much smaller footprint than X11 and we've had trouble with X11 on linuxbios in the past. easier to just use framebuffers. On Tue, 8 Apr 2003, Gregg C Levine wrote: > Hello again from Gregg C Levine > Adam, have you used the DirectVNC arrangement in your work? And how > well does it work? > ------------------- > Gregg C Levine hansolofalcon at worldnet.att.net > ------------------------------------------------------------ > "The Force will be with you...Always." Obi-Wan Kenobi > "Use the Force, Luke."? Obi-Wan Kenobi > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > (This company dedicates this E-Mail to Master Yoda ) > > > > > -----Original Message----- > > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > > admin at clustermatic.org] On Behalf Of Adam Agnew > > Sent: Tuesday, April 08, 2003 1:38 PM > > To: Ronald G. Minnich > > Cc: mike hjorleifsson; linuxbios at clustermatic.org > > Subject: Re: Newbie Question > > > > In the interest of being really fast, and saving some binary size to > boot, > > i just want to point out DirectVNC > > http://www.adam-lilienthal.de/directvnc/ > > > > On Tue, 8 Apr 2003, Ronald G. Minnich wrote: > > > > > On Tue, 8 Apr 2003, mike hjorleifsson wrote: > > > > > > > Could linuxbios be used to boot a "thin client" so that load > times and > > > > bios issues are reduced ? > > > > > > yes, indeed. Ollie Lho had linuxbios booting an sis 630 that > immediately > > > went into microwindows and then started up VNC -- in a few > seconds. He > > > demo'd this at ALS 2000. > > > > > > We are following on this work to build thin clients, with Plan 9 > thin > > > clients as one possibility, although X11 will be first. > > > > > > ron > > > > > > _______________________________________________ > > > Linuxbios mailing list > > > Linuxbios at clustermatic.org > > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From aip at cwlinux.com Tue Apr 8 21:57:00 2003 From: aip at cwlinux.com (Andrew Ip) Date: Tue Apr 8 21:57:00 2003 Subject: Newbie Question In-Reply-To: <00d101c2fdf7$dcba1fd0$83fea8c0@NETXINC.NET>; from mike hjorleifsson on Tue, Apr 08, 2003 at 01:54:10PM -0400 References: <20030408133518.N48819-100000@www.missl.cs.umd.edu> <00d101c2fdf7$dcba1fd0$83fea8c0@NETXINC.NET> Message-ID: <20030409102735.B13201@mail.cwlinux.com> Mike, > Actually I need to run a citrix client... > Or could I setup a linux server as a "proxy" then use directvnc to that > server and run the citrix client from the server ? > This sounds like it could be hell though doing screen scrape after > screen scrape. Actually we tried LinuxBIOS + etherboot + ltsp + rdesktop. It gives me a Win2000 terminal clients. The funny part is Win2000 outside and Linux inside. I expect Citrix would be the same. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From rminnich at lanl.gov Tue Apr 8 22:09:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 8 22:09:01 2003 Subject: Newbie Question In-Reply-To: <20030409102735.B13201@mail.cwlinux.com> Message-ID: I think this 'remote desktop in a box' is an application for linuxbios that is just crying for someone to sell it as a turnkey system. my $.02 ron From mikeh at netxinc.com Tue Apr 8 22:11:01 2003 From: mikeh at netxinc.com (MikeH) Date: Tue Apr 8 22:11:01 2003 Subject: Newbie Question References: Message-ID: <00f801c2fe5a$5b791990$0401a8c0@laptop> add my .02 ================= Mike Hjorleifsson Chief Operating Officer NetX, Inc. 1110 South Ave Staten Island, NY 10314 Ph: 347-223-1331 Fx: 718-247-1673 PLEASE NOTE: This e-mail message may contain confidential and privileged material for the sole use of the intended recipient(s). If you are not the intended recipient you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please contact me immediately by e-mail or by telephone at (347) 273-1331 and delete the original message and any copies. Thank you. ----- Original Message ----- From: "ron minnich" To: "Andrew Ip" Cc: "mike hjorleifsson" ; "'Adam Agnew'" ; Sent: Tuesday, April 08, 2003 7:39 PM Subject: Re: Newbie Question > I think this 'remote desktop in a box' is an application for linuxbios > that is just crying for someone to sell it as a turnkey system. > > my $.02 > > ron > > From hansolofalcon at worldnet.att.net Tue Apr 8 23:10:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Tue Apr 8 23:10:01 2003 Subject: Newbie Question In-Reply-To: <00f801c2fe5a$5b791990$0401a8c0@laptop> Message-ID: <000001c2fe49$d4aa6640$1ac4580c@who5> Hello again from Gregg C Levine To quote a phrase that's even older then I am, "Count me in!". Besides, on another Linux related list that I belong to, someone suggested rebuilding the LTSP project for that processor. Of course, since that design isn't as capable for being a fileserver, or a terminal server, as an x86 configuration, and the same someone who suggested, made that point.... Oh, and Mike, I'm based in Queens. Astoria to be exact. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of MikeH > Sent: Wednesday, April 09, 2003 1:39 AM > To: ron minnich; Andrew Ip > Cc: 'Adam Agnew'; linuxbios at clustermatic.org > Subject: Re: Newbie Question > > add my .02 > ================= > Mike Hjorleifsson > Chief Operating Officer > NetX, Inc. > 1110 South Ave > Staten Island, NY 10314 > Ph: 347-223-1331 > Fx: 718-247-1673 > > PLEASE NOTE: This e-mail message may contain confidential and privileged > material for the sole use of the intended recipient(s). If you are not the > intended recipient you are hereby notified that you have received this > document in error and that any review, dissemination, distribution, or > copying of this message is strictly prohibited. If you have received this > communication in error, please contact me immediately by e-mail or by > telephone at (347) 273-1331 and delete the original message and any copies. > Thank you. > ----- Original Message ----- > From: "ron minnich" > To: "Andrew Ip" > Cc: "mike hjorleifsson" ; "'Adam Agnew'" > ; > Sent: Tuesday, April 08, 2003 7:39 PM > Subject: Re: Newbie Question > > > > I think this 'remote desktop in a box' is an application for linuxbios > > that is just crying for someone to sell it as a turnkey system. > > > > my $.02 > > > > ron > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From agnew at cs.umd.edu Tue Apr 8 23:39:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Tue Apr 8 23:39:01 2003 Subject: Newbie Question In-Reply-To: <00d101c2fdf7$dcba1fd0$83fea8c0@NETXINC.NET> Message-ID: <20030408235534.A51099-100000@www.missl.cs.umd.edu> In that case, maybe you should just try DirectFB + XDirectFB. It will get you a rootless x server to live on top of DirectFB. And still with a pretty small footprint. www.directfb.org There's also KDrive http://www.pps.jussieu.fr/~jch/software/kdrive.html Both can run a lot of X-Windows applications pretty much unchanged. On top of that, directfb has gtk ported over to directfb, so gnome applications can run easily too. My main point is just that XFree86 is probably overkill for these thin-client applications. By the way, fbvnc is another framebuffer vnc client that wont need directfb http://www.w-m-p.com/pocketworkstation/fbvnc.html On Tue, 8 Apr 2003, mike hjorleifsson wrote: > Actually I need to run a citrix client... > Or could I setup a linux server as a "proxy" then use directvnc to that > server and run the citrix client from the server ? > This sounds like it could be hell though doing screen scrape after > screen scrape. > > -----Original Message----- > From: Adam Agnew [mailto:agnew at cs.umd.edu] > Sent: Tuesday, April 08, 2003 1:38 PM > To: Ronald G. Minnich > Cc: mike hjorleifsson; linuxbios at clustermatic.org > Subject: Re: Newbie Question > > > In the interest of being really fast, and saving some binary size to > boot, i just want to point out DirectVNC > http://www.adam-lilienthal.de/directvnc/ > > On Tue, 8 Apr 2003, Ronald G. Minnich wrote: > > > On Tue, 8 Apr 2003, mike hjorleifsson wrote: > > > > > Could linuxbios be used to boot a "thin client" so that load times > > > and bios issues are reduced ? > > > > yes, indeed. Ollie Lho had linuxbios booting an sis 630 that > > immediately went into microwindows and then started up VNC -- in a few > > > seconds. He demo'd this at ALS 2000. > > > > We are following on this work to build thin clients, with Plan 9 thin > > clients as one possibility, although X11 will be first. > > > > ron > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From Lawrence_Morrison at Dell.com Wed Apr 9 09:17:01 2003 From: Lawrence_Morrison at Dell.com (Morrison, Lawrence) Date: Wed Apr 9 09:17:01 2003 Subject: Newbie Question Message-ID: <50D868C51A5C8243816C05835C1FF034042101@ausx2kmpc105.aus.amer.dell.com> I want in on this : ) Here's my $.02 as well!! -----Original Message----- From: MikeH [mailto:mikeh at netxinc.com] Sent: Wednesday, April 09, 2003 12:39 AM To: ron minnich; Andrew Ip Cc: 'Adam Agnew'; linuxbios at clustermatic.org Subject: Re: Newbie Question add my .02 ================= Mike Hjorleifsson Chief Operating Officer NetX, Inc. 1110 South Ave Staten Island, NY 10314 Ph: 347-223-1331 Fx: 718-247-1673 PLEASE NOTE: This e-mail message may contain confidential and privileged material for the sole use of the intended recipient(s). If you are not the intended recipient you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please contact me immediately by e-mail or by telephone at (347) 273-1331 and delete the original message and any copies. Thank you. ----- Original Message ----- From: "ron minnich" To: "Andrew Ip" Cc: "mike hjorleifsson" ; "'Adam Agnew'" ; Sent: Tuesday, April 08, 2003 7:39 PM Subject: Re: Newbie Question > I think this 'remote desktop in a box' is an application for linuxbios > that is just crying for someone to sell it as a turnkey system. > > my $.02 > > ron > > _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From dkotian3 at vsnl.net Wed Apr 9 14:31:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 14:31:00 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it Message-ID: <003c01c2feca$63d75320$d92041db@vsnl.net> HI, Looking at the burn mtd script, it seems that the linux kernel image also needs to be burnt on the BIOS flash along with linuxbios.In this case, if the customized kernel image is around 900KB ,then it may be difficult to fit it on the FLASH ROMS which is usually 2-4MBit. Is there a way to boot it from IDE device itself instead of the above process. Could some please clarify on this. Thanks and Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From dkotian3 at vsnl.net Wed Apr 9 14:32:48 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 14:32:48 2003 Subject: Kernel patches significance in LinuxBIOS cvs References: <50D868C51A5C8243816C05835C1FF034042101@ausx2kmpc105.aus.amer.dell.com> Message-ID: <003b01c2feca$63635ba0$d92041db@vsnl.net> Hi, Could someone please clarify on this. 1. Is there a kernel patch needed for all the motherboards to work out LinuxBIOS? 2. How do I know that I need a kernel patch to work out LinuxBIOS on my motherboard. Please suggest Thanks and Regards Deepak From dkotian3 at vsnl.net Wed Apr 9 14:32:52 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 14:32:52 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS Message-ID: <003d01c2feca$645d9a20$d92041db@vsnl.net> Hi, This may be a very basic simple question, but I do not know a very good answer/solution to do this. If I want to burn LinuxBIOS, I can use the burn_mtd script , it may write it successfully as well, but if linuxbios is not correctly built. I cannot get the any O.S started on this machine, which means I would need to replace the BIOS FLASH ROM with a one with proper BIOS and bring up the machine. Is there a way I can re-program the old BIOS on to the corrupted FLASH ROMS. Please note,I do not have a PLC or any kind of FLASH programmer with me. Can I do without it. If there are any suggestions, please let me know. Thanks and Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From agnew at cs.umd.edu Wed Apr 9 14:35:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 14:35:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <003c01c2feca$63d75320$d92041db@vsnl.net> Message-ID: <20030409150504.I53761-100000@www.missl.cs.umd.edu> 3 ways. 1. Etherboot 5.0.6 with polled ide patch: download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for the patch. 2. Etherboot development series: download etherboot-5.1.7 3. ADLO see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most work, but wont require you to modify your kernel when you're done) - Adam Agnew On Thu, 10 Apr 2003, Deepak Kotian wrote: > HI, > > Looking at the burn mtd script, it seems that the linux kernel image also needs > to be burnt on the BIOS flash along with linuxbios.In this case, if the customized kernel > image is around 900KB ,then it may be difficult to fit it on the FLASH ROMS which is > usually 2-4MBit. > Is there a way to boot it from IDE device itself instead of the above process. > > Could some please clarify on this. > > Thanks and Regards > Deepak > > From dkotian3 at vsnl.net Wed Apr 9 14:52:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 14:52:00 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it References: <20030409150504.I53761-100000@www.missl.cs.umd.edu> Message-ID: <00dc01c2fecd$55bd1100$d92041db@vsnl.net> Thanks for the quick response. If I use ADLO, will the machine boot up really fast, would there be performance degradation because of this. My main intention is to improve the startup time to around 2/3 seconds. About ADLO, I could not see it in the utils folder in the linuxbios source I have. Can the cvs snapshot be updated please. I do not have access to LINUX box as of now. Regards Deepak ----- Original Message ----- From: "Adam Agnew" To: "Deepak Kotian" Cc: Sent: Thursday, April 10, 2003 12:37 AM Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it > 3 ways. > > 1. Etherboot 5.0.6 with polled ide patch: > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > the patch. > > 2. Etherboot development series: > download etherboot-5.1.7 > > 3. ADLO > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > work, but wont require you to modify your kernel when you're done) > > - Adam Agnew > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > HI, > > > > Looking at the burn mtd script, it seems that the linux kernel image also needs > > to be burnt on the BIOS flash along with linuxbios.In this case, if the customized kernel > > image is around 900KB ,then it may be difficult to fit it on the FLASH ROMS which is > > usually 2-4MBit. > > Is there a way to boot it from IDE device itself instead of the above process. > > > > Could some please clarify on this. > > > > Thanks and Regards > > Deepak > > > > > > From rminnich at lanl.gov Wed Apr 9 14:53:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 9 14:53:00 2003 Subject: Kernel patches significance in LinuxBIOS cvs In-Reply-To: <003b01c2feca$63635ba0$d92041db@vsnl.net> Message-ID: On Wed, 9 Apr 2003, Deepak Kotian wrote: > 1. Is there a kernel patch needed for all the motherboards to work out > LinuxBIOS? no. > 2. How do I know that I need a kernel patch to work out LinuxBIOS > on my motherboard. if linuxbios comes up so fast that IDE is not ready, you need the IDE patch. ron From ebiederman at lnxi.com Wed Apr 9 14:59:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 14:59:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <003d01c2feca$645d9a20$d92041db@vsnl.net> References: <003d01c2feca$645d9a20$d92041db@vsnl.net> Message-ID: "Deepak Kotian" writes: > Hi, > > This may be a very basic simple question, but I do not know a very good > answer/solution > > to do this. > > If I want to burn LinuxBIOS, I can use the burn_mtd script , it may write it > > successfully as well, but if linuxbios is not correctly built. I cannot get > the > > any O.S started on this machine, which means I would need to replace the > BIOS FLASH ROM with a one with proper BIOS and bring up the machine. > Is there a way I can re-program the old BIOS on to the corrupted FLASH ROMS. > > Please note,I do not have a PLC or any kind of FLASH programmer with me. Can > > I do without it. Usually it is recommeded for bootstraping and development that you either have a spare rom chip, which you can hot swap in to flash, or that you have an external burner. Eric From agnew at cs.umd.edu Wed Apr 9 15:07:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 15:07:00 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <00dc01c2fecd$55bd1100$d92041db@vsnl.net> Message-ID: <20030409153422.S53761-100000@www.missl.cs.umd.edu> You don't want to *touch* that cvs snapshot its really old. give me a few minutes and i'll send you an email with where you can download a new tarbell from today. On Thu, 10 Apr 2003, Deepak Kotian wrote: > > Thanks for the quick response. > > > If I use ADLO, will the machine boot up really fast, would there be > performance > degradation because of this. My main intention is to improve the startup > time to around > 2/3 seconds. > > About ADLO, I could not see it in the utils folder in the linuxbios source I > have. > Can the cvs snapshot be updated please. I do not have access to LINUX box as > of now. > > Regards > Deepak > > > > > ----- Original Message ----- > From: "Adam Agnew" > To: "Deepak Kotian" > Cc: > Sent: Thursday, April 10, 2003 12:37 AM > Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt > on it > > > > 3 ways. > > > > 1. Etherboot 5.0.6 with polled ide patch: > > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > > the patch. > > > > 2. Etherboot development series: > > download etherboot-5.1.7 > > > > 3. ADLO > > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > > work, but wont require you to modify your kernel when you're done) > > > > - Adam Agnew > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > HI, > > > > > > Looking at the burn mtd script, it seems that the linux kernel image > also needs > > > to be burnt on the BIOS flash along with linuxbios.In this case, if > the customized kernel > > > image is around 900KB ,then it may be difficult to fit it on the > FLASH ROMS which is > > > usually 2-4MBit. > > > Is there a way to boot it from IDE device itself instead of the > above process. > > > > > > Could some please clarify on this. > > > > > > Thanks and Regards > > > Deepak > > > > > > > > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Wed Apr 9 15:08:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 15:08:00 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS References: <003d01c2feca$645d9a20$d92041db@vsnl.net> Message-ID: <016c01c2fecf$764df0e0$d92041db@vsnl.net> Thanks Eric. I have a spare FLASH ROM. Could you please elaborate on the process of "hot swap", is it that I can replace the FLASH ROM with the corrupted one with the backup flash ROM, then once it boots up, I can just replace the corrupted FLASH ROM while the O.S. is up and burn the backed up bios from the harddisk onto the corrupted FLAH ROMS. Is this correct and feasible, please elaborate.. ----- Original Message ----- From: "Eric W. Biederman" To: "Deepak Kotian" Cc: Sent: Thursday, April 10, 2003 1:00 AM Subject: Re: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS > "Deepak Kotian" writes: > > > Hi, > > > > This may be a very basic simple question, but I do not know a very good > > answer/solution > > > > to do this. > > > > If I want to burn LinuxBIOS, I can use the burn_mtd script , it may write it > > > > successfully as well, but if linuxbios is not correctly built. I cannot get > > the > > > > any O.S started on this machine, which means I would need to replace the > > BIOS FLASH ROM with a one with proper BIOS and bring up the machine. > > Is there a way I can re-program the old BIOS on to the corrupted FLASH ROMS. > > > > Please note,I do not have a PLC or any kind of FLASH programmer with me. Can > > > > I do without it. > > Usually it is recommeded for bootstraping and development that you either > have a spare rom chip, which you can hot swap in to flash, or that you have > an external burner. > > > Eric > From agnew at cs.umd.edu Wed Apr 9 15:09:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 15:09:00 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <00dc01c2fecd$55bd1100$d92041db@vsnl.net> Message-ID: <20030409153944.D53761-100000@www.missl.cs.umd.edu> Oh, and ADLO slows things down a little. mostly because it definitely needs the hard drives to be spun up already, while the linux kernel can be tricked into thinking they already are. Same with etherboot. If you want 3 second boot up, booting from IDE is out of the question. The other thing you can try is a DoC Millenium which is a bios chip with 8 megs of storage. On Thu, 10 Apr 2003, Deepak Kotian wrote: > > Thanks for the quick response. > > > If I use ADLO, will the machine boot up really fast, would there be > performance > degradation because of this. My main intention is to improve the startup > time to around > 2/3 seconds. > > About ADLO, I could not see it in the utils folder in the linuxbios source I > have. > Can the cvs snapshot be updated please. I do not have access to LINUX box as > of now. > > Regards > Deepak > > > > > ----- Original Message ----- > From: "Adam Agnew" > To: "Deepak Kotian" > Cc: > Sent: Thursday, April 10, 2003 12:37 AM > Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt > on it > > > > 3 ways. > > > > 1. Etherboot 5.0.6 with polled ide patch: > > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > > the patch. > > > > 2. Etherboot development series: > > download etherboot-5.1.7 > > > > 3. ADLO > > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > > work, but wont require you to modify your kernel when you're done) > > > > - Adam Agnew > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > HI, > > > > > > Looking at the burn mtd script, it seems that the linux kernel image > also needs > > > to be burnt on the BIOS flash along with linuxbios.In this case, if > the customized kernel > > > image is around 900KB ,then it may be difficult to fit it on the > FLASH ROMS which is > > > usually 2-4MBit. > > > Is there a way to boot it from IDE device itself instead of the > above process. > > > > > > Could some please clarify on this. > > > > > > Thanks and Regards > > > Deepak > > > > > > > > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Wed Apr 9 15:14:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 15:14:00 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it References: <20030409153944.D53761-100000@www.missl.cs.umd.edu> Message-ID: <019201c2fed0$6435df20$d92041db@vsnl.net> OK.Thanks for the information. What would be best boot up time with IDE with LinuxBIOS ? Any inputs from anyone. Thanks and Regards Deepak ----- Original Message ----- From: "Adam Agnew" To: "Deepak Kotian" Cc: Sent: Thursday, April 10, 2003 1:11 AM Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it > Oh, and ADLO slows things down a little. mostly because it definitely > needs the hard drives to be spun up already, while the linux kernel can be > tricked into thinking they already are. Same with etherboot. If you want 3 > second boot up, booting from IDE is out of the question. The other thing > you can try is a DoC Millenium which is a bios chip with 8 megs of > storage. > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > Thanks for the quick response. > > > > > > If I use ADLO, will the machine boot up really fast, would there be > > performance > > degradation because of this. My main intention is to improve the startup > > time to around > > 2/3 seconds. > > > > About ADLO, I could not see it in the utils folder in the linuxbios source I > > have. > > Can the cvs snapshot be updated please. I do not have access to LINUX box as > > of now. > > > > Regards > > Deepak > > > > > > > > > > ----- Original Message ----- > > From: "Adam Agnew" > > To: "Deepak Kotian" > > Cc: > > Sent: Thursday, April 10, 2003 12:37 AM > > Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt > > on it > > > > > > > 3 ways. > > > > > > 1. Etherboot 5.0.6 with polled ide patch: > > > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > > > the patch. > > > > > > 2. Etherboot development series: > > > download etherboot-5.1.7 > > > > > > 3. ADLO > > > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > > > work, but wont require you to modify your kernel when you're done) > > > > > > - Adam Agnew > > > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > > > HI, > > > > > > > > Looking at the burn mtd script, it seems that the linux kernel image > > also needs > > > > to be burnt on the BIOS flash along with linuxbios.In this case, if > > the customized kernel > > > > image is around 900KB ,then it may be difficult to fit it on the > > FLASH ROMS which is > > > > usually 2-4MBit. > > > > Is there a way to boot it from IDE device itself instead of the > > above process. > > > > > > > > Could some please clarify on this. > > > > > > > > Thanks and Regards > > > > Deepak > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > From ebiederman at lnxi.com Wed Apr 9 15:21:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 15:21:01 2003 Subject: [ANNOUNCE] linuxbios utilities. Message-ID: At: ftp://ftp.lnxi.com/pub/src/linuxbios/utilities/ There are several utilities that LinuxNetworx is pleased to make available. lbflash which flashes linuxbios using the mtd drivers cmos_util which changes the few LinuxBIOS settings from Linux. bios_copy which is a simple utility to save/restore an arbitrary BIOS, using the mtd drivers. Eric From agnew at cs.umd.edu Wed Apr 9 15:25:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 15:25:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <20030409153944.D53761-100000@www.missl.cs.umd.edu> Message-ID: <20030409155255.G53761-100000@www.missl.cs.umd.edu> http://www.missl.cs.umd.edu/~agnew/freebios-snapshot-for-deepak.tgz On Wed, 9 Apr 2003, Adam Agnew wrote: > Oh, and ADLO slows things down a little. mostly because it definitely > needs the hard drives to be spun up already, while the linux kernel can be > tricked into thinking they already are. Same with etherboot. If you want 3 > second boot up, booting from IDE is out of the question. The other thing > you can try is a DoC Millenium which is a bios chip with 8 megs of > storage. > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > Thanks for the quick response. > > > > > > If I use ADLO, will the machine boot up really fast, would there be > > performance > > degradation because of this. My main intention is to improve the startup > > time to around > > 2/3 seconds. > > > > About ADLO, I could not see it in the utils folder in the linuxbios source I > > have. > > Can the cvs snapshot be updated please. I do not have access to LINUX box as > > of now. > > > > Regards > > Deepak > > > > > > > > > > ----- Original Message ----- > > From: "Adam Agnew" > > To: "Deepak Kotian" > > Cc: > > Sent: Thursday, April 10, 2003 12:37 AM > > Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt > > on it > > > > > > > 3 ways. > > > > > > 1. Etherboot 5.0.6 with polled ide patch: > > > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > > > the patch. > > > > > > 2. Etherboot development series: > > > download etherboot-5.1.7 > > > > > > 3. ADLO > > > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > > > work, but wont require you to modify your kernel when you're done) > > > > > > - Adam Agnew > > > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > > > HI, > > > > > > > > Looking at the burn mtd script, it seems that the linux kernel image > > also needs > > > > to be burnt on the BIOS flash along with linuxbios.In this case, if > > the customized kernel > > > > image is around 900KB ,then it may be difficult to fit it on the > > FLASH ROMS which is > > > > usually 2-4MBit. > > > > Is there a way to boot it from IDE device itself instead of the > > above process. > > > > > > > > Could some please clarify on this. > > > > > > > > Thanks and Regards > > > > Deepak > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Wed Apr 9 15:39:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 15:39:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it References: <20030409155255.G53761-100000@www.missl.cs.umd.edu> Message-ID: <01ef01c2fed3$ca9f7e80$d92041db@vsnl.net> Thanks very much. I have download it.Appreciated your help. Now I can see all over there. ----- Original Message ----- From: "Adam Agnew" To: "Deepak Kotian" Cc: Sent: Thursday, April 10, 2003 1:27 AM Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it > > http://www.missl.cs.umd.edu/~agnew/freebios-snapshot-for-deepak.tgz > > > On Wed, 9 Apr 2003, Adam Agnew wrote: > > > Oh, and ADLO slows things down a little. mostly because it definitely > > needs the hard drives to be spun up already, while the linux kernel can be > > tricked into thinking they already are. Same with etherboot. If you want 3 > > second boot up, booting from IDE is out of the question. The other thing > > you can try is a DoC Millenium which is a bios chip with 8 megs of > > storage. > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > > > > Thanks for the quick response. > > > > > > > > > If I use ADLO, will the machine boot up really fast, would there be > > > performance > > > degradation because of this. My main intention is to improve the startup > > > time to around > > > 2/3 seconds. > > > > > > About ADLO, I could not see it in the utils folder in the linuxbios source I > > > have. > > > Can the cvs snapshot be updated please. I do not have access to LINUX box as > > > of now. > > > > > > Regards > > > Deepak > > > > > > > > > > > > > > > ----- Original Message ----- > > > From: "Adam Agnew" > > > To: "Deepak Kotian" > > > Cc: > > > Sent: Thursday, April 10, 2003 12:37 AM > > > Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt > > > on it > > > > > > > > > > 3 ways. > > > > > > > > 1. Etherboot 5.0.6 with polled ide patch: > > > > download etherboot-5.0.6 and see http://www.missl.cs.umd.edu/~agnew/ for > > > > the patch. > > > > > > > > 2. Etherboot development series: > > > > download etherboot-5.1.7 > > > > > > > > 3. ADLO > > > > see freebios/utils/ADLO/ from the linuxbios cvs. (this one is the most > > > > work, but wont require you to modify your kernel when you're done) > > > > > > > > - Adam Agnew > > > > > > > > On Thu, 10 Apr 2003, Deepak Kotian wrote: > > > > > > > > > HI, > > > > > > > > > > Looking at the burn mtd script, it seems that the linux kernel image > > > also needs > > > > > to be burnt on the BIOS flash along with linuxbios.In this case, if > > > the customized kernel > > > > > image is around 900KB ,then it may be difficult to fit it on the > > > FLASH ROMS which is > > > > > usually 2-4MBit. > > > > > Is there a way to boot it from IDE device itself instead of the > > > above process. > > > > > > > > > > Could some please clarify on this. > > > > > > > > > > Thanks and Regards > > > > > Deepak > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > Linuxbios mailing list > > > Linuxbios at clustermatic.org > > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From agnew at cs.umd.edu Wed Apr 9 15:45:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 15:45:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <3E947D3E.4020109@bitworks.com> Message-ID: <20030409161427.R54263-100000@www.missl.cs.umd.edu> I stand corrected, Richard is right. You may find a CF on IDE adapter is your best choice Deepak. On Wed, 9 Apr 2003, Richard Smith wrote: > Adam Agnew wrote: > > > tricked into thinking they already are. Same with etherboot. If you want 3 > > second boot up, booting from IDE is out of the question. The other thing > > From a mechanical IDE drive yes, but not from a Compact Flash which is > ready almostly instantly. > > -- > Richard A. Smith > rsmith at bitworks.com > > > From ebiederman at lnxi.com Wed Apr 9 15:51:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 15:51:01 2003 Subject: Kernel patches significance in LinuxBIOS cvs In-Reply-To: References: Message-ID: ron minnich writes: > On Wed, 9 Apr 2003, Deepak Kotian wrote: > > > 1. Is there a kernel patch needed for all the motherboards to work out > > LinuxBIOS? > > no. > > > 2. How do I know that I need a kernel patch to work out LinuxBIOS > > on my motherboard. > > if linuxbios comes up so fast that IDE is not ready, you need the IDE > patch. Note that while it works it is not the correct fix. There is a place in the probe code where the kernel should spin on the busy bit. But so far Linux has ignored that. The PowerPC guys are seeing this problem as well so there is chance the clean fix will get into the kernel. Eric From dkotian3 at vsnl.net Wed Apr 9 15:51:06 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 9 15:51:06 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it References: <20030409161427.R54263-100000@www.missl.cs.umd.edu> Message-ID: <021001c2fed5$8cdedc60$d92041db@vsnl.net> Thanks. I have my kernel image on a 32 MB compact IDE flash. I guess, I can still achieve a 3 seconds startup then with linuxbios for my machine.. ----- Original Message ----- From: "Adam Agnew" To: "Richard Smith" Cc: Sent: Thursday, April 10, 2003 1:46 AM Subject: Re: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it > > I stand corrected, Richard is right. You may find a CF on IDE adapter is > your best choice Deepak. > > On Wed, 9 Apr 2003, Richard Smith wrote: > > > Adam Agnew wrote: > > > > > tricked into thinking they already are. Same with etherboot. If you want 3 > > > second boot up, booting from IDE is out of the question. The other thing > > > > From a mechanical IDE drive yes, but not from a Compact Flash which is > > ready almostly instantly. > > > > -- > > Richard A. Smith > > rsmith at bitworks.com > > > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rsmith at bitworks.com Wed Apr 9 16:13:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 9 16:13:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <003d01c2feca$645d9a20$d92041db@vsnl.net> References: <003d01c2feca$645d9a20$d92041db@vsnl.net> Message-ID: <3E9485B3.4070004@bitworks.com> Deepak Kotian wrote: > any O.S started on this machine, which means I would need to replace the > BIOS FLASH ROM with a one with proper BIOS and bring up the machine. > Is there a way I can re-program the old BIOS on to the corrupted FLASH ROMS. > Please note,I do not have a PLC or any kind of FLASH programmer with me. Can > I do without it. > > If there are any suggestions, please let me know. Deepak, Something you might also consider is an ROM Emulator. One that will do a larger flash is still expensive but not as expensive as a good flash/eeprom programmer. With a ROM emulator you remove the original BIOS chip and plug in the emulator. Then you download you code into the emulator. The target never knows the difference. This also saves a lot of time in the code-test-recode cycle as the emulator is usually much faster than re-flashing a chip. Of course if your chip is soldered on the board it's not really an option. I've done a good bit of my developemnt with a 1-Mbit ROM emulator from tech-tools called the EconoROM and a DIP to 32pin PLCC adapter. 1Mbit gets you 128k of space which will hold Linuxbios fine. The emulator + adapter would probally run you $400-$500. -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Wed Apr 9 16:29:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 9 16:29:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <019201c2fed0$6435df20$d92041db@vsnl.net> Message-ID: On Thu, 10 Apr 2003, Deepak Kotian wrote: > OK.Thanks for the information. > What would be best boot up time with IDE with LinuxBIOS ? Any inputs from > anyone. I got to < 10 seconds once when I was able to put linux in a machine with 1 Mbyte flash. ron From rminnich at lanl.gov Wed Apr 9 16:31:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 9 16:31:01 2003 Subject: Kernel patches significance in LinuxBIOS cvs In-Reply-To: Message-ID: On 9 Apr 2003, Eric W. Biederman wrote: > Note that while it works it is not the correct fix. agreed, my patch is very ugly. > There is a place in the probe code where the kernel should spin > on the busy bit. But so far Linux has ignored that. linux has not gotten this right for 3 years now, I gave up waiting. ron From ebiederman at lnxi.com Wed Apr 9 19:20:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 19:20:00 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: <17273545427.20030410031351@reactor.ru> References: <17273545427.20030410031351@reactor.ru> Message-ID: Alexander Amelkin writes: > Hello Eric, > > Wednesday, April 9, 2003, 11:34:26 PM, you wrote: > > EWB> cmos_util which changes the few LinuxBIOS settings from Linux. > > Do we really use the CMOS already? What do we store there? There are several other people who could have answered and this is a question of general interest so back to the list, with this. I am frequently very busy so I do not appreciate LinuxBIOS mail directed exclusively at me. Yes. For normal vs fallback. For serial port baud rates, for the boot order that etherboot should use. Check out the appropriate cmos.layout files. Eric From ebiederman at lnxi.com Wed Apr 9 19:25:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 19:25:01 2003 Subject: Kernel patches significance in LinuxBIOS cvs In-Reply-To: References: Message-ID: ron minnich writes: > On 9 Apr 2003, Eric W. Biederman wrote: > > > Note that while it works it is not the correct fix. > > agreed, my patch is very ugly. > > > There is a place in the probe code where the kernel should spin > > on the busy bit. But so far Linux has ignored that. > > linux has not gotten this right for 3 years now, I gave up waiting. My points are: 1) There is a correct solution polling the BSY bit. And except in one case that only Ollie has seen that is sufficient. 2) The correct code currently exists in etherboot. 3) There are PPC kernel developers who are seeing this as well, and agree with me about the correct fix. 4) The PPC guys have it in there tree and were busily pushing it into the mainstream kernel. For just the LinuxBIOS case Andre Hendricks figured we have a borked BIOS. For the PPC case there was enough justification to get the code in. This is just a status report of where that conversation was last time I looked. But the code was accepted in principle now if we can just get a clear development path on the IDE layer something might actually happen. Eric From joey at joescan.com Wed Apr 9 20:42:01 2003 From: joey at joescan.com (Joey Nelson) Date: Wed Apr 9 20:42:01 2003 Subject: PIRQ problems on Geode GX1 board. Message-ID: <001201c2fefd$ee848b40$6501a8c0@pink> Thanks to lots of great help from Eric on the Etherboot list I have my system mostly working, except there appear to be some problems with the PCI IRQ routing stuff. I ran getpir on the target with its standard bios, and got a good irq_tables.c, and HAVE_PIRQ_TABLE=1 is set in the Config. But this line is troubling: Verifing priq routing tables copy at 0xf0000...failed Any idea why this is failing? When the kernel is loading I get the following complaint: PCI: No IRQ known for interrupt pin A of device 00:0d.0. Please try using pci=biosirq. I'm also hoping this be why my IDE does not appear to be working. I don't imagine it is the memcpy call that is bad, so I'm guessing there is a problem with my memory setup. Any suggestions? Thanks, Joey Nelson LinuxBIOS-1.0.0 Wed Apr 2 15:06:02 PST 2003 starting... Setting up default parameters for memory Sizing memory Probing for DIMM0 Found DIMM0 Page Size: 00000800 Component Banks: 4 Module Banks: 1 DIMM size: 02000000 Probing for DIMM1 Memory sizing done, MC_BANK_CFG = 0x00701310 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.0.0 Wed Apr 2 15:06:02 PST 2003 booting... Finding PCI configuration type. PCI: Using configuration type 1 handle_superio start, nsuperio 1 handle_superio Pass 0, check #0, s 0000a920 s->super 0000ab64 handle_superio: Pass 0, Superio SMC fdc37b78x handle_superio port 0x3f0, defaultport 0x3f0 handle_superio Using port 0x3f0 handle_superio Pass 0, done #0 handle_superio done Scanning PCI bus...PCI: pci_scan_bus for bus 0 PCI: 00:00.0 [1078/0001] PCI: 00:0d.0 [10ec/8139] PCI: 00:12.0 [1078/0100] PCI: 00:12.1 [1078/0101] PCI: 00:12.2 [1078/0102] PCI: 00:12.3 [1078/0103] PCI: 00:12.4 [1078/0104] PCI: 00:13.0 [0e11/a0f8] PCI: pci_scan_bus returning with max=00 done Allocating PCI resources... ASSIGN RESOURCES, bus 0 PCI: 00:0d.0 10 <- [0x00001000 - 0x000010ff] io PCI: 00:0d.0 14 <- [0xfebfd000 - 0xfebfd0ff] mem PCI: 00:12.1 10 <- [0xfebfe000 - 0xfebfe0ff] mem PCI: 00:12.2 20 <- [0x00001400 - 0x0000147f] io PCI: 00:12.3 10 <- [0xfebff000 - 0xfebff07f] mem PCI: 00:12.4 10 <- [0xfebfb000 - 0xfebfbfff] mem PCI: 00:13.0 10 <- [0xfebfc000 - 0xfebfcfff] mem ASSIGNED RESOURCES, bus 0 Allocating VGA resource done. Enabling PCI resourcess...PCI: 00:00.0 cmd <- 07 PCI: 00:0d.0 cmd <- 03 PCI: 00:12.0 cmd <- 0f PCI: 00:12.1 cmd <- 02 PCI: 00:12.2 cmd <- 01 PCI: 00:12.3 cmd <- 02 PCI: 00:12.4 cmd <- 03 PCI: 00:13.0 cmd <- 02 done. Initializing PCI devices... PCI devices initialized DIMM0: 32MB (2kB page size, 4 component banks, 1 module banks) DIMM1: empty Reserving 4096kB for video memory BC_DRAM_TOP = 0x01bfffff MC_GBASE_ADD = 0x00000038 totalram: 28M Initializing CPU #0 Enabling cache...done. Max cpuid index : 2 Vendor ID : Geode by NSC Processor Type : 0x00 Processor Family : 0x05 Processor Model : 0x04 Processor Mask : 0x00 Processor Stepping : 0x00 Feature flags : 0x00808131 Cache/TLB descriptor values: 1 reads required Desc 0x70 : UNKNOWN Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null Desc 0x80 : UNKNOWN Desc 0x00 : null Desc 0x00 : null Desc 0x00 : null done. CPU #0 Initialized Enable FLASH Set F0/0x52 to 0xee cs5530: Enabling Primary IDE Controller cs5530: Enabling Secondary IDE Controller Set F0/0x5b to |= 1 << 5(0x38) handle_superio start, nsuperio 1 handle_superio Pass 1, check #0, s 0000a920 s->super 0000ab64 handle_superio: Pass 1, Superio SMC fdc37b78x handle_superio port 0x3f0, defaultport 0x3f0 handle_superio Using port 0x3f0 Call init Enabling com device: 04 Com1 configured Disabling com device: 05 handle_superio Pass 1, done #0 handle_superio done Final southbridge fixup F0/5b = 0x38 cs5530: Enabling Primary IDE Controller cs5530: Enabling Secondary IDE Controller F0/5b = 0x38 Final mainboard fixup handle_superio start, nsuperio 1 handle_superio Pass 2, check #0, s 0000a920 s->super 0000ab64 handle_superio: Pass 2, Superio SMC fdc37b78x handle_superio port 0x3f0, defaultport 0x3f0 handle_superio Using port 0x3f0 handle_superio Pass 2, done #0 handle_superio done Checking IRQ routing tables... /home/highfreq/tt/freebios/src/arch/i386/lib/pirq_routing.c: 24:check_pirq_routing_table() - irq_routing_table located at: 0x00008c20 done. Copying IRQ routing tables to 0xf0000...done. Verifing priq routing tables copy at 0xf0000...failed Wrote linuxbios table at: 00000500 - 00000634 checksum f470 Welcome to elfboot, the open sourced starter. January 2002, Eric Biederman. Version 1.2 203:init_bytes() - zkernel_start:0xfffc0000 zkernel_mask:0x0000ffff Searching for 16 byte tags 64:rom_read_bytes() - overflowed source buffer. max_block = 3 init_bytes found 0 tags Found ELF candiate at offset 0 Loading Etherboot version: 5.1.7 Dropping non PT_LOAD segment New segment addr 0x20000 size 0xeaa4 offset 0xb0 filesize 0x5d60 (cleaned up) New segment addr 0x20000 size 0xeaa4 offset 0xb0 filesize 0x5d60 Loading Segment: addr: 0x0000000001b825a8 memsz: 0x000000000000eaa4 filesz: 0x0000000000005d60 Clearing Segment: addr: 0x0000000001b88308 memsz: 0x0000000000008d44 Jumping to boot code at 0x20000 ROM segment 0x0002 length 0x0ce4 reloc 0x00020000 CPU 138 Mhz Etherboot 5.1.7 (GPL) ELF for [IDE] Relocating _text from: [00020000,0002eab0) to [01bf1550,01c00000) Probing pci disk... [IDE]disk-1 7872k cap: 0200 Searching for image... ................................(ELF)... Loading Linux version: Loading image... ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ ........................................................................ .......................................done Unknown bootloader class! type=0x00000000 data=0x00000000 Firmware type: LinuxBIOS Uncompressing Linux... Ok, booting the kernel. BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 0000000000000678 (reserved) BIOS-e820: 0000000000000678 - 00000000000a0000 (usable) BIOS-e820: 0000000000100000 - 0000000001c00000 (usable) 28MB LOWMEM available. hm, page 00000000 reserved twice. On node 0 totalpages: 7168 zone(0): 4096 pages. zone(1): 3072 pages. zone(2): 0 pages. Kernel command line: console=ttyS0,38400n8 debug root=/dev/hda2 Initializing CPU#0 Working around Cyrix MediaGX virtual DMA bugs. Detected 133.636 MHz processor. Linux version 2.4.19 (root at bender) (gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110)) #21 Wed Apr 9 16:46:42 PDT 2003 BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 0000000000000678 (reserved) BIOS-e820: 0000000000000678 - 00000000000a0000 (usable) BIOS-e820: 0000000000100000 - 0000000001c00000 (usable) 28MB LOWMEM available. hm, page 00000000 reserved twice. On node 0 totalpages: 7168 zone(0): 4096 pages. zone(1): 3072 pages. zone(2): 0 pages. Kernel command line: console=ttyS0,38400n8 debug root=/dev/hda2 nitializing CPU#0 Working around Cyrix MediaGX virtual DMA bugs. Detected 133.636 MHz processor. Calibrating delay loop... 266.24 BogoMIPS Memory: 26504k/28672k available (860k kernel code, 1780k reserved, 335k data, 52k init, 0k highmem) Checking if this processor honours the WP bit even in supervisor mode... Ok. Dentry cache hash table entries: 4096 (order: 3, 32768 bytes) Inode cache hash table entries: 2048 (order: 2, 16384 bytes) Mount-cache hash table entries: 512 (order: 0, 4096 bytes) Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes) Page-cache hash table entries: 8192 (order: 3, 32768 bytes) CPU: Before vendor init, caps: 00808131 01818131 00000000, vendor = 8 Working around Cyrix MediaGX virtual DMA bugs. CPU: After vendor init, caps: 00808131 00818131 00000000 00000001 CPU: After generic, caps: 00808131 00818131 00000000 00000001 CPU: Common caps: 00808131 00818131 00000000 00000001 CPU: NSC Geode(TM) Integrated Processor by National Semi stepping 02 Checking 'hlt' instruction... OK. POSIX conformance testing by UNIFIX PCI: Using configuration type 1 PCI: Probing PCI hardware Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer Society NET3.039 Initializing RT netlink socket Starting kswapd Installing knfsd (copyright (C) 1996 okir at monad.swb.de). pty: 256 Unix98 ptys configured Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled ttyS00 at 0x03f8 (irq = 4) is a 16550A hda: 0MB, CHS=0/0/0 RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize 8139too Fast Ethernet driver 0.9.25 PCI: No IRQ known for interrupt pin A of device 00:0d.0. Please try using pci=biosirq. eth0: RealTek RTL8139 Fast Ethernet at 0xc2800000, 00:04:a7:02:0c:af, IRQ 0 eth0: Identified 8139 chip type 'RTL-8139C' NET4: Linux TCP/IP 1.0 for NET4.0 IP Protocols: ICMP, UDP, TCP, IGMP IP: routing cache hash table of 512 buckets, 4Kbytes TCP: Hash tables configured (established 2048 bind 2048) NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. end_request: I/O error, dev 03:02 (hard disk), sector 2 EXT2-fs: unable to read superblock end_request: I/O error, dev 03:02 (hard disk), sector 0 FAT: unable to read boot sector Kernel panic: VFS: Unable to mount root fs on 03:02 From klos1980 at sina.com Wed Apr 9 20:58:00 2003 From: klos1980 at sina.com (klos) Date: Wed Apr 9 20:58:00 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: Message-ID: <000401c2fefe$93b1b340$7ca8a8c0@sh.corp.tyan.com> Hi eric, I find cmos_util can change the value in option_table that was loaded with irq_table together, am I right? but I cannot find the place to use the option_table, how we control the fallback or normal image? Or cmos_util change the cmos value indeed? -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org] On Behalf Of Eric W. Biederman Sent: 2003?4?10? 7:52 To: Alexander Amelkin Cc: LinuxBIOS Subject: Re: [ANNOUNCE] linuxbios utilities. Alexander Amelkin writes: > Hello Eric, > > Wednesday, April 9, 2003, 11:34:26 PM, you wrote: > > EWB> cmos_util which changes the few LinuxBIOS settings from Linux. > > Do we really use the CMOS already? What do we store there? There are several other people who could have answered and this is a question of general interest so back to the list, with this. I am frequently very busy so I do not appreciate LinuxBIOS mail directed exclusively at me. Yes. For normal vs fallback. For serial port baud rates, for the boot order that etherboot should use. Check out the appropriate cmos.layout files. Eric _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From xw at chinahost.com Wed Apr 9 21:04:00 2003 From: xw at chinahost.com (David Xiong) Date: Wed Apr 9 21:04:00 2003 Subject: What the $.02 is? References: <20030409160001.4526.81547.Mailman@nwn.definitive.org> Message-ID: <006101c2ff01$3572c780$b40a0a0a@xw> Hi,guys!I am on the list for a while.But what is the $.02? David > > I think this 'remote desktop in a box' is an application for linuxbios > > that is just crying for someone to sell it as a turnkey system. > > > > my $.02 > > > > ron > > > > > From ebiederman at lnxi.com Wed Apr 9 21:14:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 9 21:14:01 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: <000401c2fefe$93b1b340$7ca8a8c0@sh.corp.tyan.com> References: <000401c2fefe$93b1b340$7ca8a8c0@sh.corp.tyan.com> Message-ID: "klos" writes: > Hi eric, > I find cmos_util can change the value in option_table that was > loaded with irq_table together, am I right? > but I cannot find the place to use the option_table, how we > control the fallback or normal image? Or cmos_util change the cmos value > indeed? option_table describes the layout of options in the cmos. cmos_util changes options in the cmos described by option_table. Eric From hansolofalcon at worldnet.att.net Wed Apr 9 21:15:00 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Wed Apr 9 21:15:00 2003 Subject: What the $.02 is? In-Reply-To: <006101c2ff01$3572c780$b40a0a0a@xw> Message-ID: <000401c2ff02$e138aac0$51a3580c@who5> Hello again from Gregg C Levine Shouldn't it be obvious? That is what happens when someone chips in something that should be obvious. My two credits. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of David Xiong > Sent: Wednesday, April 09, 2003 9:34 PM > To: linuxbios at clustermatic.org > Subject: What the $.02 is? > > Hi,guys!I am on the list for a while.But what is the $.02? > > David > > > > I think this 'remote desktop in a box' is an application for linuxbios > > > that is just crying for someone to sell it as a turnkey system. > > > > > > my $.02 > > > > > > ron > > > > > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From rsmith at bitworks.com Wed Apr 9 21:55:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 9 21:55:01 2003 Subject: What the $.02 is? In-Reply-To: <006101c2ff01$3572c780$b40a0a0a@xw> References: <20030409160001.4526.81547.Mailman@nwn.definitive.org> <006101c2ff01$3572c780$b40a0a0a@xw> Message-ID: <3E94D5CA.4060001@bitworks.com> David Xiong wrote: > Hi,guys!I am on the list for a while.But what is the $.02? > It's short hand for the phrase "My two cents" and basically means my opinion on the subject. I also found the following after doing a little web research: Many years ago, the United States used gold and silver coins as a common currency. The value of the coin would be equal to the value of the gold or silver contained in the coin. Occasionally, an individual would have to pay in increments smaller than the coin's denominations. In this case, the occasion is a person's "ante" in a poker game, which was roughly 25 cents. The gold and silver coins would be split into halfs, quarters and eighths. The eighths of a coin piece was referred to as a "piece of eight" or a "bit". Hence, one's "two bits", as the orignal phrase went, were their ante in a game of poker; their contribution in order to play the game. -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Wed Apr 9 22:28:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 9 22:28:01 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: Message-ID: for the plan9 nodes another use we made of CMOS was to store authentication data. That was pretty neat. ron From rminnich at lanl.gov Wed Apr 9 22:34:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 9 22:34:00 2003 Subject: What the $.02 is? In-Reply-To: <006101c2ff01$3572c780$b40a0a0a@xw> Message-ID: On Thu, 10 Apr 2003, David Xiong wrote: > Hi,guys!I am on the list for a while.But what is the $.02? > > > I think this 'remote desktop in a box' is an application for linuxbios > > > that is just crying for someone to sell it as a turnkey system. > > > > > > my $.02 It is a slang term which means "That is my opinion". Sorry for the obscure slang. :) ron From agnew at cs.umd.edu Wed Apr 9 22:36:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 9 22:36:01 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: Message-ID: <20030409230641.M55525-100000@www.missl.cs.umd.edu> Is there a map of bits of the cmos used? In ADLO we set a few bits because the Bochs BIOS it built to rely on an AMI cmos for settings. On Wed, 9 Apr 2003, ron minnich wrote: > > for the plan9 nodes another use we made of CMOS was to store > authentication data. That was pretty neat. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From terryc at tyanchina.com Wed Apr 9 23:44:00 2003 From: terryc at tyanchina.com (Terry B. Chen) Date: Wed Apr 9 23:44:00 2003 Subject: about BOOT_IDE Message-ID: <441383BA85E81D48964152E537886C9D2BF94D@mail.sh.corp.tyan.com> Dear all: I decide to elf_boot use BOOT_IDE, because the rom is too small to settle a big kernel that support. Can some one tell me the usage of the options such as ONE_TRACK and how to put the elf to hard disk, I think it must be dangerous, it seem to read the sector directly. Someone can teach me a little? Best regards Terry chen -------------- next part -------------- An HTML attachment was scrubbed... URL: From agnew at cs.umd.edu Thu Apr 10 00:02:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 10 00:02:01 2003 Subject: about BOOT_IDE In-Reply-To: <441383BA85E81D48964152E537886C9D2BF94D@mail.sh.corp.tyan.com> Message-ID: <20030410002910.O55525-100000@www.missl.cs.umd.edu> Hi Terry, This was actually discussed just yesturday as well, so in the interest of not covering the subject again so soon... Please see the thread starting with: ---- From: Deepak Kotian To: linuxbios at clustermatic.org Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it --- For three other options for booting from hard disk. Your actually looking at the fourth option, which i forgot earlier. Also keep in mind that using a flash disk, like perhaps a compact flash card on an ide adaptor (cwlinux.com sells such a beast) you can keep your lightning fast boot up speeds! If you don't want to put your kernel image on the beginning sectors of your hard disk, your easiest option is probably etherboot 5.0.6 + the polled ide patch. - Adam Agnew On Thu, 10 Apr 2003, Terry B. Chen wrote: > Dear all: > I decide to elf_boot use BOOT_IDE, because the rom is too small > to settle a big kernel that support. > Can some one tell me the usage of the options such as ONE_TRACK and how > to put the elf to hard disk, I think it must be dangerous, it seem to > read the sector directly. Someone can teach me a little? > > Best regards > Terry chen > From hansolofalcon at worldnet.att.net Thu Apr 10 00:13:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Thu Apr 10 00:13:01 2003 Subject: Warnings while building the Small C compiler that Eric designed Message-ID: <000201c2ff1b$d747c500$7fc3580c@who5> Hello again from Gregg C Levine On my Linux box, which is using Slackware 8.0, along with the reasonably current tool chain for that release, I downloaded the Small C compiler that Eric designed. I then extracted the files, and ran the make program, inside its directory. During the build process, it threw up a series of warnings, nothing threatened the executable which was built so they did not accumulate into the classic error messages that do that. Anyway I've attached a script file explaining them. Eric, care to comment? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: romcc.txt URL: From agnew at cs.umd.edu Thu Apr 10 00:25:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 10 00:25:01 2003 Subject: Warnings while building the Small C compiler that Eric designed In-Reply-To: <000201c2ff1b$d747c500$7fc3580c@who5> Message-ID: <20030410005451.P55525-100000@www.missl.cs.umd.edu> Eric's small-c compiler is still very beta. Most of the warnings you say come from the "#warning" pre-processor feature. They allow the programmer to intentionally make warning messages for his own reference later. (ie Eric knows they're there and its quite intentional. Each one alerts him to something he needs to do still.) The other warning messages were just unused variables. Absolutely nothing to worry about. On Thu, 10 Apr 2003, Gregg C Levine wrote: > Hello again from Gregg C Levine > On my Linux box, which is using Slackware 8.0, along with the > reasonably current tool chain for that release, I downloaded the Small > C compiler that Eric designed. > > I then extracted the files, and ran the make program, inside its > directory. During the build process, it threw up a series of warnings, > nothing threatened the executable which was built so they did not > accumulate into the classic error messages that do that. Anyway I've > attached a script file explaining them. Eric, care to comment? > ------------------- > Gregg C Levine hansolofalcon at worldnet.att.net > ------------------------------------------------------------ > "The Force will be with you...Always." Obi-Wan Kenobi > "Use the Force, Luke."? Obi-Wan Kenobi > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > (This company dedicates this E-Mail to Master Yoda ) > > > > From ijpriya at hotmail.com Thu Apr 10 00:33:00 2003 From: ijpriya at hotmail.com (Priya) Date: Thu Apr 10 00:33:00 2003 Subject: Driver Message-ID: <3E94FB9A.000003.01028@johncy> Hello, I need a driver for audio codec (National's LM4550) Linux OS. Where shall I get one -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: IMSTP.gif Type: image/gif Size: 494 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: BackGrnd.jpg Type: image/jpeg Size: 1431 bytes Desc: not available URL: From steve at nexpath.com Thu Apr 10 00:39:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 10 00:39:01 2003 Subject: about BOOT_IDE In-Reply-To: <441383BA85E81D48964152E537886C9D2BF94D@mail.sh.corp.tyan.com> References: <441383BA85E81D48964152E537886C9D2BF94D@mail.sh.corp.tyan.com> Message-ID: <3E950055.6090701@nexpath.com> Terry B. Chen wrote: > Dear all: > > I decide to elf_boot use BOOT_IDE, because the rom is too small > to settle a big kernel that support. > > Can some one tell me the usage of the options such as ONE_TRACK and how > to put the elf to hard disk, I think it must be dangerous, it seem to > read the sector directly. Someone can teach me a little? > > See pcchips787.config in util/config for complete configuration. option BOOT_IDE=1 This enables booting from IDE, the file to use is linux.bin.gz: option IDE_BOOT_DRIVE=2 If you do not use drive 0 (default), then you can set which drive to boot; (0,1,2,3) are the four standard PC drives: option ONE_TRACK=32 The linux.bin.gz file is put in raw form at partition 1, ie, the first partition on the disk. This is located just past the partition table. The partition table size varies, it is "one track" from the beginning of the disk. "one track" in c/h/s notation is "s" or the number of sectors per track. ONE_TRACK is in sectors, the software multiplies by 512. Most disks are 63 sectors per track (the default), but my CF is 32 sectors per track. eg, the partion table is 63x512 or 32x512 bytes. You can partition your disk as you want, but linux goes raw in partition 1; just make sure partition 1 is big enough, not a problem on today's disks. You could put the linux root file system on partition 2, for example. In pcchips787.config, I put the linux root file system on IDE 0, partition 2 (I was experimenting with linux in partition 1), but I eventually put linux on drive 2 using CF. You are right, copying of linux.bin.gz raw to the partition is dangerous, and something like "cat linux.bin.gz > /dev/hda1" will definitely screw the disk if you put the wrong disk or partition. I recommend a shell script, fingers cannot be trusted. You can also use "dd" but "cat" works. From hansolofalcon at worldnet.att.net Thu Apr 10 00:40:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Thu Apr 10 00:40:01 2003 Subject: Driver In-Reply-To: <3E94FB9A.000003.01028@johncy> Message-ID: <000801c2ff1f$a3e5a340$dec6580c@who5> Hello from Gregg C Levine You might try restating your question, and actually explaining what it is you want. And without sending the message in ugly HTML mail. It tends to slow the list down sometimes. Also some list servers consider that to be spam, and will apply punitive measures. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org] On Behalf Of Priya Sent: Thursday, April 10, 2003 1:06 AM To: linuxbios at clustermatic.org Subject: Driver Hello,???? ?????? I need a driver for audio codec? (National's LM4550) Linux OS. Where shall I get one ____________________________________________________ ? IncrediMail - Email has finally evolved - Click Here From yenneo at oreka.com Thu Apr 10 00:44:01 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Thu Apr 10 00:44:01 2003 Subject: newbie question: where to begin with the code =?iso-8859-1?q?=3F?= Message-ID: <200304100514.HAA16941@mailhub5.isdnet.net> I?m currently trying to understand the code. I?ve tried with the file crt0.s in my build directory but i really have big difficulties. I wonder if it?s the way to begin with this file. Other question : how does work ljmp **,** ? Cause i only know ljmp with one argument. I didn?t find this information. Thanks in advance -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From ijpriya at hotmail.com Thu Apr 10 01:02:00 2003 From: ijpriya at hotmail.com (Devi Priya) Date: Thu Apr 10 01:02:00 2003 Subject: (no subject) Message-ID: _________________________________________________________________ Visit Malaysia. Win a free trip! http://server1.msn.co.in/msnspecials/malaysiatourism/index.asp Have an exotic holiday From ijpriya at hotmail.com Thu Apr 10 01:12:01 2003 From: ijpriya at hotmail.com (Devi Priya) Date: Thu Apr 10 01:12:01 2003 Subject: Codec Message-ID: Hello, I have been assigned the job of writing BIOS. My device is based upon SC1200 chip(National) and it has external devices like ethernet, Audio codec LM4550. My device has linux OS.I need a driver for my LM4550. So I need to know where I shall get this driver either the source or an exe? _________________________________________________________________ Vrroooom?? Fasten your seatbelts. http://server1.msn.co.in/msnSpecials/formula2003/index.asp Get set for F1 2003 From bendany at mistdl.com Thu Apr 10 01:36:00 2003 From: bendany at mistdl.com (bendany) Date: Thu Apr 10 01:36:00 2003 Subject: How to get LinuxBIOS work with Xwindows? Message-ID: <006801c2ffa5$13670300$2a00a8c0@ben> HI,all I am trying to use LinuxBIOS work with my machine, My mainboard is K7SEM. And When I start the Xwindows, I get a blank screen. Does any idea that LinuxBIOS can work with the Xwindows? thanks! ---- Bendany Qian bendany at mistdl.com "Let's make life easy!" -------------- next part -------------- An HTML attachment was scrubbed... URL: From agnew at cs.umd.edu Thu Apr 10 01:43:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 10 01:43:01 2003 Subject: How to get LinuxBIOS work with Xwindows? In-Reply-To: <006801c2ffa5$13670300$2a00a8c0@ben> Message-ID: <20030410021312.C56213-100000@www.missl.cs.umd.edu> I assume you're using a framebuffer right now? Easiest thing to do is search the web for a howto on using X-Windows with the frame buffer server. keywords: xfree86 fbdev On Thu, 10 Apr 2003, bendany wrote: > HI,all > I am trying to use LinuxBIOS work with my machine, My mainboard is K7SEM. > And When I start the Xwindows, I get a blank screen. > Does any idea that LinuxBIOS can work with the Xwindows? > thanks! > ---- > Bendany Qian bendany at mistdl.com > "Let's make life easy!" > From dkotian3 at vsnl.net Thu Apr 10 02:09:00 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Thu Apr 10 02:09:00 2003 Subject: newbie question: where to begin with the code =?iso-8859-1?q?=3F?= Message-ID: <20030410071326.838DD500E6@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From gnuorder at tampabay.rr.com Thu Apr 10 02:26:00 2003 From: gnuorder at tampabay.rr.com (GNUOrder) Date: Thu Apr 10 02:26:00 2003 Subject: Codec In-Reply-To: References: Message-ID: <200304100656.h3A6uqmn027144@ms-smtp-02.tampabay.rr.com> LinuxBIOS wouldn't have anything to do with the audio chip. It leaves things like that up to the OS itself. Find out which linux drivers work with it, alsa, oss or built into the kernel. You should be able to search google for LM4550 and linux to find out what drivers are available. If you are looking for the PnP setup, that would be in the kernel source. If you are using linux as the OS, you could very well skip all that yourself and let linux handle it. GO On Thursday 10 April 2003 01:39, Devi Priya wrote: > Hello, > I have been assigned the job of writing BIOS. My device is based upon > SC1200 chip(National) and it has external devices like ethernet, Audio > codec LM4550. My device has linux OS.I need a driver for my LM4550. So I > need to know where I shall get this driver either the source or an exe? > > > > _________________________________________________________________ > Vrroooom Fasten your seatbelts. > http://server1.msn.co.in/msnSpecials/formula2003/index.asp Get set for F1 > 2003 > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From SAtherton at HPTI.com Thu Apr 10 03:00:01 2003 From: SAtherton at HPTI.com (Atherton, Stephen) Date: Thu Apr 10 03:00:01 2003 Subject: Etherboot/LinuxBios on Supermicro P4DPE Message-ID: <22699FB2708ED611ADA100105AC5C9A0E70744@hpti_main.hpti.com> I'm trying to get etherboot and linuxbios running on a supermicro p4dpe. The config file I'm using is below. It builds okay, and I get a romimage of around 460k (not sure why it isn't 524288), but when I burn it (using flash_and_burn), it doesn't seem to work. I get no output on the serial port. Note that I have BOOT_IDE set to 1 because at this point I would be happy just to see it boot anything, and I have linux on the ide hard disk. Does BOOT_IDE do what I think it does, or is that for something else? Is there something wrong with my config file, or should I use some other flash program for this board? I tried the one from supermicro, and doesn't seem to like my romimage file, it exits with "cannot close filehandle". Also, a few other questions... Does linuxbios cause a beep or do anything so that when the machine is powered on I would have some clue that it is working? Since I'm using the etherboot elf image as a payload, the linux and commandline directives in the config file do nothing, right? What does USE_FALLBACK_IMAGE do exactly? LinuxBios Config file: target supermicro mainboard supermicro/p4dpe option USE_FALLBACK_IMAGE=0 option ROM_SIZE=524288 option ROM_IMAGE_SIZE=49152 option SERIAL_CONSOLE=1 option TTYS0_BAUD=9600 option TTYS0_BASE=0x3f8 option TTYS0_LCS=0x3 option DEFAULT_CONSOLE_LOGLEVEL=9 option MAXIMUM_CONSOLE_LOGLEVEL=8 option USE_ELF_BOOT=1 option BOOT_IDE=1 payload ../eepro100.elf option CPU_CLOCK_MULTIPLIER=XEON_X17 option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON linux /usr/src/linux-2.4.18-14 commandline root=/dev/hda2 console=ttyS0,9600 single -------------- next part -------------- An HTML attachment was scrubbed... URL: From terryc at tyanchina.com Thu Apr 10 05:02:00 2003 From: terryc at tyanchina.com (Terry B. Chen) Date: Thu Apr 10 05:02:00 2003 Subject: about BOOT_IDE Message-ID: <441383BA85E81D48964152E537886C9D2BF9A5@mail.sh.corp.tyan.com> Thank you for your advice, I did as you say, but the ide_init() failed, do you meet the problem ? or you have change the code? -----Original Message----- From: Steve Gehlbach [mailto:steve at nexpath.com] Sent: 2003?4?10? 13:26 To: Terry B. Chen Cc: LinuxBIOS Subject: Re: about BOOT_IDE Terry B. Chen wrote: > Dear all: > > I decide to elf_boot use BOOT_IDE, because the rom is too small > to settle a big kernel that support. > > Can some one tell me the usage of the options such as ONE_TRACK and how > to put the elf to hard disk, I think it must be dangerous, it seem to > read the sector directly. Someone can teach me a little? > > See pcchips787.config in util/config for complete configuration. option BOOT_IDE=1 This enables booting from IDE, the file to use is linux.bin.gz: option IDE_BOOT_DRIVE=2 If you do not use drive 0 (default), then you can set which drive to boot; (0,1,2,3) are the four standard PC drives: option ONE_TRACK=32 The linux.bin.gz file is put in raw form at partition 1, ie, the first partition on the disk. This is located just past the partition table. The partition table size varies, it is "one track" from the beginning of the disk. "one track" in c/h/s notation is "s" or the number of sectors per track. ONE_TRACK is in sectors, the software multiplies by 512. Most disks are 63 sectors per track (the default), but my CF is 32 sectors per track. eg, the partion table is 63x512 or 32x512 bytes. You can partition your disk as you want, but linux goes raw in partition 1; just make sure partition 1 is big enough, not a problem on today's disks. You could put the linux root file system on partition 2, for example. In pcchips787.config, I put the linux root file system on IDE 0, partition 2 (I was experimenting with linux in partition 1), but I eventually put linux on drive 2 using CF. You are right, copying of linux.bin.gz raw to the partition is dangerous, and something like "cat linux.bin.gz > /dev/hda1" will definitely screw the disk if you put the wrong disk or partition. I recommend a shell script, fingers cannot be trusted. You can also use "dd" but "cat" works. From terryc at tyanchina.com Thu Apr 10 05:56:01 2003 From: terryc at tyanchina.com (Terry B. Chen) Date: Thu Apr 10 05:56:01 2003 Subject: 533Mhz FSB of E7501 Message-ID: <441383BA85E81D48964152E537886C9D2BF9B2@mail.sh.corp.tyan.com> We tested s2723 that use E7501 source code; I find linuxbios can not support dual channel memory. Then I try to disable hyper threading, the spot_check meet error again? I have compare the E7501 with E7500 Bios spec, it have a lot of differences between these two chipsets including RCOMP programming. I suppose it need to modify in many site. This is the two the lspci file. I suppose you just need 00:00;0 400 FSB: 00:00.0 Host bridge: Intel Corp.: Unknown device 254c (rev 01) 00: 86 80 4c 25 06 01 90 00 01 00 00 06 00 00 00 00 10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 30: 00 00 00 00 40 00 00 00 00 00 00 00 00 00 00 00 40: 09 00 05 11 00 00 00 00 00 00 00 00 00 00 00 00 50: 04 60 0d 00 00 00 00 00 00 10 11 01 00 00 33 33 60: 00 00 08 08 08 08 08 08 00 00 00 00 00 00 00 00 70: 00 03 00 00 00 00 00 00 1f 04 01 39 79 02 67 20 80: b1 0b 71 00 00 00 00 00 00 98 10 d2 0d 00 00 00 90: 00 00 00 00 00 00 00 00 55 05 55 05 01 1a 38 00 a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 c0: 44 c0 50 11 00 20 ff 03 00 00 00 00 00 00 00 00 d0: 02 28 00 0e 03 00 00 33 80 09 31 b5 00 00 01 01 e0: 1d 1d 00 00 00 00 00 00 56 46 00 00 00 00 00 00 f0: 00 00 00 00 74 00 30 40 40 0f 00 00 00 00 00 00 533 MHZ: 00:00.0 Host bridge: Intel Corp.: Unknown device 254c (rev 01) 00: 86 80 4c 25 06 01 90 00 01 00 00 06 00 00 00 00 10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 30: 00 00 00 00 40 00 00 00 00 00 00 00 00 00 00 00 40: 09 00 05 11 00 00 00 00 00 00 00 00 00 00 00 00 50: 04 60 0d 00 00 00 00 00 00 10 11 01 00 00 33 33 60: 00 00 08 08 08 08 08 08 00 00 00 00 00 00 00 00 70: 00 03 00 00 00 00 00 00 04 02 01 28 79 02 67 20 80: 62 06 71 00 00 00 00 00 00 98 10 d2 8d 00 00 00 90: 00 00 00 00 00 00 00 00 55 05 55 05 01 1a 38 00 a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 c0: 44 c0 50 11 00 20 ff 03 00 00 00 00 00 00 00 00 d0: 02 28 00 0e 03 00 00 33 80 09 31 b5 00 00 01 01 e0: 1d 1d 00 00 00 00 00 00 54 4a 00 00 00 00 00 00 f0: 00 00 00 00 74 00 30 40 40 0f 00 00 00 00 00 00 -----Original Message----- From: steven james [mailto:pyro at linuxlabs.com] Sent: 2003?4?8? 23:16 To: Terry B. Chen Cc: linuxbios at clustermatic.org Subject: Re: 533Mhz FSB of E7501 Greetings, Most likely, there is an undocumented register issue in the northbridge. With the OEM BIOS, can you send me an lspci with a 400 FSB and a 533 FSB. That should narrow things down. G'day, sjames On Tue, 8 Apr 2003, Terry B. Chen wrote: > Dear all; > I am trying the Tyan 2723(E7501), but I find it will report > spot_check error if the board run in 533 FSB. But if the CPU or the > memory is 400 FSB it will be ok. Does anyone else meet the same problem? > > Best regards > Terry chen > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From stepan at suse.de Thu Apr 10 08:05:01 2003 From: stepan at suse.de (Stefan Reinauer) Date: Thu Apr 10 08:05:01 2003 Subject: PIRQ problems on Geode GX1 board. In-Reply-To: <001201c2fefd$ee848b40$6501a8c0@pink> References: <001201c2fefd$ee848b40$6501a8c0@pink> Message-ID: <20030410123539.GC28399@suse.de> * Joey Nelson [030410 03:10]: > I ran getpir on the target with its standard bios, and got a good > irq_tables.c, and HAVE_PIRQ_TABLE=1 is set in the Config. Please check whether the PIRQ table is really ok. I dumped a "good" pirq table from my geode box, but unfortunately, it was for the wrong chipset (gx5520 instead of 5530) In the Supertek ST-3WT target i fixed this by setting interrupts manually to the mapping on that board. I guess this is what the award bios on there did before instead of using the wrong pirq table. Stefan -- The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offense. -- E. W. Dijkstra From pyro at linuxlabs.com Thu Apr 10 08:42:01 2003 From: pyro at linuxlabs.com (steven james) Date: Thu Apr 10 08:42:01 2003 Subject: Etherboot/LinuxBios on Supermicro P4DPE In-Reply-To: <22699FB2708ED611ADA100105AC5C9A0E70744@hpti_main.hpti.com> Message-ID: Greetings, LinuxBIOS for that board (and many others) supports two copies of LinuxBIOS, fallback and primary. The fallback image is intended to be old and stable and it's purpose is to act as a sort of rescue image. The primary image is the one that gets updated to the latest and greatest, and is meant to be more versatile and complete. The fallback image is loaded at the top of the flash, and always recieves control at power on or reset. It does very minimal setup (switch to 32bit flat memory protected mode), checks and clears the CMOS boot bit, and if it was set, jumps to the primary image. This way, if the primary fails, hitting reset will boot from the fallback image alone. From pyro at linuxlabs.com Thu Apr 10 09:13:01 2003 From: pyro at linuxlabs.com (steven james) Date: Thu Apr 10 09:13:01 2003 Subject: 533Mhz FSB of E7501 In-Reply-To: <441383BA85E81D48964152E537886C9D2BF9B2@mail.sh.corp.tyan.com> Message-ID: Greetings, I'll have to do a bit of studying, but the one thing that jumps out at me immediatly is bit 7 of register 0x8c. and 0x78-0x7b. The big difference there is that DRAM timings are backed off a bit, and 0x8c selects for 133 (266) MHz operation. There will likely be a bit of trial and error with differences to the RAM configuration since there is no way to dump those settings out (unless someone knows something I don't). 0xe8-9 and 0x80-3 are not documented by Intel. 0x80 appears to involve CAS latency according to Eric (it certainly seems that way). That setting is at line 189 of e7501/raminit.inc. G'day, sjames On Thu, 10 Apr 2003, Terry B. Chen wrote: > We tested s2723 that use E7501 source code; I find linuxbios can not support dual channel memory. Then I try to disable hyper threading, the spot_check meet error again?? > I have compare the E7501 with E7500 Bios spec, it have a lot of differences between these two chipsets including RCOMP programming. I suppose it need to modify in many site. > This is the two the lspci file. I suppose you just need 00:00;0 > 400 FSB: > 00:00.0 Host bridge: Intel Corp.: Unknown device 254c (rev 01) > 00: 86 80 4c 25 06 01 90 00 01 00 00 06 00 00 00 00 > 10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > 30: 00 00 00 00 40 00 00 00 00 00 00 00 00 00 00 00 > 40: 09 00 05 11 00 00 00 00 00 00 00 00 00 00 00 00 > 50: 04 60 0d 00 00 00 00 00 00 10 11 01 00 00 33 33 > 60: 00 00 08 08 08 08 08 08 00 00 00 00 00 00 00 00 > 70: 00 03 00 00 00 00 00 00 1f 04 01 39 79 02 67 20 > 80: b1 0b 71 00 00 00 00 00 00 98 10 d2 0d 00 00 00 > 90: 00 00 00 00 00 00 00 00 55 05 55 05 01 1a 38 00 > a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > c0: 44 c0 50 11 00 20 ff 03 00 00 00 00 00 00 00 00 > d0: 02 28 00 0e 03 00 00 33 80 09 31 b5 00 00 01 01 > e0: 1d 1d 00 00 00 00 00 00 56 46 00 00 00 00 00 00 > f0: 00 00 00 00 74 00 30 40 40 0f 00 00 00 00 00 00 > 533 MHZ: > 00:00.0 Host bridge: Intel Corp.: Unknown device 254c (rev 01) > 00: 86 80 4c 25 06 01 90 00 01 00 00 06 00 00 00 00 > 10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > 30: 00 00 00 00 40 00 00 00 00 00 00 00 00 00 00 00 > 40: 09 00 05 11 00 00 00 00 00 00 00 00 00 00 00 00 > 50: 04 60 0d 00 00 00 00 00 00 10 11 01 00 00 33 33 > 60: 00 00 08 08 08 08 08 08 00 00 00 00 00 00 00 00 > 70: 00 03 00 00 00 00 00 00 04 02 01 28 79 02 67 20 > 80: 62 06 71 00 00 00 00 00 00 98 10 d2 8d 00 00 00 > 90: 00 00 00 00 00 00 00 00 55 05 55 05 01 1a 38 00 > a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > c0: 44 c0 50 11 00 20 ff 03 00 00 00 00 00 00 00 00 > d0: 02 28 00 0e 03 00 00 33 80 09 31 b5 00 00 01 01 > e0: 1d 1d 00 00 00 00 00 00 54 4a 00 00 00 00 00 00 > f0: 00 00 00 00 74 00 30 40 40 0f 00 00 00 00 00 00 > > > -----Original Message----- > From: steven james [mailto:pyro at linuxlabs.com] > Sent: 2003??4??8?? 23:16 > To: Terry B. Chen > Cc: linuxbios at clustermatic.org > Subject: Re: 533Mhz FSB of E7501 > > Greetings, > > Most likely, there is an undocumented register issue in the northbridge. > > With the OEM BIOS, can you send me an lspci with a 400 FSB and a 533 > FSB. That should narrow things down. > > G'day, > sjames > > > On Tue, 8 Apr 2003, Terry B. Chen wrote: > > > Dear all; > > I am trying the Tyan 2723(E7501), but I find it will report > > spot_check error if the board run in 533 FSB. But if the CPU or the > > memory is 400 FSB it will be ok. Does anyone else meet the same problem? > > > > Best regards > > Terry chen > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Thu Apr 10 10:34:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 10 10:34:00 2003 Subject: Etherboot/LinuxBios on Supermicro P4DPE In-Reply-To: Message-ID: to expand on fallback vs. primary. You need to build TWO linuxbioses, a fallback and a primary. The fallback is 64k. The primary is (romsize-64k). To build a romimage, you build the fallback, build the primary, then: cat primary/romimage fallback/romimage > final_romimage flash_rom final_romimage ron From stuge-linuxbios at cdy.org Thu Apr 10 10:59:00 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Thu Apr 10 10:59:00 2003 Subject: Codec In-Reply-To: References: Message-ID: <20030410152036.GA11695@foo.birdnet.se> On Thu, Apr 10, 2003 at 05:39:34AM +0000, Devi Priya wrote: > Hello, > I have been assigned the job of writing BIOS. My device is based upon > SC1200 chip(National) and it has external devices like ethernet, Audio > codec LM4550. My device has linux OS.I need a driver for my LM4550. So I > need to know where I shall get this driver either the source or an exe? Hi. The SCx2xx series have this feature where binary-only code gets executed in SMM (System Management Mode) when you (or kernel drivers) access certain addresses. It's called VSA, Virtual System Architecture. There's also VSA2 which adds something. NSC supplies some binary VSMs (Virtual System Modules) with their royalty free BIOS kit BLDT (BootLoader Development Toolkit), aka XpressLOADER. Among others, there are VSMs for enabling audio and video. LinuxBIOS does not currently implement VSA or VSA2. I see no reason why it couldn't, though, I even think it could be made as a separate ELF object loaded in the same way the bare bones toolkit is, I seem to remember the idea of being able to plug multiple ELF objects into the bootup sequence but this still requires some twiddling, right? NSC offers documentation on VSA/VSA2 but I'm not sure about availability.. NSC also offer a Linux audio driver for the SCx2xx family but this driver requires the XpressAUDIO VSM to work. //Peter From jerj at coplanar.net Thu Apr 10 11:31:01 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Thu Apr 10 11:31:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <003d01c2feca$645d9a20$d92041db@vsnl.net> References: <003d01c2feca$645d9a20$d92041db@vsnl.net> Message-ID: <1049990528.1246.0.camel@contact.skynet.coplanar.net> There is work being done to allow the EEPROM socket of certain network cards to be used as a flash programmer. Also see the devbios project. Regards, Jeremy On Wed, 2003-04-09 at 14:52, Deepak Kotian wrote: > Hi, > > This may be a very basic simple question, but I do not know a very > good answer/solution > to do this. > > If I want to burn LinuxBIOS, I can use the burn_mtd script , it > may write it > successfully as well, but if linuxbios is not correctly built. I > cannot get the > any O.S started on this machine, which means I would need to > replace the > BIOS FLASH ROM with a one with proper BIOS and bring up the > machine. > Is there a way I can re-program the old BIOS on to the corrupted > FLASH ROMS. > Please note,I do not have a PLC or any kind of FLASH programmer > with me. Can > I do without it. > > If there are any suggestions, please let me know. > > Thanks and Regards > Deepak > > > From dpilon at atipa.com Thu Apr 10 12:18:00 2003 From: dpilon at atipa.com (Denis Pilon) Date: Thu Apr 10 12:18:00 2003 Subject: What do I need ... Message-ID: <1049993320.21821.120.camel@atipa-dp> What do I need to get started with linuxbios on a new motherboard ? I am trying to get linuxbios working on a Supermicro X5DPE-G2. But don't know where to start. Denis -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From agnew at cs.umd.edu Thu Apr 10 13:04:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 10 13:04:01 2003 Subject: What do I need ... In-Reply-To: <1049993320.21821.120.camel@atipa-dp> Message-ID: <20030410133531.U58066-100000@www.missl.cs.umd.edu> The output of # lspci helps tell you which parts of the motherboard are already supported by LinuxBIOS. On 10 Apr 2003, Denis Pilon wrote: > What do I need to get started with linuxbios on a new motherboard ? > > I am trying to get linuxbios working on a Supermicro X5DPE-G2. > > But don't know where to start. > > Denis > > From ebiederman at lnxi.com Thu Apr 10 13:30:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 10 13:30:01 2003 Subject: [ANNOUNCE] linuxbios utilities. In-Reply-To: <20030409230641.M55525-100000@www.missl.cs.umd.edu> References: <20030409230641.M55525-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > Is there a map of bits of the cmos used? Yes. It tends to leave out things like Ron's uses but what LinuxBIOS and etherboot use are in the map. They cmos.layout files are the source of the map, and the map is provided in the LinuxBIOS table. > In ADLO we set a few bits because the Bochs BIOS it built to rely on an > AMI cmos for settings. You probably want to take a look at how either etherboot or cmos_util get the map. We can probably included your Bochs BIOS settings as well. So far I have steered clear of legacy locations to reduce the chance of conflicts with hard codes but ultimately it is better to put everyone into the map. A big thing on the todo list is to figure out how to move some of the settings into the Rom chip. Very useful for board without a CMOS battery or at least without a reliable one. Eric From dkotian3 at vsnl.net Thu Apr 10 13:49:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 10 13:49:00 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS References: <003d01c2feca$645d9a20$d92041db@vsnl.net> <016c01c2fecf$764df0e0$d92041db@vsnl.net> Message-ID: <00f301c2ff8d$982aab40$323e41db@vsnl.net> Hi , Are there any more suggestions or easier ways to do the burning process so that I avoid any kind of Flash programmer or emulator. Basically which would save time during the development process. Buying an emulator or flash programmer would be the last thing for me. I guess, The below concept is not very feasible. Please let me know your views on the same. Thanks and Regards Deepak ----- Original Message ----- From: "Deepak Kotian" To: "Eric W. Biederman" Cc: Sent: Thursday, April 10, 2003 1:07 AM Subject: Re: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS > Thanks Eric. > I have a spare FLASH ROM. > Could you please elaborate on the process of "hot swap", is it that I > can replace the FLASH ROM with the corrupted one with > the backup flash ROM, then once it boots up, I can just replace the > corrupted FLASH ROM while the O.S. is up and burn the > backed up bios from the harddisk onto the corrupted FLAH ROMS. > Is this correct and feasible, please elaborate.. > > > ----- Original Message ----- > From: "Eric W. Biederman" > To: "Deepak Kotian" > Cc: > Sent: Thursday, April 10, 2003 1:00 AM > Subject: Re: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS > > > > "Deepak Kotian" writes: > > > > > Hi, > > > > > > This may be a very basic simple question, but I do not know a very > good > > > answer/solution > > > > > > to do this. > > > > > > If I want to burn LinuxBIOS, I can use the burn_mtd script , it may > write it > > > > > > successfully as well, but if linuxbios is not correctly built. I > cannot get > > > the > > > > > > any O.S started on this machine, which means I would need to replace > the > > > BIOS FLASH ROM with a one with proper BIOS and bring up the machine. > > > Is there a way I can re-program the old BIOS on to the corrupted > FLASH ROMS. > > > > > > Please note,I do not have a PLC or any kind of FLASH programmer with > me. Can > > > > > > I do without it. > > > > Usually it is recommeded for bootstraping and development that you either > > have a spare rom chip, which you can hot swap in to flash, or that you > have > > an external burner. > > > > > > Eric > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Thu Apr 10 14:11:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 10 14:11:01 2003 Subject: What does one derive from lspci -x -x References: <20030410133531.U58066-100000@www.missl.cs.umd.edu> Message-ID: <014a01c2ff90$c30edd60$323e41db@vsnl.net> Hi , What does one derive from lspci -v -v and lspci -x , lspci -x gives the hexdump, is this hexdump helpfull for me to use this data for the chipset which is not implement in linuxbios, if someone can share some more details on it, it would be great. Definitely, lspci gives out the chip information and one scan the linuxbios tree to see if the chipset is supported or not. But lspci -x is not very clear to me yet. Thanks and Regards Deepak ----- Original Message ----- From: "Adam Agnew" To: "Denis Pilon" Cc: Sent: Thursday, April 10, 2003 11:06 PM Subject: Re: What do I need ... > The output of > # lspci > helps tell you which parts of the motherboard are already supported by > LinuxBIOS. > > > On 10 Apr 2003, Denis Pilon wrote: > > > What do I need to get started with linuxbios on a new motherboard ? > > > > I am trying to get linuxbios working on a Supermicro X5DPE-G2. > > > > But don't know where to start. > > > > Denis > > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rminnich at lanl.gov Thu Apr 10 14:23:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 10 14:23:00 2003 Subject: What do I need ... In-Reply-To: <1049993320.21821.120.camel@atipa-dp> Message-ID: On 10 Apr 2003, Denis Pilon wrote: > I am trying to get linuxbios working on a Supermicro X5DPE-G2. do an lspci and find a very similar motherboard. Then make a src/mainboard/supermicro/x5dpe-g2 directory and populate it with files from the similar board. The modify to suit. ron From rminnich at lanl.gov Thu Apr 10 14:31:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 10 14:31:00 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <00f301c2ff8d$982aab40$323e41db@vsnl.net> Message-ID: I can rotate flash parts in seconds on my geode nodes, so I don't see any problem with just flashing the part in the motherboard. ron From dkotian3 at vsnl.net Thu Apr 10 14:38:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 10 14:38:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS References: Message-ID: <01ba01c2ff94$7c3b50e0$323e41db@vsnl.net> Thanks for the information. Sorry, Could you please elaborate on "geode nodes" , howto set it up,etc. Thanks and Regards Deepak ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: Sent: Friday, April 11, 2003 12:31 AM Subject: Re: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS > I can rotate flash parts in seconds on my geode nodes, so I don't see any > problem with just flashing the part in the motherboard. > > ron > From rsmith at bitworks.com Thu Apr 10 14:55:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Thu Apr 10 14:55:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <00f301c2ff8d$982aab40$323e41db@vsnl.net> References: <003d01c2feca$645d9a20$d92041db@vsnl.net> <016c01c2fecf$764df0e0$d92041db@vsnl.net> <00f301c2ff8d$982aab40$323e41db@vsnl.net> Message-ID: <3E95C4FF.2070102@bitworks.com> Deepak Kotian wrote: > so that I avoid any kind of Flash programmer or emulator. Basically which > would save time during the development process. Buying an emulator or > flash programmer would be the last thing for me. If you are really wanting to save time then an emulator is by _far_ one of the best investments you could possibly make. The speedup in flash programming time alone will probally pay back what you spend. -- Richard A. Smith rsmith at bitworks.com From aip at cwlinux.com Thu Apr 10 21:18:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Thu Apr 10 21:18:01 2003 Subject: Suggested mechanism of burning a LinuxBIOS on FLASH ROMS In-Reply-To: <00f301c2ff8d$982aab40$323e41db@vsnl.net>; from Deepak Kotian on Thu, Apr 10, 2003 at 11:47:39PM +0530 References: <003d01c2feca$645d9a20$d92041db@vsnl.net> <016c01c2fecf$764df0e0$d92041db@vsnl.net> <00f301c2ff8d$982aab40$323e41db@vsnl.net> Message-ID: <20030411094831.A10255@mail.cwlinux.com> Deepak, > Are there any more suggestions or easier ways to do the burning process > so that I avoid any kind of Flash programmer or emulator. Basically which > would save time during the development process. Buying an emulator or > flash programmer would be the last thing for me. > I guess, The below concept is not very feasible. BIOS Savior is a pretty good alternative. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From terryc at tyanchina.com Fri Apr 11 00:20:01 2003 From: terryc at tyanchina.com (Terry B. Chen) Date: Fri Apr 11 00:20:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it Message-ID: <441383BA85E81D48964152E537886C9D2BF9FE@mail.sh.corp.tyan.com> Hi, I have read the ADLO install and README files, I know a lot from the files. But I have some questions. If I use Linuxbios + ADLO, which options should be set? Should the ouput payout replace the linuxkernel? ADLO support all chipset? Have some guide to teach how to merge ADLO with linuxbios? Thank you! Best regards Terry From adam at cfar.umd.edu Fri Apr 11 00:45:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Fri Apr 11 00:45:01 2003 Subject: Burning BIOS on FLASH ROMS, is the kernel also needs to burnt on it In-Reply-To: <441383BA85E81D48964152E537886C9D2BF9FE@mail.sh.corp.tyan.com> Message-ID: <20030411010113.X59960-100000@www.missl.cs.umd.edu> > I have read the ADLO install and README files, I know a lot from > the files. But I have some questions. If I use Linuxbios + ADLO, which > options should be set? Should the ouput payout replace the linuxkernel? you configure it whatever way you want to use it. aldo+bochs bios are an ELF image. so if you want both in one chip, then it just LinuxBIOS with (ADLO+BOCHS) elf image after it. just like you would have etherboot (in elf format) after it or linux kernel (in elf format as well). of course different configurations are possible. for example I used to place ALDO+BOCHS on hdd (with etherboot used to load it from hdd) > ADLO support all chipset? you must know how to setup shadow ram for your chipset. other that that it should work with any chipset that is supported by linuxbios. > Have some guide to teach how to merge ADLO with linuxbios? well if you are familar with using linxubios it is just an incremental step. you simply replace your usual ELF module with an ELF Module which as (ADLO+BOCHS BIOS) inside of it. still somewhat sticky steep right now is video bios. We still use binary only module. So if there's something wrong with it, there's no easy way to fix it :-( -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From shubhangi.jadhav at patni.com Fri Apr 11 06:36:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Fri Apr 11 06:36:01 2003 Subject: How does the OS boot OS Message-ID: Hi, I'm trying to understand the working of LinuxBIOS. The FAQ says that it uses a real OS is used to load the OS. Can someone elaborate on this ? Is it that the kernel image which you burn along with linuxbios on the BIOS Flash ROM is copied to RAM, performs the hardware initialization and then it invokes the other OS which in my case would be linux? How is the other OS invoked ? Thanks, Shubhangi From pyro at linuxlabs.com Fri Apr 11 08:19:00 2003 From: pyro at linuxlabs.com (steven james) Date: Fri Apr 11 08:19:00 2003 Subject: What do I need ... In-Reply-To: Message-ID: Greetings, intel/Clearwater533 will be quite similar. Note that the e7501 code has reported problems with 533MHz CPUs. It looks like there are just a few registers in the northbridge that will need to be tweaked. I need to get a pair of CPUs to fix that unless someone beats me to it (patches greatfully accepted :-) G'day, sjames On Thu, 10 Apr 2003, ron minnich wrote: > On 10 Apr 2003, Denis Pilon wrote: > > > I am trying to get linuxbios working on a Supermicro X5DPE-G2. > > do an lspci and find a very similar motherboard. Then make a > src/mainboard/supermicro/x5dpe-g2 directory and populate it with files > from the similar board. The modify to suit. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From SAtherton at HPTI.com Fri Apr 11 08:32:01 2003 From: SAtherton at HPTI.com (Atherton, Stephen) Date: Fri Apr 11 08:32:01 2003 Subject: Still not having success with SM P4DPE Message-ID: <22699FB2708ED611ADA100105AC5C9A0E70748@hpti_main.hpti.com> Ron and James, thanks for clueing me in on the fallback image stuff. I'm still not quite doing something right though, it seems. My config files are below. I based these on the samples in mainboard/supermicro/p4dpe. The fallback image is 81k and contains the etherboot payload...this doesn't seem right to me but I don't know what else to do. If I don't specify the etherboot payload, it tries to use vmlinux which is much, much bigger. So I run the config script on both files, then go to the fallback/ and primary/ directories and run make, and I get a primary romimage of 458752 bytes and a fallback image of 81920 bytes. That doesn't seem right. What am I doing wrong? Oh and though I was doubtful, I did do a "cat primary/romimage fallback/romimage > finalimage" and burned the final image. Also, from the looks of the makeprocess my two image files are basically the same binary stuff but one has a lot of null padding. They both contain the eepro100 elf image and linuxbios.rom... Anyone see what is wrong with this picture? -Steve fallback.config -------------------------------------------- target ./fallback mainboard supermicro/p4dpe option USE_FALLBACK_IMAGE=1 option ROM_SIZE=524288 option ROM_IMAGE_SIZE=49152 option SERIAL_CONSOLE=1 option TTYS0_BAUD=9600 option TTYS0_BASE=0x3f8 option TTYS0_LCS=0x3 option DEFAULT_CONSOLE_LOGLEVEL=9 option MAXIMUM_CONSOLE_LOGLEVEL=6 option USE_ELF_BOOT=1 option USE_GENERIC_ROM=1 payload ../eepro100.elf option CPU_CLOCK_MULTIPLIER=XEON_X8 option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON primary.config -------------------------------------------------- target primary mainboard supermicro/p4dpe linux /usr/src/linux-2.4.18-14 commandline root=/dev/hda2 console=ttyS0,9600 single option USE_FALLBACK_IMAGE=0 option ROM_SIZE=524288 option ROM_IMAGE_SIZE=49152 option SERIAL_CONSOLE=1 option TTYS0_BAUD=9600 option TTYS0_BASE=0x3f8 option TTYS0_LCS=0x3 option DEFAULT_CONSOLE_LOGLEVEL=9 option MAXIMUM_CONSOLE_LOGLEVEL=8 option USE_ELF_BOOT=1 option BOOT_IDE=1 payload ../eepro100.elf option CPU_CLOCK_MULTIPLIER=XEON_X8 option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON -------------- next part -------------- An HTML attachment was scrubbed... URL: From pyro at linuxlabs.com Fri Apr 11 08:55:01 2003 From: pyro at linuxlabs.com (steven james) Date: Fri Apr 11 08:55:01 2003 Subject: Still not having success with SM P4DPE In-Reply-To: <22699FB2708ED611ADA100105AC5C9A0E70748@hpti_main.hpti.com> Message-ID: Greetings, It looks like the fallback image is being built wrong. This can be hacked together manually. linuxbios.strip should be the actual LinuxBIOS code. The objective is to assemble a 64K (exactly) block containing: 1. etherboot 2. null padding (can be generated with dd if=/dev/zero) 3. linuxbios.strip. I just use calctool to figure out how much padding to use and cat eepro100.elf pad linuxbios.strip >fallback.bin the 64K fallback.bin can then be appended to the primary image to get the full rom image. That's a little ugly, but it should get the job done. Just as a final verification, in linuxbios.map, ZKERNEL_START for the fallback image should be 0xffff0000. That is whjere it will look for the fallback etherboot. Also, it will ljmp to ZKERNEL_START-8 to transfer control to the primary image. I'm slowly trying to disentangle that on the boards I have. Unfortunatly, I don't have a P4DPE. G'day, sjames On Fri, 11 Apr 2003, Atherton, Stephen wrote: > Ron and James, thanks for clueing me in on the fallback image stuff. I'm > still not quite doing something right though, it seems. > > My config files are below. I based these on the samples in > mainboard/supermicro/p4dpe. The fallback image is 81k and contains the > etherboot payload...this doesn't seem right to me but I don't know what else > to do. If I don't specify the etherboot payload, it tries to use vmlinux > which is much, much bigger. > > So I run the config script on both files, then go to the fallback/ and > primary/ directories and run make, and I get a primary romimage of 458752 > bytes and a fallback image of 81920 bytes. That doesn't seem right. What > am I doing wrong? Oh and though I was doubtful, I did do a "cat > primary/romimage fallback/romimage > finalimage" and burned the final image. > > > Also, from the looks of the makeprocess my two image files are basically the > same binary stuff but one has a lot of null padding. They both contain the > eepro100 elf image and linuxbios.rom... > > Anyone see what is wrong with this picture? > > -Steve > > > > > fallback.config -------------------------------------------- > > target ./fallback > mainboard supermicro/p4dpe > option USE_FALLBACK_IMAGE=1 > option ROM_SIZE=524288 > option ROM_IMAGE_SIZE=49152 > option SERIAL_CONSOLE=1 > option TTYS0_BAUD=9600 > option TTYS0_BASE=0x3f8 > option TTYS0_LCS=0x3 > option DEFAULT_CONSOLE_LOGLEVEL=9 > option MAXIMUM_CONSOLE_LOGLEVEL=6 > option USE_ELF_BOOT=1 > option USE_GENERIC_ROM=1 > payload ../eepro100.elf > option CPU_CLOCK_MULTIPLIER=XEON_X8 > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > > > primary.config -------------------------------------------------- > > target primary > mainboard supermicro/p4dpe > linux /usr/src/linux-2.4.18-14 > commandline root=/dev/hda2 console=ttyS0,9600 single > option USE_FALLBACK_IMAGE=0 > option ROM_SIZE=524288 > option ROM_IMAGE_SIZE=49152 > option SERIAL_CONSOLE=1 > option TTYS0_BAUD=9600 > option TTYS0_BASE=0x3f8 > option TTYS0_LCS=0x3 > option DEFAULT_CONSOLE_LOGLEVEL=9 > option MAXIMUM_CONSOLE_LOGLEVEL=8 > option USE_ELF_BOOT=1 > option BOOT_IDE=1 > payload ../eepro100.elf > option CPU_CLOCK_MULTIPLIER=XEON_X8 > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Fri Apr 11 09:52:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 09:52:01 2003 Subject: How does the OS boot OS In-Reply-To: Message-ID: On Fri, 11 Apr 2003, Shubhangi Jadhav wrote: > Is it that the kernel image which you burn along with linuxbios on the > BIOS Flash ROM is copied to RAM, performs the hardware initialization and > then it invokes the other OS which in my case would be linux? How is the > other OS invoked ? that is pretty much how it works. Here at LANL we use a system call two kernel monte; do a google on it. ron From rminnich at lanl.gov Fri Apr 11 09:53:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 09:53:00 2003 Subject: What do I need ... In-Reply-To: Message-ID: On Fri, 11 Apr 2003, steven james wrote: > intel/Clearwater533 will be quite similar. Note that the e7501 code has > reported problems with 533MHz CPUs. It looks like there are just a few > registers in the northbridge that will need to be tweaked. I need to get a > pair of CPUs to fix that unless someone beats me to it (patches greatfully > accepted :-) I called Intel yesterday about this 533 Mhz. issue and the lack of accurate docs. The person I talked to said he will try to help. ron From SAtherton at HPTI.com Fri Apr 11 09:53:04 2003 From: SAtherton at HPTI.com (Atherton, Stephen) Date: Fri Apr 11 09:53:04 2003 Subject: Still not having success with SM P4DPE Message-ID: <22699FB2708ED611ADA100105AC5C9A0E7074A@hpti_main.hpti.com> Linuxbios.strip is 49152 bytes, and eepro100.elf is 22368. So that gives a total of 71520...larger than 65536. So one of them is too big. Any idea which? -Steve -----Original Message----- From: steven james To: Atherton, Stephen Cc: 'linuxbios at clustermatic.org' Sent: 4/11/03 9:25 AM Subject: Re: Still not having success with SM P4DPE Greetings, It looks like the fallback image is being built wrong. This can be hacked together manually. linuxbios.strip should be the actual LinuxBIOS code. The objective is to assemble a 64K (exactly) block containing: 1. etherboot 2. null padding (can be generated with dd if=/dev/zero) 3. linuxbios.strip. I just use calctool to figure out how much padding to use and cat eepro100.elf pad linuxbios.strip >fallback.bin the 64K fallback.bin can then be appended to the primary image to get the full rom image. That's a little ugly, but it should get the job done. Just as a final verification, in linuxbios.map, ZKERNEL_START for the fallback image should be 0xffff0000. That is whjere it will look for the fallback etherboot. Also, it will ljmp to ZKERNEL_START-8 to transfer control to the primary image. I'm slowly trying to disentangle that on the boards I have. Unfortunatly, I don't have a P4DPE. G'day, sjames On Fri, 11 Apr 2003, Atherton, Stephen wrote: > Ron and James, thanks for clueing me in on the fallback image stuff. I'm > still not quite doing something right though, it seems. > > My config files are below. I based these on the samples in > mainboard/supermicro/p4dpe. The fallback image is 81k and contains the > etherboot payload...this doesn't seem right to me but I don't know what else > to do. If I don't specify the etherboot payload, it tries to use vmlinux > which is much, much bigger. > > So I run the config script on both files, then go to the fallback/ and > primary/ directories and run make, and I get a primary romimage of 458752 > bytes and a fallback image of 81920 bytes. That doesn't seem right. What > am I doing wrong? Oh and though I was doubtful, I did do a "cat > primary/romimage fallback/romimage > finalimage" and burned the final image. > > > Also, from the looks of the makeprocess my two image files are basically the > same binary stuff but one has a lot of null padding. They both contain the > eepro100 elf image and linuxbios.rom... > > Anyone see what is wrong with this picture? > > -Steve > > > > > fallback.config -------------------------------------------- > > target ./fallback > mainboard supermicro/p4dpe > option USE_FALLBACK_IMAGE=1 > option ROM_SIZE=524288 > option ROM_IMAGE_SIZE=49152 > option SERIAL_CONSOLE=1 > option TTYS0_BAUD=9600 > option TTYS0_BASE=0x3f8 > option TTYS0_LCS=0x3 > option DEFAULT_CONSOLE_LOGLEVEL=9 > option MAXIMUM_CONSOLE_LOGLEVEL=6 > option USE_ELF_BOOT=1 > option USE_GENERIC_ROM=1 > payload ../eepro100.elf > option CPU_CLOCK_MULTIPLIER=XEON_X8 > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > > > primary.config -------------------------------------------------- > > target primary > mainboard supermicro/p4dpe > linux /usr/src/linux-2.4.18-14 > commandline root=/dev/hda2 console=ttyS0,9600 single > option USE_FALLBACK_IMAGE=0 > option ROM_SIZE=524288 > option ROM_IMAGE_SIZE=49152 > option SERIAL_CONSOLE=1 > option TTYS0_BAUD=9600 > option TTYS0_BASE=0x3f8 > option TTYS0_LCS=0x3 > option DEFAULT_CONSOLE_LOGLEVEL=9 > option MAXIMUM_CONSOLE_LOGLEVEL=8 > option USE_ELF_BOOT=1 > option BOOT_IDE=1 > payload ../eepro100.elf > option CPU_CLOCK_MULTIPLIER=XEON_X8 > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Fri Apr 11 09:57:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 09:57:00 2003 Subject: Still not having success with SM P4DPE In-Reply-To: <22699FB2708ED611ADA100105AC5C9A0E70748@hpti_main.hpti.com> Message-ID: to simplify this I'm going to just show you my files, which work. I just built a linuxbios yesterday that worked fine on the P4DPE-G2. my romimage sizes for this are 64K and (512k-64k) Note that fallback size is 40K which with the 24k etherbooth gives you 64k. I think your normal file has a typo on the rom image size. ron --- fallback ## This will make a target directory of ./fallback ## This is relative to where the configuration file resides in the filesystem target ./p4dpe-g2-fallback mainboard supermicro/p4dpe/g2 ## Build a fallback not a normal image. option USE_FALLBACK_IMAGE=1 ## Build an image for a 512KB rom ## ./fallback/romimage is just the last 64KB which we reserve for the fallback image. option ROM_SIZE=524288 #option ROM_SIZE=1048576 ## Select the maximum size the linuxBIOS code can compile to. ## Allow linuxBIOS to be up to 48KB in size option ROM_IMAGE_SIZE=40960 ## ### The Serial Console ## ## Hardware flow control is currently ignored. ## Enable the Serial Console option SERIAL_CONSOLE=1 ## Select the serial console baud rate. option TTYS0_BAUD=115200 #option TTYS0_BAUD=57600 #option TTYS0_BAUD=38400 #option TTYS0_BAUD=19200 #option TTYS0_BAUD=9600 #option TTYS0_BAUD=4800 #option TTYS0_BAUD=2400 #option TTYS0_BAUD=1200 # Select the serial console base port option TTYS0_BASE=0x3f8 # Select the serial protocol # This defaults to 8 data bits, 1 stop bit, and no parity option TTYS0_LCS=0x3 ## ### Select the linuxBIOS loglevel ## ## EMERG 1 system is unusable ## ALERT 2 action must be taken immediately ## CRIT 3 critical conditions ## ERR 4 error conditions ## WARNING 5 warning conditions ## NOTICE 6 normal but significant condition ## INFO 7 informational ## DEBUG 8 debug-level messages ## SPEW 9 Way too many details ## Request this level of debugging output option DEFAULT_CONSOLE_LOGLEVEL=7 ## At a maximum only compile in this level of debugging option MAXIMUM_CONSOLE_LOGLEVEL=7 ## Use the elf bootloader option USE_ELF_BOOT=1 ## Select the boot device option USE_GENERIC_ROM=1 #option BOOT_FLOPPY=1 #option USE_SERIAL_FILL_INBUF=1 #option BOOT_IDE=1 # Load etherboot with the elf bootloader # The payload command is relative the build directory # So .. is the directory this config file resides in payload ../lnxieepro100.ebi ## ## Cpu Speed ## option CPU_CLOCK_MULTIPLIER=XEON_X8 #option CPU_CLOCK_MULTIPLIER=XEON_X9 #option CPU_CLOCK_MULTIPLIER=XEON_X10 #option CPU_CLOCK_MULTIPLIER=XEON_X11 #option CPU_CLOCK_MULTIPLIER=XEON_X12 #option CPU_CLOCK_MULTIPLIER=XEON_X13 #option CPU_CLOCK_MULTIPLIER=XEON_X14 #option CPU_CLOCK_MULTIPLIER=XEON_X15 #option CPU_CLOCK_MULTIPLIER=XEON_X16 #option CPU_CLOCK_MULTIPLIER=XEON_X17 #option CPU_CLOCK_MULTIPLIER=XEON_X18 #option CPU_CLOCK_MULTIPLIER=XEON_X19 #option CPU_CLOCK_MULTIPLIER=XEON_X19 #option CPU_CLOCK_MULTIPLIER=XEON_X20 #option CPU_CLOCK_MULTIPLIER=XEON_X21 #option CPU_CLOCK_MULTIPLIER=XEON_X22 #option CPU_CLOCK_MULTIPLIER=XEON_X23 ## ## Select power on after power fail setting option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON #option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON --- normal ## This will make a target directory of ./normal ## This is relative to where the configuration file resides in the filesystem target ./p4dpe-g2-normal mainboard supermicro/p4dpe/g2 ## Build a normal not a fallback image. option USE_FALLBACK_IMAGE=0 ## Build an image for a 512KB rom ## ./normal/romimage is the entire rom image except for the last 64KB ## which are reserved for the fallback image. option ROM_SIZE=524288 #option ROM_SIZE=1048576 ## Select the maximum size the linuxBIOS code can compile to. ## Allow linuxBIOS to be up to 48KB in size option ROM_IMAGE_SIZE=49152 ## ### The Serial Console ## ## Hardware flow control is currently ignored. ## Enable the Serial Console option SERIAL_CONSOLE=1 ## Select the serial console baud rate. option TTYS0_BAUD=115200 #option TTYS0_BAUD=57600 #option TTYS0_BAUD=38400 #option TTYS0_BAUD=19200 #option TTYS0_BAUD=9600 #option TTYS0_BAUD=4800 #option TTYS0_BAUD=2400 #option TTYS0_BAUD=1200 # Select the serial console base port option TTYS0_BASE=0x3f8 # Select the serial protocol # This defaults to 8 data bits, 1 stop bit, and no parity option TTYS0_LCS=0x3 ## ### Select the linuxBIOS loglevel ## ## EMERG 1 system is unusable ## ALERT 2 action must be taken immediately ## CRIT 3 critical conditions ## ERR 4 error conditions ## WARNING 5 warning conditions ## NOTICE 6 normal but significant condition ## INFO 7 informational ## DEBUG 8 debug-level messages ## SPEW 9 Way too many details ## Request this level of debugging output option DEFAULT_CONSOLE_LOGLEVEL=4 ## At a maximum only compile in this level of debugging option MAXIMUM_CONSOLE_LOGLEVEL=5 ## Use the elf bootloader option USE_ELF_BOOT=1 ## Select the boot device option USE_GENERIC_ROM=1 #option BOOT_FLOPPY=1 #option USE_SERIAL_FILL_INBUF=1 #option BOOT_IDE=1 # Load etherboot with the elf bootloader # The payload command is relative the build directory # So .. is the directory this config file resides in payload ../lnxieepro100.ebi ## ## Cpu Speed ## #option CPU_CLOCK_MULTIPLIER=XEON_X8 #option CPU_CLOCK_MULTIPLIER=XEON_X9 #option CPU_CLOCK_MULTIPLIER=XEON_X10 #option CPU_CLOCK_MULTIPLIER=XEON_X11 #option CPU_CLOCK_MULTIPLIER=XEON_X12 #option CPU_CLOCK_MULTIPLIER=XEON_X13 #option CPU_CLOCK_MULTIPLIER=XEON_X14 #option CPU_CLOCK_MULTIPLIER=XEON_X15 #option CPU_CLOCK_MULTIPLIER=XEON_X16 option CPU_CLOCK_MULTIPLIER=XEON_X17 #option CPU_CLOCK_MULTIPLIER=XEON_X18 #option CPU_CLOCK_MULTIPLIER=XEON_X19 #option CPU_CLOCK_MULTIPLIER=XEON_X19 #option CPU_CLOCK_MULTIPLIER=XEON_X20 #option CPU_CLOCK_MULTIPLIER=XEON_X21 #option CPU_CLOCK_MULTIPLIER=XEON_X22 #option CPU_CLOCK_MULTIPLIER=XEON_X23 ## ## Select power on after power fail setting option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON #option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON From rminnich at lanl.gov Fri Apr 11 10:01:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 10:01:01 2003 Subject: Still not having success with SM P4DPE In-Reply-To: <22699FB2708ED611ADA100105AC5C9A0E7074A@hpti_main.hpti.com> Message-ID: On Fri, 11 Apr 2003, Atherton, Stephen wrote: > Linuxbios.strip is 49152 bytes, and eepro100.elf is 22368. So that gives a > total of 71520...larger than 65536. So one of them is too big. Any idea > which? use those files I just sent. You need to back up the linxubios size to 40K. etherboot has grown just a tad. ron From steve at nexpath.com Fri Apr 11 11:00:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 11 11:00:01 2003 Subject: How does the OS boot OS In-Reply-To: References: Message-ID: <3E96E149.6080807@nexpath.com> ron minnich wrote: > that is pretty much how it works. Here at LANL we use a system call two > kernel monte; do a google on it. > Does monte work on 2.4? What about kexec for 2.4? -Steve From rminnich at lanl.gov Fri Apr 11 11:03:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 11:03:00 2003 Subject: How does the OS boot OS In-Reply-To: <3E96E149.6080807@nexpath.com> Message-ID: On Fri, 11 Apr 2003, Steve Gehlbach wrote: > Does monte work on 2.4? What about kexec for 2.4? both work on 2.4, and kexec works on 2.5 ron From steve at nexpath.com Fri Apr 11 11:23:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 11 11:23:00 2003 Subject: How does the OS boot OS In-Reply-To: References: Message-ID: <3E96E6CC.5060209@nexpath.com> ron minnich wrote: > On Fri, 11 Apr 2003, Steve Gehlbach wrote: >>Does monte work on 2.4? What about kexec for 2.4? > > > both work on 2.4, and kexec works on 2.5 > I must be getting the wrong files or something. The last time I looked at kmonte the docs said 2.2 and 2.3 only, and I could not get it to compile under 2.4. And kexec, I could only find the files for 2.5. Maybe you can help me find the right locations. This is where I was getting kmonte: http://www.scyld.com/products/beowulf/software/monte.html From rminnich at lanl.gov Fri Apr 11 11:32:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 11:32:01 2003 Subject: How does the OS boot OS In-Reply-To: <3E96E6CC.5060209@nexpath.com> Message-ID: On Fri, 11 Apr 2003, Steve Gehlbach wrote: > This is where I was getting kmonte: > http://www.scyld.com/products/beowulf/software/monte.html The Scyld stuff is for older Linux kernels; you need to go to the sourceforge and look for beoboot and bproc, etc. ron From steve at nexpath.com Fri Apr 11 13:07:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 11 13:07:00 2003 Subject: How does the OS boot OS In-Reply-To: References: Message-ID: <3E96FF38.9010607@nexpath.com> ron minnich wrote: > The Scyld stuff is for older Linux kernels; you need to go to the > sourceforge and look for beoboot and bproc, etc. > Thanks, I didn't see it in beowulf, but I did find http://sourceforge.net/projects/monte/, and did a "cvs co" of monte, and built it and insmod'd it on 2.4, and all seems well, though I haven't actually executed it yet. Thanks for the hints; it's too bad that the scyld stuff comes up on google first, I looked to use this a year ago on 2.4 and gave up assuming it only worked on 2.2. Kind of cryptic to find the right files; the monte project says production stable, but no files released. Someone needs to put a note on scyld. -Steve From dpilon at atipa.com Fri Apr 11 16:59:00 2003 From: dpilon at atipa.com (Denis Pilon) Date: Fri Apr 11 16:59:00 2003 Subject: Etherboot ... Message-ID: <1050096575.1531.317.camel@atipa-dp> OK. Got an image built using etherboot-5.0.8 I think it works, but the e1000 driver is to old so it can't network boot. Using the etherboot-5.1.3, the elf image is to big. How do I shrink the romimage to fit 524288 ? Denis -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rminnich at lanl.gov Fri Apr 11 18:44:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 18:44:00 2003 Subject: mkelfImage problem Message-ID: I've got an elfImage that has a text segment at 800000. When I run mkelfImage from the latest CVS, the elfImage that is created has that chunk at 800000 moved down to 100000. The segment in elfImage at 100000 has the exact size of the segment at 800000 in the input elf file. OK, anybody seen anything like this? any hints :-) ron From ebiederman at lnxi.com Fri Apr 11 21:05:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 11 21:05:01 2003 Subject: mkelfImage problem In-Reply-To: References: Message-ID: ron minnich writes: > I've got an elfImage that has a text segment at 800000. Is this a linux kernel? If not mkelfImage 1.x does not apply. > When I run mkelfImage from the latest CVS, the elfImage that is created > has that chunk at 800000 moved down to 100000. The segment in elfImage at > 100000 has the exact size of the segment at 800000 in the input elf file. > > OK, anybody seen anything like this? any hints :-) Have you tried the latest mkelfImage from my site. Sorry for being slow in this sync up I have a nasty cold right now, plus quite a few things are catching up to me that I put off while writing romcc. Eric From rminnich at lanl.gov Fri Apr 11 21:18:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 21:18:01 2003 Subject: mkelfImage problem In-Reply-To: Message-ID: On 11 Apr 2003, Eric W. Biederman wrote: > ron minnich writes: > > > I've got an elfImage that has a text segment at 800000. > > Is this a linux kernel? If not mkelfImage 1.x does not apply. never mind, it was a big mistake on my part. I was reconstructing last summer's plan9 work and my brain glitched :-) Sorry if I made your cold worse :-) ron From ebiederman at lnxi.com Fri Apr 11 21:25:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 11 21:25:01 2003 Subject: mkelfImage problem In-Reply-To: References: Message-ID: ron minnich writes: > On 11 Apr 2003, Eric W. Biederman wrote: > > > ron minnich writes: > > > > > I've got an elfImage that has a text segment at 800000. > > > > Is this a linux kernel? If not mkelfImage 1.x does not apply. > > never mind, it was a big mistake on my part. > > I was reconstructing last summer's plan9 work and my brain glitched :-) Ah. If you can send me a patch for mkelfImage-2.3 I would appreciate it. -t plan9 sounds nice. > Sorry if I made your cold worse :-) No problem I just figured an explanation was useful. Eric From aip at cwlinux.com Fri Apr 11 21:29:00 2003 From: aip at cwlinux.com (Andrew Ip) Date: Fri Apr 11 21:29:00 2003 Subject: How does the OS boot OS In-Reply-To: ; from ron minnich on Fri, Apr 11, 2003 at 09:32:44AM -0600 References: <3E96E149.6080807@nexpath.com> Message-ID: <20030412095947.A25650@mail.cwlinux.com> > both work on 2.4, and kexec works on 2.5 I have backported kexec to 2.4.19, and 2.4.20 can use this patch also. Although it isn't the lastest, it works ok. I have tried to port the latest to 2.4, but it requires much more work than I expect. It is what I have so far. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From ebiederman at lnxi.com Fri Apr 11 21:55:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 11 21:55:01 2003 Subject: [COMMIT] LNXI synch up Pass 2 Message-ID: This gets the easy stuff. There were a few oddities in the mainboard code that I think should be fixed but the work so I left them. Most notably we now have to mainboard entries for the s2466 and the p4dpeg2. Someone who has more energy can sit down and compare them. - intel.h removed false comment - p4dpe sync with LNXI tree * Make switching to fallback mode more robust Added boot_countdown to count the number of times booting has failed. * Added boot index to allow fully forcing the boot device with cmos options * Added missing irqs to the mptable. - p4dpr sync with LNXI tree * Make switching to fallback mode more robust Added boot_countdown to count the number of times booting has failed. * Added boot index to allow fully forcing the boot device with cmos options - tyan/guiness repair and update the Config file - northbridge/amd/amd76x/ * Support for up to 4GB of ram. * Correct handling of strange dimm sizes - northbridge/intel/E7500 * Tunned settings for better memory performance - northbridge/intel/E7501 * Fixed calculations based on a 100Mhz to use a 133Mhz clock. * Replaced hard codes for the supermicro x5dpr * Misc bug fixes - src/ram/spotcheck.inc * Removed someones temporary debugging code - src/sdram/generci_zero_ecc_sdram.inc * Handle addresses > 2GB - src/southbridge/amd/amd768/ * Changed references to the amd766 to the amd768 * misc fixes * Added a count to the failover code so we trigger fallback much less easily. * more attempts to disable the amd768 watchdog... * reset the board on a timeout reading from the smbus. - src/southbridge/intel/82801 * added a count of boot failers so we trigger fallback mode much leass easily. - Added motherboards p4dpeg2, x5dpr, s2466, s2469 Eric From rminnich at lanl.gov Fri Apr 11 23:10:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 11 23:10:01 2003 Subject: [COMMIT] LNXI synch up Pass 2 In-Reply-To: Message-ID: On 11 Apr 2003, Eric W. Biederman wrote: > - Added motherboards p4dpeg2, x5dpr, s2466, s2469 So did you keep or delete p4dpe/g2? I was hoping to keep that in there. ron From ebiederman at lnxi.com Fri Apr 11 23:12:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 11 23:12:00 2003 Subject: [COMMIT] LNXI synch up Pass 2 In-Reply-To: References: Message-ID: ron minnich writes: > On 11 Apr 2003, Eric W. Biederman wrote: > > > - Added motherboards p4dpeg2, x5dpr, s2466, s2469 > > So did you keep or delete p4dpe/g2? I was hoping to keep that in there. See the other part that says we have two copies of the p4dpeg2 and the s2466. Removing the redundancies is more than I have the oomph for at the moment. I will investigate that a little later. I am more worried about a few pieces of the baremetal tool kit that have slipped into the main tree. Eric From NEWBELL7 at magicn.com Sat Apr 12 00:44:00 2003 From: NEWBELL7 at magicn.com (NEWBELL7 at magicn.com) Date: Sat Apr 12 00:44:00 2003 Subject: ADLO for EPIA Message-ID: <669801c300b2$929bf960$81cda8c0@magicn.net> Hello! I'm also trying to enable VGA console for EPIA. Maybe, I think it difficult that I get some VGA BIOS from anywhere. I just read that you've enabled EPIA VGA. And your boot flow is linuxbios -> etherboot -> ADLO -> ? . By the way, why are you using ADLO ? I heard that ADLO was only bootloader, such as etherboot, doesn't it ? and what is the Boch ? I read all mailing list , but didn't understand its apearance. can you explain me that ? And please ! can you tell me any hints about the successful story of EPIA's VGA ? Also, What Should I do to enable VGA ? Maybe, I survey frame buffer .... How long will it take to enable VGA and Console , then see any penguin ??? I'm waiting for reply. Thank you. ======================================================================== === This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C2EC72.5263E320 Content-Type: text/plain; charset="iso-2022-kr" Thanks. It works and I can see something (just splash of colors) on my CRT. The screen, though it is not usable, is brought up by original VGABIOS using ADLO. Heechul. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pyro at linuxlabs.com Sat Apr 12 07:30:01 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 12 07:30:01 2003 Subject: Still not having success with SM P4DPE In-Reply-To: <22699FB2708ED611ADA100105AC5C9A0E7074A@hpti_main.hpti.com> Message-ID: Greetings, Possibly both :-) First, try: strip eepro100.elf objcopy -R .note -R .comment eepro100.elf eepro100.strip That should shrink it a bit. If that isn't enough, you can probably squeeze a few more bytes out of linuxbios.strip by adjusting ROM_IMAGE_SIZE in the mainboard directory's Config file. Unfortunatly the build process currently faces a sort of chicken and egg problem. The resolution is to use that variable to pick a size. Too small and the build will fail, too large and you get null padding inside the image (which can't be cut out since that would screw up jmp offsets. Between those two things, you should be able to get the two pieces to fit neatly into the boot block. G'day, sjames On Fri, 11 Apr 2003, Atherton, Stephen wrote: > Linuxbios.strip is 49152 bytes, and eepro100.elf is 22368. So that gives a > total of 71520...larger than 65536. So one of them is too big. Any idea > which? > > -Steve > > -----Original Message----- > From: steven james > To: Atherton, Stephen > Cc: 'linuxbios at clustermatic.org' > Sent: 4/11/03 9:25 AM > Subject: Re: Still not having success with SM P4DPE > > Greetings, > > It looks like the fallback image is being built wrong. > > This can be hacked together manually. linuxbios.strip should be the > actual > LinuxBIOS code. The objective is to assemble a 64K (exactly) block > containing: > > 1. etherboot > 2. null padding (can be generated with dd if=/dev/zero) > 3. linuxbios.strip. > > I just use calctool to figure out how much padding to use and > cat eepro100.elf pad linuxbios.strip >fallback.bin > > the 64K fallback.bin can then be appended to the primary image to get > the > full rom image. > > That's a little ugly, but it should get the job done. > > Just as a final verification, in linuxbios.map, ZKERNEL_START for the > fallback image should be 0xffff0000. That is whjere it will look for the > fallback etherboot. Also, it will ljmp to ZKERNEL_START-8 to transfer > control to the primary image. I'm slowly trying to disentangle that on > the > boards I have. Unfortunatly, I don't have a P4DPE. > > G'day, > sjames > > > On Fri, 11 Apr 2003, Atherton, Stephen wrote: > > > Ron and James, thanks for clueing me in on the fallback image stuff. > I'm > > still not quite doing something right though, it seems. > > > > My config files are below. I based these on the samples in > > mainboard/supermicro/p4dpe. The fallback image is 81k and contains > the > > etherboot payload...this doesn't seem right to me but I don't know > what else > > to do. If I don't specify the etherboot payload, it tries to use > vmlinux > > which is much, much bigger. > > > > So I run the config script on both files, then go to the fallback/ and > > primary/ directories and run make, and I get a primary romimage of > 458752 > > bytes and a fallback image of 81920 bytes. That doesn't seem right. > What > > am I doing wrong? Oh and though I was doubtful, I did do a "cat > > primary/romimage fallback/romimage > finalimage" and burned the final > image. > > > > > > Also, from the looks of the makeprocess my two image files are > basically the > > same binary stuff but one has a lot of null padding. They both > contain the > > eepro100 elf image and linuxbios.rom... > > > > Anyone see what is wrong with this picture? > > > > -Steve > > > > > > > > > > fallback.config -------------------------------------------- > > > > target ./fallback > > mainboard supermicro/p4dpe > > option USE_FALLBACK_IMAGE=1 > > option ROM_SIZE=524288 > > option ROM_IMAGE_SIZE=49152 > > option SERIAL_CONSOLE=1 > > option TTYS0_BAUD=9600 > > option TTYS0_BASE=0x3f8 > > option TTYS0_LCS=0x3 > > option DEFAULT_CONSOLE_LOGLEVEL=9 > > option MAXIMUM_CONSOLE_LOGLEVEL=6 > > option USE_ELF_BOOT=1 > > option USE_GENERIC_ROM=1 > > payload ../eepro100.elf > > option CPU_CLOCK_MULTIPLIER=XEON_X8 > > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > > > > > > primary.config -------------------------------------------------- > > > > target primary > > mainboard supermicro/p4dpe > > linux /usr/src/linux-2.4.18-14 > > commandline root=/dev/hda2 console=ttyS0,9600 single > > option USE_FALLBACK_IMAGE=0 > > option ROM_SIZE=524288 > > option ROM_IMAGE_SIZE=49152 > > option SERIAL_CONSOLE=1 > > option TTYS0_BAUD=9600 > > option TTYS0_BASE=0x3f8 > > option TTYS0_LCS=0x3 > > option DEFAULT_CONSOLE_LOGLEVEL=9 > > option MAXIMUM_CONSOLE_LOGLEVEL=8 > > option USE_ELF_BOOT=1 > > option BOOT_IDE=1 > > payload ../eepro100.elf > > option CPU_CLOCK_MULTIPLIER=XEON_X8 > > option MAINBOARD_POWER_ON_AFTER_POWER_FAIL=MAINBOARD_POWER_ON > > > > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From pyro at linuxlabs.com Sat Apr 12 07:37:00 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 12 07:37:00 2003 Subject: What do I need ... In-Reply-To: Message-ID: Greetings, Excellent! Perhaps that will help. Meanwhile, I've ordered a pair of 533 FSB CPUs for my board. If all else fails, trial and error (and error and error) will provide the answers. From pyro at linuxlabs.com Sat Apr 12 14:18:00 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 12 14:18:00 2003 Subject: [commit] tiger-i7501 Message-ID: Greetings, Might as well join the party. I've comitted the tiger-i7501. For anyone working on a 7501 board, Eric's made some adjustments to the raminit code, it's worth updating your tree and trying the 533MHz FSB again. G'day, sjames -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Sat Apr 12 15:10:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sat Apr 12 15:10:01 2003 Subject: [commit] tiger-i7501 In-Reply-To: Message-ID: what is the part # of the i7501? this is great to see 7501 commits coming in! ron From pyro at linuxlabs.com Sat Apr 12 15:39:00 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 12 15:39:00 2003 Subject: [commit] tiger-i7501 In-Reply-To: Message-ID: Greetings, I suppose that might help! It's the S2723 G'day, sjames On Sat, 12 Apr 2003, ron minnich wrote: > what is the part # of the i7501? > > this is great to see 7501 commits coming in! > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From hansolofalcon at worldnet.att.net Sat Apr 12 23:53:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sat Apr 12 23:53:01 2003 Subject: How does the OS boot OS In-Reply-To: <20030412095947.A25650@mail.cwlinux.com> Message-ID: <000b01c30174$a3cf1ec0$f5c8580c@who5> Hello again from Gregg C Levine I might as well chime in on this discussion right here. Andrew are you certain that your patch for the 2.4.19 kernel for the kernel exec function can be applied to the 2.4.20 kernel? Granted I should think it would work, with out too much fuss, except from patch itself, but I'm not an expert. And more importantly, the make elf image tool, normally requires as one of its parameters, a ram disk. Any suggestions for building one? Does your LinuxBIOS SDK have anything to offer for this one, Andrew? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of Andrew Ip > Sent: Friday, April 11, 2003 10:00 PM > To: ron minnich > Cc: Steve Gehlbach; Shubhangi Jadhav; linuxbios at clustermatic.org > Subject: Re: How does the OS boot OS > > > > both work on 2.4, and kexec works on 2.5 > I have backported kexec to 2.4.19, and 2.4.20 can use this patch also. > Although it isn't the lastest, it works ok. I have tried to port the > latest to 2.4, but it requires much more work than I expect. It > is what I have so far. > > -Andrew > > -- > Andrew Ip > Email: aip at cwlinux.com > Tel: (852) 2542 2046 > Fax: (852) 2542 2036 > Mobile: (852) 9201 9866 > > Cwlinux Limited > Unit 202B 2/F Lai Cheong Factory Building, > 479-479A Castle Peak Road, > Lai Chi Kok, Kowloon, > Hong Kong. > > Tel: (852)2542 2046 > Fax: (852)2542 2036 > > For public pgp key, please obtain it from http://www.keyserver.net/en. > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From hansolofalcon at worldnet.att.net Sun Apr 13 02:21:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sun Apr 13 02:21:01 2003 Subject: Kexec and kernel oops Message-ID: <000001c30189$42487ce0$24c4580c@who5> Hello from Gregg C Levine Here is a strange bit of events, I just finished recompiling my Linux box's 2.4.20 kernel, after applying the kernel patch supplied by Andrew Ip. The patch applied with the usual amount of fuss. No problems surfaced during the build process. Nor with anything else. Except for the kernel exec tool. Since I wasn't sure if Andrew had back ported the ones from the 2.5 series to the 2.4.20, and 2.4.19 ones, I chose the original command line one, and built that. I chose the original 2..2.19 kernel as the one to exec. Original that is, because this version of Slackware is 8.0, and that's the version that works for it to install correctly here. Also the tools that I have used, are from that release version. Each of the different versions of the make elf Image tools from 1.6, to 2.0 that I tried,. gave me an oops message, and caused the program to abort with a segmentation fault. Here's a cut and paste of my oops message from the dmesg log: Script started on Sun Apr 13 02:36:24 2003 root at who4:/usr/src/LoBos/work# kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b6f18 edi: 0002df48 ebp: c0eafd40 esp: c0eafd20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 131, stackpage=c0eaf000) Stack: 04000000 00010740 c10b6f9c c0eafd9c 00000080 c10b6f1c 00000282 c105f2c0 c0eafd9c c014f00e c10b6f1c 00000080 c0eafdd0 c0eafe9c 00000000 00010740 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b69c8 edi: 0002df48 ebp: c0e8dd40 esp: c0e8dd20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 146, stackpage=c0e8d000) Stack: 04000000 00010740 c10b6a4c c0e8dd9c 00000080 c10b69cc 00000286 c105f2c0 c0e8dd9c c014f00e c10b69cc 00000080 c0e8ddd0 c0e8de9c 00000000 00010740 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b6f18 edi: 0002df48 ebp: c0e8dd40 esp: c0e8dd20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 150, stackpage=c0e8d000) Stack: 04000000 00010740 c10b6f9c c0e8dd9c 00000080 c10b6f1c 00000282 c105f2c0 c0e8dd9c c014f00e c10b6f1c 00000080 c0e8ddd0 c0e8de9c 00000000 00010740 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b69c8 edi: 0002df48 ebp: c1e83d40 esp: c1e83d20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 170, stackpage=c1e83000) Stack: 04000000 00010740 c10b6a4c c1e83d9c 00000080 c10b69cc 00000286 c105f2c0 c1e83d9c c014f00e c10b69cc 00000080 c1e83dd0 c1e83e9c 00000000 00010740 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b6f18 edi: 0002df48 ebp: c1727d40 esp: c1727d20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 213, stackpage=c1727000) Stack: 04000000 000107a0 c10b6f9c c1727d9c 00000080 c10b6f1c 00000282 c105f2c0 c1727d9c c014f00e c10b6f1c 00000080 c1727dd0 c1727e9c 00000000 000107a0 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b69c8 edi: 0002df48 ebp: c1727d40 esp: c1727d20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 840, stackpage=c1727000) Stack: 04000000 000107a0 c10b6a4c c1727d9c 00000080 c10b69cc 00000286 c105f2c0 c1727d9c c014f00e c10b69cc 00000080 c1727dd0 c1727e9c 00000000 000107a0 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b6f18 edi: 0002df48 ebp: c19d9d40 esp: c19d9d20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 845, stackpage=c19d9000) Stack: 04000000 000107a0 c10b6f9c c19d9d9c 00000080 c10b6f1c 00000282 c105f2c0 c19d9d9c c014f00e c10b6f1c 00000080 c19d9dd0 c19d9e9c 00000000 000107a0 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 kernel BUG at slab.c:1437! invalid operand: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010016 eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 esi: c10b6f18 edi: 0002df48 ebp: c1a6fd40 esp: c1a6fd20 ds: 0018 es: 0018 ss: 0018 Process kexec (pid: 868, stackpage=c1a6f000) Stack: 04000000 000107a0 c10b6f9c c1a6fd9c 00000080 c10b6f1c 00000282 c105f2c0 c1a6fd9c c014f00e c10b6f1c 00000080 c1a6fdd0 c1a6fe9c 00000000 000107a0 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 00000001 Call Trace: [] [] [] [] [] [] [] [] [] [] Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 root at who4:/usr/src/LoBos/work# exit Script done on Sun Apr 13 02:36:28 2003 As you can see, that's what caused the message. This time I tripped the kernel exec on a remote console, and watched the main one. The main one displayed the oops, and the remote console gave a segmentation fault as its error. So? Anybody out there, recognize the fault? I personally think it has to do with my choice of kernel exec command line tools. Also, versions 2.1, and 2.2. and 2.3 of the make elf Image too, didn't work here. I'll grab that as a script file for the latest release, and present it later. Andrew did you indeed back port the 2.5 series of kernel exec patch functions to the 2..4.19, and 2.4.20, as I have now decided? ------------------- P.S. If that's the case, then don't worry about it any. It is my fault for choosing the wrong tool to use first. --------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) From adam at cfar.umd.edu Sun Apr 13 02:24:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sun Apr 13 02:24:00 2003 Subject: Kexec and kernel oops In-Reply-To: <000001c30189$42487ce0$24c4580c@who5> Message-ID: <20030413025719.N70156-100000@www.missl.cs.umd.edu> for oops report to be usefull it needs to be decoded. see linux/Documentation/oops-tracing.txt basically you need to turn them thuru ksymoops program first. On Sun, 13 Apr 2003, Gregg C Levine wrote: > Hello from Gregg C Levine > Here is a strange bit of events, I just finished recompiling my Linux > box's 2.4.20 kernel, after applying the kernel patch supplied by > Andrew Ip. The patch applied with the usual amount of fuss. No > problems surfaced during the build process. Nor with anything else. > Except for the kernel exec tool. Since I wasn't sure if Andrew had > back ported the ones from the 2.5 series to the 2.4.20, and 2.4.19 > ones, I chose the original command line one, and built that. > > I chose the original 2..2.19 kernel as the one to exec. Original that > is, because this version of Slackware is 8.0, and that's the version > that works for it to install correctly here. Also the tools that I > have used, are from that release version. > Each of the different versions of the make elf Image tools from 1.6, > to 2.0 that I tried,. gave me an oops message, and caused the program > to abort with a segmentation fault. Here's a cut and paste of my oops > message from the dmesg log: > Script started on Sun Apr 13 02:36:24 2003 > > root at who4:/usr/src/LoBos/work# > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b6f18 edi: 0002df48 ebp: c0eafd40 esp: c0eafd20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 131, stackpage=c0eaf000) > Stack: 04000000 00010740 c10b6f9c c0eafd9c 00000080 c10b6f1c 00000282 > c105f2c0 > c0eafd9c c014f00e c10b6f1c 00000080 c0eafdd0 c0eafe9c 00000000 > 00010740 > 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b69c8 edi: 0002df48 ebp: c0e8dd40 esp: c0e8dd20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 146, stackpage=c0e8d000) > Stack: 04000000 00010740 c10b6a4c c0e8dd9c 00000080 c10b69cc 00000286 > c105f2c0 > c0e8dd9c c014f00e c10b69cc 00000080 c0e8ddd0 c0e8de9c 00000000 > 00010740 > 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b6f18 edi: 0002df48 ebp: c0e8dd40 esp: c0e8dd20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 150, stackpage=c0e8d000) > Stack: 04000000 00010740 c10b6f9c c0e8dd9c 00000080 c10b6f1c 00000282 > c105f2c0 > c0e8dd9c c014f00e c10b6f1c 00000080 c0e8ddd0 c0e8de9c 00000000 > 00010740 > 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b69c8 edi: 0002df48 ebp: c1e83d40 esp: c1e83d20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 170, stackpage=c1e83000) > Stack: 04000000 00010740 c10b6a4c c1e83d9c 00000080 c10b69cc 00000286 > c105f2c0 > c1e83d9c c014f00e c10b69cc 00000080 c1e83dd0 c1e83e9c 00000000 > 00010740 > 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b6f18 edi: 0002df48 ebp: c1727d40 esp: c1727d20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 213, stackpage=c1727000) > Stack: 04000000 000107a0 c10b6f9c c1727d9c 00000080 c10b6f1c 00000282 > c105f2c0 > c1727d9c c014f00e c10b6f1c 00000080 c1727dd0 c1727e9c 00000000 > 000107a0 > 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b69c8 edi: 0002df48 ebp: c1727d40 esp: c1727d20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 840, stackpage=c1727000) > Stack: 04000000 000107a0 c10b6a4c c1727d9c 00000080 c10b69cc 00000286 > c105f2c0 > c1727d9c c014f00e c10b69cc 00000080 c1727dd0 c1727e9c 00000000 > 000107a0 > 00000004 c10b69cc 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b6f18 edi: 0002df48 ebp: c19d9d40 esp: c19d9d20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 845, stackpage=c19d9000) > Stack: 04000000 000107a0 c10b6f9c c19d9d9c 00000080 c10b6f1c 00000282 > c105f2c0 > c19d9d9c c014f00e c10b6f1c 00000080 c19d9dd0 c19d9e9c 00000000 > 000107a0 > 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > kernel BUG at slab.c:1437! > invalid operand: 0000 > CPU: 0 > EIP: 0010:[] Not tainted > EFLAGS: 00010016 > eax: 5a2cf071 ebx: c10b6020 ecx: c105f2c0 edx: 6444c888 > esi: c10b6f18 edi: 0002df48 ebp: c1a6fd40 esp: c1a6fd20 > ds: 0018 es: 0018 ss: 0018 > Process kexec (pid: 868, stackpage=c1a6f000) > Stack: 04000000 000107a0 c10b6f9c c1a6fd9c 00000080 c10b6f1c 00000282 > c105f2c0 > c1a6fd9c c014f00e c10b6f1c 00000080 c1a6fdd0 c1a6fe9c 00000000 > 000107a0 > 00000004 c10b6f1c 464c457f 00010101 00000000 00000000 00030002 > 00000001 > Call Trace: [] [] [] [] > [] > [] [] [] [] [] > > Code: 0f 0b 9d 05 80 89 22 c0 8b 7d fc b8 71 f0 2c 5a 89 f2 03 57 > > root at who4:/usr/src/LoBos/work# exit > Script done on Sun Apr 13 02:36:28 2003 > > As you can see, that's what caused the message. This time I tripped > the kernel exec on a remote console, and watched the main one. The > main one displayed the oops, and the remote console gave a > segmentation fault as its error. So? Anybody out there, recognize the > fault? I personally think it has to do with my choice of kernel exec > command line tools. Also, versions 2.1, and 2.2. and 2.3 of the make > elf Image too, didn't work here. I'll grab that as a script file for > the latest release, and present it later. > > Andrew did you indeed back port the 2.5 series of kernel exec patch > functions to the 2..4.19, and 2.4.20, as I have now decided? > ------------------- > P.S. If that's the case, then don't worry about it any. It is my fault > for choosing the wrong tool to use first. > --------------------- > Gregg C Levine hansolofalcon at worldnet.att.net > ------------------------------------------------------------ > "The Force will be with you...Always." Obi-Wan Kenobi > "Use the Force, Luke."? Obi-Wan Kenobi > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > (This company dedicates this E-Mail to Master Yoda ) > > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From aip at cwlinux.com Sun Apr 13 02:33:00 2003 From: aip at cwlinux.com (Andrew Ip) Date: Sun Apr 13 02:33:00 2003 Subject: Kexec and kernel oops In-Reply-To: <000001c30189$42487ce0$24c4580c@who5>; from Gregg C Levine on Sun, Apr 13, 2003 at 02:52:31AM -0400 References: <000001c30189$42487ce0$24c4580c@who5> Message-ID: <20030413150402.A8980@mail.cwlinux.com> Greg, > As you can see, that's what caused the message. This time I tripped > the kernel exec on a remote console, and watched the main one. The > main one displayed the oops, and the remote console gave a > segmentation fault as its error. So? Anybody out there, recognize the > fault? I personally think it has to do with my choice of kernel exec > command line tools. Also, versions 2.1, and 2.2. and 2.3 of the make > elf Image too, didn't work here. I'll grab that as a script file for > the latest release, and present it later. > Andrew did you indeed back port the 2.5 series of kernel exec patch > functions to the 2..4.19, and 2.4.20, as I have now decided? Can you try 2.4.19? It did work when I boot off from DOC and then kexec 2.4.19 kernel from IDE hd. For 2.4.20, I will have to double check. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From hansolofalcon at worldnet.att.net Sun Apr 13 11:24:00 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sun Apr 13 11:24:00 2003 Subject: Kernel exec functions and new problems (long) Message-ID: <000201c301d5$170d1120$cac7580c@who5> Hello again from Gregg C Levine Well, switching to the latest of Eric's kernel exec tool, has made a difference. Instead of getting those annoying "oops" messages, I am now getting something far more meaningful. Although it means nothing to me, (or it should and I don't know what to look for), it does indicate, that your assertions Andrew are leaning towards being correct. Here's the message: Script started on Sun Apr 13 11:36:57 2003 root at who4:/usr/src/LoBos/work# ./kexec kernel.elf kexec_load failed: Function not implemented entry = (nil) nr_segments = 4 segment[0].buf = 0x80a8ee8 segment[0].bufsz = 158 segment[0].mem = (nil) segment[0].memsz = 158 segment[1].buf = 0x80a3888 segment[1].bufsz = 55e0 segment[1].mem = 0x10000 segment[1].memsz = 55e0 segment[2].buf = 0x80a8e70 segment[2].bufsz = 0 segment[2].mem = 0x91000 segment[2].memsz = 0 segment[3].buf = 0x40001008 segment[3].bufsz = b48dc segment[3].mem = 0x100000 segment[3].memsz = 700000 root at who4:/usr/src/LoBos/work# exit Script done on Sun Apr 13 11:37:15 2003 At a guess, I'd say, it is telling me that the function was not correctly implemented on this kernel, which is 2.4.20. So, this does backup that notion that I should grab a copy of the raw 2.4.19 sources. I'll do that later today, and try again. However while building the version 1.8 tool, I received a number of warnings, nothing severe just simple warnings. I'll post them after the signature block. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) Script started on Sun Apr 13 11:38:51 2003 root at who4:/usr/src/LoBos/kexec-tools-1.8# make mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/kexec.o -c kexec/kexec.c mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/kexec-syscall.o -c kexec/kexec-syscall.c mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/kexec-x86.o -c kexec/kexec-x86.c mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/kexec-elf32-x86.o -c kexec/kexec-elf32-x86.c mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/kexec-bzImage.o -c kexec/kexec-bzImage.c kexec/kexec-bzImage.c: In function `bzImage_load': kexec/kexec-bzImage.c:163: warning: `fp_ramdisk' might be used uninitialized in this function mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/ifdown.o -c kexec/ifdown.c mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/x86-setup-32.o -c kexec/x86-setup-32.S mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/x86-setup-16.o -c kexec/x86-setup-16.S mkdir -p objdir/kexec gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -o objdir/kexec/x86-setup-16-debug.o -c kexec/x86-setup-16-debug.S /tmp/cc12pRzT.s: Assembler messages: /tmp/cc12pRzT.s:254: Warning: indirect jmp without `*' mkdir -p objdir/util_lib gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -c -o objdir/util_lib/compute_ip_checksum.o util_lib/compute_ip_checksum.c mkdir -p objdir ar rs ./objdir/libutil.a ./objdir/util_lib/compute_ip_checksum.o mkdir -p objdir/build/sbin gcc -Wall -g -O2 -fno-strict-aliasing -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -static -o objdir/build/sbin/kexec ./objdir/kexec/kexec.o ./objdir/kexec/kexec-syscall.o ./objdir/kexec/kexec-x86.o ./objdir/kexec/kexec-elf32-x86.o ./objdir/kexec/kexec-bzImage.o ./objdir/kexec/ifdown.o ./objdir/kexec/x86-setup-32.o ./objdir/kexec/x86-setup-16.o ./objdir/kexec/x86-setup-16-debug.o ./objdir/libutil.a mkdir -p objdir/kexec_test gcc -x assembler-with-cpp -E -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -DRELOC=0x10000 kexec_test/kexec_test16.S > objdir/kexec_test/kexec_test16.s mkdir -p objdir/kexec_test as -o objdir/kexec_test/kexec_test16.o objdir/kexec_test/kexec_test16.s mkdir -p objdir/kexec_test gcc -x assembler-with-cpp -E -I./util_lib/include -DVERSION='"1.8"' -DRELEASE_DATE='"01 December 2002"' -DRELOC=0x10000 kexec_test/kexec_test.S > objdir/kexec_test/kexec_test.s mkdir -p objdir/kexec_test as -o objdir/kexec_test/kexec_test.o objdir/kexec_test/kexec_test.s mkdir -p objdir/build/bin ld -e _start -Ttext 0x10000 ./objdir/kexec_test/kexec_test16.o ./objdir/kexec_test/kexec_test.o -o objdir/build/bin/kexec_test root at who4:/usr/src/LoBos/kexec-tools-1.8# exit exit Script done on Sun Apr 13 11:39:39 2003 And that's what it said. Now where do I install everything? From helpijp at yahoo.co.in Sun Apr 13 23:28:00 2003 From: helpijp at yahoo.co.in (=?iso-8859-1?q?J=20P?=) Date: Sun Apr 13 23:28:00 2003 Subject: Geode Message-ID: <20030414035848.95060.qmail@web8201.mail.in.yahoo.com> Hi, I am using geode sc1200 chip. I have to do some initialization to my sc1200 and then load my linux OS from hard disk to SDRAM. Can linux bios satisfy all these? What else can linux bios do? Catch all the cricket action. Download Yahoo! Score tracker -------------- next part -------------- An HTML attachment was scrubbed... URL: From hansolofalcon at worldnet.att.net Sun Apr 13 23:34:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sun Apr 13 23:34:01 2003 Subject: Geode In-Reply-To: <20030414035848.95060.qmail@web8201.mail.in.yahoo.com> Message-ID: <000b01c3023b$2d39e200$d1c3580c@who5> Hello from Gregg C Levine As far as I know, about that family of processors, and associated systems, a lot of things. But please be more specific. We can not help you, if you do not provide sufficient information. I believe this list has searchable archives as well. Before you submit your questions, please try searching them, you should find a clear enough answer for that issue. I do know that one of the main sticking points, is still IDE issues. Also please try not to use HTML, it wastes network bandwidth, among other issues. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org] On Behalf Of J P Sent: Sunday, April 13, 2003 11:59 PM To: linuxbios at clustermatic.org Subject: Geode Hi, ???? I am using geode sc1200 chip. I have to do some initialization to my sc1200 and then load my linux OS from hard disk to SDRAM. Can linux bios satisfy all these? What else can linux bios do? Catch all the cricket action. Download Yahoo! Score tracker From stuge-linuxbios at cdy.org Mon Apr 14 07:33:00 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Mon Apr 14 07:33:00 2003 Subject: Geode In-Reply-To: <20030414035848.95060.qmail@web8201.mail.in.yahoo.com> References: <20030414035848.95060.qmail@web8201.mail.in.yahoo.com> Message-ID: <20030414115502.GA7819@foo.birdnet.se> On Mon, Apr 14, 2003 at 04:58:48AM +0100, J P wrote: > Hi, I am using geode sc1200 chip. I have to do some initialization to > my sc1200 and then load my linux OS from hard disk to SDRAM. Can linux > bios satisfy all these? What else can linux bios do? Checkout the current source from cvs and look at the code for the nano system written by Christer Weinigel. The nano is based on the SC2200 but almost all of the code applies to SC1200 as well. Using etherboot or some patches to LinuxBIOS you'll be able to load an OS from an IDE drive as well. //Peter From rminnich at lanl.gov Mon Apr 14 09:52:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 09:52:00 2003 Subject: Geode In-Reply-To: <20030414035848.95060.qmail@web8201.mail.in.yahoo.com> Message-ID: On Mon, 14 Apr 2003, J P wrote: > Hi, I am using geode sc1200 chip. I have to do some initialization to my > sc1200 and then load my linux OS from hard disk to SDRAM. Can linux bios > satisfy all these? What else can linux bios do? Catch all the cricket > action. Download Yahoo! Score tracker yes, it can do all these things. ron From wdinkel at atipa.com Mon Apr 14 12:14:01 2003 From: wdinkel at atipa.com (Will Dinkel) Date: Mon Apr 14 12:14:01 2003 Subject: flash_rom and tiger-i7501 Message-ID: <1050337956.12148.90.camel@zappa> flash_rom experts: Has anyone managed to successfully flash a tiger-i7501 with flash_rom? On all of my attempts it can't manage to find an EEPROM: ---------------- Calibrating timer since microsleep sucks ... takes a second Setting up microsecond timing loop 485M loops per second OK, calibrated, now do the deed Enabling flash write on E7500...OK Trying Am29F040B, 512 KB probe_29f040b: id1 0x1f, id2 0x8b Trying At29C040A, 512 KB probe_jedec: id1 0xbf, id2 0x60 Trying Mx29f002, 256 KB probe_29f002: id1 191, id2 96 Trying SST29EE020A, 256 KB probe_jedec: id1 0xbf, id2 0x60 Trying SST28SF040A, 512 KB probe_28sf040: id1 0x1f, id2 0x8b Trying SST39SF020A, 256 KB probe_39sf020: id1 0xbf, id2 0x60 Trying SST39VF020, 256 KB probe_39sf020: id1 0xbf, id2 0x60 Trying W29C011, 128 KB probe_jedec: id1 0xbf, id2 0x60 Trying W29C020C, 256 KB probe_jedec: id1 0xbf, id2 0x60 Trying W49F002U, 256 KB probe_49f002: id1 0xbf, id2 0x60 Trying M29F400BT, 512 KB probe_m29f400bt: id1 0x1f, id2 0x8 Trying 82802ab, 512 KB probe_82802ab: id1 0x1f, id2 0x8b Trying 82802ac, 1024 KB probe_82802ab: id1 0x1f, id2 0x8b EEPROM not found ----------------- -- Will Dinkel Atipa Technologies wdinkel at atipa.com (785) 841-9513 x3111 http://www.atipa.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rminnich at lanl.gov Mon Apr 14 13:08:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 13:08:01 2003 Subject: flash_rom and tiger-i7501 In-Reply-To: <1050337956.12148.90.camel@zappa> Message-ID: On 14 Apr 2003, Will Dinkel wrote: > flash_rom experts: > > Has anyone managed to successfully flash a tiger-i7501 with flash_rom? > On all of my attempts it can't manage to find an EEPROM: This is a flash enable issue, almost certainly. What we need is tyan to tell us how to enable flash writes, and there is a tyan person on this list. HELP! ron From ebiederman at lnxi.com Mon Apr 14 13:09:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 13:09:00 2003 Subject: Dear sir In-Reply-To: <20030414154439.43476.qmail@web41008.mail.yahoo.com> References: <20030414154439.43476.qmail@web41008.mail.yahoo.com> Message-ID: gururaj thirtha writes: > Dear Mr.Eric Biederman > > > How can I proceed to implement LinuxBIOS ? > > pls send me the details as known by you. > > Is new kernel building is necessary for that project ? I am bouncing your question to the LinuxBIOS mailing list. Eric From dpilon at atipa.com Mon Apr 14 16:06:01 2003 From: dpilon at atipa.com (Denis Pilon) Date: Mon Apr 14 16:06:01 2003 Subject: supermicro ... etherboot Message-ID: <1050352671.1414.18.camel@atipa-dp> Got linuxbios coming up on a x5dpe-g2, but it will not grab an IP. It just sits there looking for dhcp server. Looking at the dhcp logs, the etherboot never sends the DHCPACK. What do I do ? FYI: using etherboot 5.1.7 Denis -- =================================================== Denis E. Pilon Systems Integration Specialist Atipa Technologies http://www.atipa.com/ dpilon at atipa.com (785)841-9513 x3111 =================================================== -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From ebiederman at lnxi.com Mon Apr 14 16:15:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 16:15:01 2003 Subject: [commit] tiger-i7501 In-Reply-To: References: Message-ID: ron minnich writes: > what is the part # of the i7501? > > this is great to see 7501 commits coming in! Now if only the code worked, in a general fashion. Eric From ebiederman at lnxi.com Mon Apr 14 16:23:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 16:23:01 2003 Subject: supermicro ... etherboot In-Reply-To: <1050352671.1414.18.camel@atipa-dp> References: <1050352671.1414.18.camel@atipa-dp> Message-ID: Denis Pilon writes: > Got linuxbios coming up on a x5dpe-g2, but it will not grab an IP. It > just sits there looking for dhcp server. > > Looking at the dhcp logs, the etherboot never sends the DHCPACK. > > What do I do ? > > FYI: > using etherboot 5.1.7 Either it is a linuxbios issue in which case the x5dpe-g2 is not a completed port, so you should debug the port. Or this is an etherboot issue in which case this is the wrong list. Eric From rminnich at lanl.gov Mon Apr 14 18:02:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 18:02:00 2003 Subject: supermicro ... etherboot In-Reply-To: <1050352671.1414.18.camel@atipa-dp> Message-ID: On 14 Apr 2003, Denis Pilon wrote: > Got linuxbios coming up on a x5dpe-g2, but it will not grab an IP. It > just sits there looking for dhcp server. > > Looking at the dhcp logs, the etherboot never sends the DHCPACK. > > What do I do ? run tcpdump. Follow the packet traffic that way. ron From rminnich at lanl.gov Mon Apr 14 18:26:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 18:26:01 2003 Subject: Power PC support added Message-ID: I just did a commit for PPC. It all looks ok but as always there may be a problem or two. Let me know. Also welcome Greg Watson (gwatson at lanl.gov) as a new committer, managing the PPC. ron From ebiederman at lnxi.com Mon Apr 14 19:17:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 19:17:00 2003 Subject: [COMMIT] sync up pass 3 Message-ID: First stab at synchronizing the questionable code. The NLBConfig.py and getpir changes are safe simple changes. The read_bytes.h and rom_fill_inbuf.c removes tag support from LinuxBIOS altogether. The code is in the baremetal toolkit so we should not need it bloating the common code. Eric include/rom/read_bytes.h - Remove baremetal toolkit code src/rom/rom_fill_inbuf.c - Remove tag support util/config/NLBConfig.py - Remove crt0.o from OBJECTS-1 util/getpir/Makefile - set LINUXBSIOSROOT to an always working path From klos1980 at sina.com Mon Apr 14 20:24:01 2003 From: klos1980 at sina.com (klos) Date: Mon Apr 14 20:24:01 2003 Subject: flash_rom and tiger-i7501 In-Reply-To: <1050337956.12148.90.camel@zappa> Message-ID: <000401c302e9$8f24b950$7ca8a8c0@sh.corp.tyan.com> You may try sst49LF004A -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org] On Behalf Of Will Dinkel Sent: 2003?4?15? 0:33 To: LinusBIOS mailing list Subject: flash_rom and tiger-i7501 flash_rom experts: Has anyone managed to successfully flash a tiger-i7501 with flash_rom? On all of my attempts it can't manage to find an EEPROM: ---------------- Calibrating timer since microsleep sucks ... takes a second Setting up microsecond timing loop 485M loops per second OK, calibrated, now do the deed Enabling flash write on E7500...OK Trying Am29F040B, 512 KB probe_29f040b: id1 0x1f, id2 0x8b Trying At29C040A, 512 KB probe_jedec: id1 0xbf, id2 0x60 Trying Mx29f002, 256 KB probe_29f002: id1 191, id2 96 Trying SST29EE020A, 256 KB probe_jedec: id1 0xbf, id2 0x60 Trying SST28SF040A, 512 KB probe_28sf040: id1 0x1f, id2 0x8b Trying SST39SF020A, 256 KB probe_39sf020: id1 0xbf, id2 0x60 Trying SST39VF020, 256 KB probe_39sf020: id1 0xbf, id2 0x60 Trying W29C011, 128 KB probe_jedec: id1 0xbf, id2 0x60 Trying W29C020C, 256 KB probe_jedec: id1 0xbf, id2 0x60 Trying W49F002U, 256 KB probe_49f002: id1 0xbf, id2 0x60 Trying M29F400BT, 512 KB probe_m29f400bt: id1 0x1f, id2 0x8 Trying 82802ab, 512 KB probe_82802ab: id1 0x1f, id2 0x8b Trying 82802ac, 1024 KB probe_82802ab: id1 0x1f, id2 0x8b EEPROM not found ----------------- -- Will Dinkel Atipa Technologies wdinkel at atipa.com (785) 841-9513 x3111 http://www.atipa.com From ebiederman at lnxi.com Mon Apr 14 20:43:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 20:43:01 2003 Subject: Still not having success with SM P4DPE In-Reply-To: References: Message-ID: steven james writes: > Greetings, > > Possibly both :-) > First, try: > strip eepro100.elf > objcopy -R .note -R .comment eepro100.elf eepro100.strip Actually building eepro100.zelf should be more helpful. On that side of the fence. And at least with the 5.1.7+ I don't think the strip helps at all. Eric From ebiederman at lnxi.com Mon Apr 14 20:47:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 20:47:00 2003 Subject: Etherboot ... In-Reply-To: <1050096575.1531.317.camel@atipa-dp> References: <1050096575.1531.317.camel@atipa-dp> Message-ID: Denis Pilon writes: > OK. > > Got an image built using etherboot-5.0.8 I think it works, but the > e1000 driver is to old so it can't network boot. > > Using the etherboot-5.1.3, the elf image is to big. 5.1.3 is a little dated but... > How do I shrink the romimage to fit 524288 ? I just built 5.1.7 with all of the pci drivers compiled in and it only came down to: ls -l tdir/linuxbios/etherboot-pci.zelf -rw-r--r-- 1 eric eric 76285 Apr 14 19:19 tdir/linuxbios/etherboot-pci.zelf So I fail to see your problem. Eric From rminnich at lanl.gov Mon Apr 14 22:28:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 22:28:01 2003 Subject: [COMMIT] sync up pass 3 In-Reply-To: Message-ID: On 14 Apr 2003, Eric W. Biederman wrote: > - Remove crt0.o from OBJECTS-1 interesting, I think Greg did this too for PPC. ron From ebiederman at lnxi.com Mon Apr 14 22:37:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 14 22:37:01 2003 Subject: [COMMIT] sync up pass 3 In-Reply-To: References: Message-ID: ron minnich writes: > On 14 Apr 2003, Eric W. Biederman wrote: > > > - Remove crt0.o from OBJECTS-1 > > interesting, I think Greg did this too for PPC. I did this quite a while ago in my tree when we were having problems with crt0.o being linked against things, so I would not forget to get it into the main tree. And I finally found the change again and committed it :) Eric From rminnich at lanl.gov Mon Apr 14 22:38:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 14 22:38:00 2003 Subject: Development fork on the way Message-ID: This is a warning, but no action has yet occured. It is time to lock down LinuxBIOS in its current state and build a development fork. What's going to happen is a feature freeze. We want to freeze the core and make sure it works for everyone. What that will mean is that once we freeze, almost nothing but files in src/mainboard will change, and we want to keep that to a minimum. One thing I need from everyone "owning" a mainboard: I would like you to email me with the names of mainboards you own. I plan to fill out the status files with that info. If a mainboard has no "owner" it will not appear in the development tree, since we have no way to test it. This means systems like the IBM T23 will disappear from the development tree. So please start thinking now about what mainboards you can take responsibility for. We need all the help we can get. thanks ron From shubhangi.jadhav at patni.com Tue Apr 15 04:39:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 15 04:39:00 2003 Subject: Is cmos.layout generated? Message-ID: Hi, Is the cmos.layout file a standard file or is it generated? which is the utility that generates it? Thanks, Shubhangi From shubhangi.jadhav at patni.com Tue Apr 15 05:01:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 15 05:01:01 2003 Subject: Significance of cpu p5/p6 in config file Message-ID: Hi, Could someone tell me the significance of cpu p5 cpu p6 I've found these in most of the mainboard config files. I'm trying to port LinuxBIOS to a motherboard with via C3 processor. Do I need to include the above statements in my config file. Thanks, Shubhangi From shubhangi.jadhav at patni.com Tue Apr 15 05:57:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 15 05:57:01 2003 Subject: Intel 82810E GMCH northbridge implementation Message-ID: Hi, To port LinuxBIOS to our motherboard, we need to implement support for the northbridge. The northbridge is Intel 82810E GMCH. May I know how easy/difficult it would be to implement the northbridge? Any suggestions on how to go about implementing this ? Thanks, Shubhangi From rezso at rdsor.ro Tue Apr 15 06:37:00 2003 From: rezso at rdsor.ro (Balint Cristian) Date: Tue Apr 15 06:37:00 2003 Subject: newbie - soyo SYK7ADA ? Message-ID: <200304151409.30244.rezso@rdsor.ro> Is there chance to work on soyo SYK7ADA ? Here is full pcilist: 00:00.0 Host bridge: Acer Laboratories Inc. [ALi] M1647 Northbridge [MAGiK 1 / MobileMAGiK 1] (rev 02) 00:01.0 PCI bridge: Acer Laboratories Inc. [ALi] PCI to AGP Controller 00:02.0 USB Controller: Acer Laboratories Inc. [ALi] USB 1.1 Controller (rev 03) 00:03.0 Multimedia audio controller: Acer Laboratories Inc. [ALi] M5451 PCI AC-Link Controller Audio Device (rev 02) 00:04.0 IDE interface: Acer Laboratories Inc. [ALi] M5229 IDE (rev c4) 00:06.0 USB Controller: Acer Laboratories Inc. [ALi] USB 1.1 Controller (rev 03) 00:07.0 ISA bridge: Acer Laboratories Inc. [ALi] M1533 PCI to ISA Bridge [Aladdin IV] 00:0b.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 00:11.0 Bridge: Acer Laboratories Inc. [ALi] M7101 PMU 01:00.0 VGA compatible controller: NVidia / SGS Thomson (Joint Venture) Riva128 (rev 10) Maybe can i recive some hints how to test/start amd makeup an linuxbios for this MB ? Thank You, cristian -- Unable to handle kernel paging request \|/ ____ \|/ "@'/ .. \`@" /_| \__/ |_\ \__U_/ setserial(104): Oops TSTATE: 0000004411f09605 TPC: 000000000051ba00 TNPC: 000000000051ba0c Y: 0600000 g0: 7001821070016068 g1: 00000000006bf5d0 g2: 0000000000000000 g3: 00000000006b4300 g4: fffff8001398f7a0 g5: 0000000000000000 g6: fffff80010638000 g7: 0000000000000000 o0: 000008810001ccb4 o1: fffff8001398f7a0 o2: 00000000005f8af4 o3: 00000000000000f8 o4: ffffffffffffffff o5: 0000000000000000 sp: fffff8001063a601 ret_pc: 000000000044a548 l0: 0000000000000043 l1: 0000000000000039 l2: 0000000000000000 l3: 0000000000000000 l4: 0000000000000000 l5: 000000000070c920 l6: 0000000000000000 l7: 0000000000000000 i0: fffff80013d55e50 i1: fffff8001063af9e i2: fffff8001063b0f0 i3: 0000000000000000 i4: 00000000006bf240 i5: ffffffffffffffff i6: fffff8001063a6e1 i7: 000000000051c34c Instruction DUMP: d05e2028 808ae001 02600004 9a102001 e64620b4 808ae080 0240000c 808ae040 From pyro at linuxlabs.com Tue Apr 15 08:59:01 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 15 08:59:01 2003 Subject: flash_rom and tiger-i7501 In-Reply-To: <1050337956.12148.90.camel@zappa> Message-ID: Greetings, I will be posting a flash utility to take care of that shortly. G'day, sjames On 14 Apr 2003, Will Dinkel wrote: > flash_rom experts: > > Has anyone managed to successfully flash a tiger-i7501 with flash_rom? > On all of my attempts it can't manage to find an EEPROM: > > ---------------- > > Calibrating timer since microsleep sucks ... takes a second > Setting up microsecond timing loop > 485M loops per second > OK, calibrated, now do the deed > Enabling flash write on E7500...OK > Trying Am29F040B, 512 KB > probe_29f040b: id1 0x1f, id2 0x8b > Trying At29C040A, 512 KB > probe_jedec: id1 0xbf, id2 0x60 > Trying Mx29f002, 256 KB > probe_29f002: id1 191, id2 96 > Trying SST29EE020A, 256 KB > probe_jedec: id1 0xbf, id2 0x60 > Trying SST28SF040A, 512 KB > probe_28sf040: id1 0x1f, id2 0x8b > Trying SST39SF020A, 256 KB > probe_39sf020: id1 0xbf, id2 0x60 > Trying SST39VF020, 256 KB > probe_39sf020: id1 0xbf, id2 0x60 > Trying W29C011, 128 KB > probe_jedec: id1 0xbf, id2 0x60 > Trying W29C020C, 256 KB > probe_jedec: id1 0xbf, id2 0x60 > Trying W49F002U, 256 KB > probe_49f002: id1 0xbf, id2 0x60 > Trying M29F400BT, 512 KB > probe_m29f400bt: id1 0x1f, id2 0x8 > Trying 82802ab, 512 KB > probe_82802ab: id1 0x1f, id2 0x8b > Trying 82802ac, 1024 KB > probe_82802ab: id1 0x1f, id2 0x8b > EEPROM not found > > ----------------- > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From pyro at linuxlabs.com Tue Apr 15 09:17:01 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 15 09:17:01 2003 Subject: [commit] tiger-i7501 In-Reply-To: Message-ID: Greetings, Which cases are you seeing failures in? I have a set of 533MHz PIVs on order, but not yet in hand. G'day, sjames On 14 Apr 2003, Eric W. Biederman wrote: > ron minnich writes: > > > what is the part # of the i7501? > > > > this is great to see 7501 commits coming in! > > Now if only the code worked, in a general fashion. > > Eric > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Tue Apr 15 09:19:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 15 09:19:00 2003 Subject: Significance of cpu p5/p6 in config file In-Reply-To: Message-ID: On Tue, 15 Apr 2003, Shubhangi Jadhav wrote: > cpu p5 > cpu p6 > I've found these in most of the mainboard config files. It depends on how compatible the C3 is with the p5 and p6. Start out with just P5, see if it works, add p6, see if it still works. Thats's the simplest test. ron From rminnich at lanl.gov Tue Apr 15 09:20:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 15 09:20:01 2003 Subject: Intel 82810E GMCH northbridge implementation In-Reply-To: Message-ID: On Tue, 15 Apr 2003, Shubhangi Jadhav wrote: > To port LinuxBIOS to our motherboard, we need to implement support > for the northbridge. The northbridge is Intel 82810E GMCH. look for the most similar intel northbridge as a basis and take it from there. This will require some research on your part at developer.intel.com. ron From rminnich at lanl.gov Tue Apr 15 09:21:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 15 09:21:00 2003 Subject: newbie - soyo SYK7ADA ? In-Reply-To: <200304151409.30244.rezso@rdsor.ro> Message-ID: On Tue, 15 Apr 2003, Balint Cristian wrote: > Is there chance to work on soyo SYK7ADA ? > > Here is full pcilist: > > 00:00.0 Host bridge: Acer Laboratories Inc. [ALi] M1647 Northbridge [MAGiK 1 / MobileMAGiK 1] (rev 02) you're going to need docs on that northbridge. Now, Acer has in the past been very good to deal with, and their parts are very well designed. You can contact them and see if they will give you the info that would allow you to write a linuxbios port. ron From ebiederman at lnxi.com Tue Apr 15 10:10:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 10:10:01 2003 Subject: Is cmos.layout generated? In-Reply-To: References: Message-ID: "Shubhangi Jadhav" writes: > Hi, > > Is the cmos.layout file a standard file or is it generated? which is the > > utility that generates it? cmos.layout is not autogenerated. Instead it describes the cmos variables that a port of LinuxBIOS uses. Eric From ebiederman at lnxi.com Tue Apr 15 10:33:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 10:33:00 2003 Subject: Releases... Message-ID: Over the last several months I have been pushing Ron to start making stable releases of the LinuxBIOS tree. I have contributed the code for Version numbers, and I have been stable releases for boards I support from my own tree. Watching the LinuxBIOS tree one of the things I have seen is that change happens, and it breaks old ports. And as a developer I want more change so new ports are easier to do and simpler to maintain. The path forward as I see it is to freeze the core of the tree, and to have a stable 1.0 release series. With the goal being that from release to release, we do not break any existing port in the tree. To achieve this we need to freeze the core of the tree. And put any chipset code into at least a slush so fixing one board does not break an old board. Once all of the working boards are available in the stable tree, there is not a need to avoid breaking things in the development branch. A working even if someday antiquated version will always be available just be fetching an old stable release. So that is my plan for breaking everything without breaking anything:) Eric From dkotian3 at vsnl.net Tue Apr 15 10:47:00 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Tue Apr 15 10:47:00 2003 Subject: Are the additional experiences or porting help for LinuxBIOS Message-ID: <20030415155201.CAFA74FE06@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From ebiederman at lnxi.com Tue Apr 15 11:08:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 11:08:00 2003 Subject: 1.1 development release targets. Message-ID: Development in the current LinuxBIOS tree is starting to get difficult because of the sheer number of backward compatible interfaces we retain even when better code is introduced into the tree. For the 1.1.x development branch leading to the stable 2.0.x series it is my intention remove all of the old interfaces and put LinuxBIOS back together using the current best of breed techniques. - Use the elf bootloader - Use romcc so we can remove most of the assembly. - Use the generic pci setup mechanism. It needs a little work to make initialization as easy as the superio case but it should work for everything. - Make hardwaremain architecture independent. - Native interrupt routing. (We don't have code but...) - Documented interfaces (I suck at doing this but it needs to happen)... I am still in the final stages of prototyping the cleanups to hardwaremain. But my first draft is below. The function that still needs things worked out is enumerate static devices. One or two boards need to be ported and the issues worked out before everything else comes along. The goal is a hardwaremain() function that does not change. Once there is a stable port in the 1.1. development branch my goal is to call that enough development make it the 2.0 branch and any future disruptive work needs to go on a new development branch. When the 2.0 branch can continue to accumulate working ports. To implement this I have started a freebios2 tree at sourceforge and I will gradually populating this tree. I am taking the expedient of copying old code over as needed. Leaving old code for now unneeded functionality to just fade away. It is the only easy incremental way I can see to make the sweeping changes I want to see in the LinuxBIOS tree. I am going to start with the Hammer port and possibly with the p4dpr, to seed the tree and work out most of the issues with the new infrastructure. After that it is moving to 2.0 and collecting additional ports, I suspect. Eric void hardwaremain(int boot_complete) { /* Processor ID of the BOOT cpu (i.e. the one running this code) */ unsigned long boot_cpu; int boot_index; /* the order here is a bit tricky. We don't want to do much of * anything that uses config registers until after PciAllocateResources * since that function also figures out what kind of config strategy * to use (type 1 or type 2). * so we turn on cache, then worry about PCI setup, then do other * things, so that the other work can use the PciRead* and PciWrite* * functions. */ struct mem_range *mem, *tmem; unsigned long totalmem; // we don't call post code for this one -- since serial post could cause real // trouble. outb(0x38, 0x80); /* displayinit MUST PRECEDE ALL PRINTK! */ displayinit(); post_code(0x39); printk_notice("LinuxBIOS-%s%s %s %s...\n", linuxbios_version, linuxbios_extra_version, linuxbios_build, (boot_complete)?"rebooting":"booting"); post_code(0x40); /* If we have already booted attempt a hard reboot */ if (boot_complete) { hard_reset(); } // pick how to scan the bus. This is first so we can get at memory size. printk_info("Finding PCI configuration type.\n"); pci_set_method(); post_code(0x5f); enumerate_static_devices(); pci_enumerate(); post_code(0x66); // Now do the real bus // we round the total ram up a lot for thing like the SISFB, which // shares high memory with the CPU. pci_configure(); post_code(0x88); pci_enable(); pci_initialize(); post_code(0x89); mem = get_ramsize(); post_code(0x70); totalmem = 0; for(tmem = mem; tmem->sizek; tmem++) { totalmem += tmem->sizek; } printk_info("totalram: %ldM\n", (totalmem + 512) >> 10); /* Round to the nearest meg */ /* Fully initialize the cpu before configuring the bus */ boot_cpu = cpu_initialize(mem); boot_index = processor_index(boot_cpu); printk_spew("BOOT CPU is %d\n", boot_cpu); processor_map[boot_index] = CPU_BOOTPROCESSOR|CPU_ENABLED; /* Now start the other cpus initializing * The sooner they start the sooner they stop. */ post_code(0x75); startup_other_cpus(processor_map); post_code(0x77); /* make certain we are the only cpu running in linuxBIOS */ wait_for_other_cpus(); /* Now that we have collected all of our information * write our configuration tables. */ write_tables(mem); elfboot(streams, get_lb_mem()); } From ebiederman at lnxi.com Tue Apr 15 11:09:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 11:09:00 2003 Subject: [commit] tiger-i7501 In-Reply-To: References: Message-ID: steven james writes: > Greetings, > > Which cases are you seeing failures in? I have a set of 533MHz PIVs on > order, but not yet in hand. Currently my coworker Tom Zimmerman is doing the work so I am not up on all of the details. But there are several undocumented registers that when set to the values the previous BIOS used allow the board to work. Otherwise the board fails. Since it is not know how to compute these magic values I consider the board broken. Because changing ram sticks could change what they need to be. In addition though less important the code should really be structured to cope with multiple memory speeds and possibly the single/dual channel issues. Eric > > On 14 Apr 2003, Eric W. Biederman wrote: > > > ron minnich writes: > > > > > what is the part # of the i7501? > > > > > > this is great to see 7501 commits coming in! > > > > Now if only the code worked, in a general fashion. > > > > Eric > > > > -- > -------------------------steven james, director of research, linux labs > ... ........ ..... .... 230 peachtree st nw ste 2701 > the original linux labs atlanta.ga.us 30303 > -since 1995 http://www.linuxlabs.com > office 404.577.7747 fax 404.577.7743 > ----------------------------------------------------------------------- From rminnich at lanl.gov Tue Apr 15 11:25:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 15 11:25:00 2003 Subject: Are the additional experiences or porting help for LinuxBIOS In-Reply-To: <20030415155201.CAFA74FE06@bom6.vsnl.net.in> Message-ID: On Tue, 15 Apr 2003 dkotian3 at vsnl.net wrote: > Or is there some kind of Porting Guide or Porting Architecture > document for LinuxBIOS not yet, but we accept contributions. ron From dkotian3 at vsnl.net Tue Apr 15 11:32:00 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Tue Apr 15 11:32:00 2003 Subject: Are the additional experiences or porting help for LinuxBIOS Message-ID: <20030415163630.5588C4FFF6@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From jerj at coplanar.net Tue Apr 15 12:00:00 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Tue Apr 15 12:00:00 2003 Subject: 1.1 development release targets. References: Message-ID: <00b001c3036c$f0169930$7c07a8c0@kennet.coplanar.net> ----- Original Message ----- From: "Eric W. Biederman" To: "LinuxBIOS" Sent: Tuesday, April 15, 2003 11:41 AM Subject: 1.1 development release targets. > To implement this I have started a freebios2 tree at sourceforge > and I will gradually populating this tree. I am taking the expedient > of copying old code over as needed. Leaving old code for now > unneeded functionality to just fade away. It is the only easy Have you considered using the branch facility of CVS? Cheers, Jeremy From steve at nexpath.com Tue Apr 15 12:18:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Tue Apr 15 12:18:00 2003 Subject: 1.1 development release targets. In-Reply-To: References: Message-ID: <3E9C39CC.2060808@nexpath.com> Eric W. Biederman wrote: > - Use the elf bootloader What about the alternative floppy/ide routines? There are a lot of people using this, although I think it could be made better by using the bzImage and jumping to startup_32 etc. -Steve From ebiederman at lnxi.com Tue Apr 15 13:25:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 13:25:01 2003 Subject: 1.1 development release targets. In-Reply-To: <3E9C39CC.2060808@nexpath.com> References: <3E9C39CC.2060808@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > > - Use the elf bootloader > > What about the alternative floppy/ide routines? Given that there are several types of rom chips we will need to continue to support several types of loaders, not just loading from a classic x86 rom chip. I think the floppy/ide routines may be a bit of unnecessary but there is no real impact on the rest of the code base. Note: All of these drivers are currently in etherboot and you can compile them into the same etherboot binary. > There are a lot of people using > this, although I think it could be made better by using the bzImage and jumping > to startup_32 etc. Not at all. Jumping straight to startup_32 is a very bad idea. First because most times LinuxBIOS is ported people don't have enough time, the code base needs to be optimized for to make simple fast ports. To achieve this everything that can be reused and is motherboard independent needs to be split off into it's own separate binary. And drawing a line at elfboot and saying no further features beyond this point keeps the code bloat down. Second startup_32 is not an officially supported kernel entry point and so it is subject to change whenever someone sees a better way of making it work. This means for example the code in the tree does not work with the redhat 9 kernel for example. A little bit of wrapper code in mkelfImage that we can change when the kernel changes and not cause people to need to reflash their BIOS is a good thing. Third people have a lot of ideas on how the user interface to a BIOS should look: Openfirmware, EFI, PCBIOS, etherboot, redboot, the baremetal toolkit, etc. And it just does not make sense to support all of these as build options of LinuxBIOS. It works a lot better to have the LinuxBIOS table with information describing the hardware, and a simple general purpose bootloader that can load any of them. And then the Interface just needs to be built. Right now with etherboot we have one very good option that the ELF loader can load. And with ADLO, we have a second option on the way. The bare metal toolkit has demonstrated how you can take cute little features like directly loading the Linux kernel still support them, but have the code loaded with the ELF bootloader. In the 1.0.x release we won't remove code but for the development release I don't see how supporting anything except the ELF bootloader makes sense. The concept has proved itself sound. And it allows us to seriously talk about things like support for the BSDs. Eric From ebiederman at lnxi.com Tue Apr 15 13:31:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 13:31:00 2003 Subject: 1.1 development release targets. In-Reply-To: <00b001c3036c$f0169930$7c07a8c0@kennet.coplanar.net> References: <00b001c3036c$f0169930$7c07a8c0@kennet.coplanar.net> Message-ID: "Jeremy Jackson" writes: > > To implement this I have started a freebios2 tree at sourceforge > > and I will gradually populating this tree. I am taking the expedient > > of copying old code over as needed. Leaving old code for now > > unneeded functionality to just fade away. It is the only easy > > Have you considered using the branch facility of CVS? Thought about it. I don't much care about the CVS history. And where there are several trees it is less then correct a number of times. I am about ready to rip out all of the $Id$ tags from the tree they are a consistent source of noise when you have two different repositories and you are comparing the differences. In addition. By pulling in source modules in on demand I have a real nice way of filtering out old dead code and showing which code works in the new development tree. With my radical changes to hardwaremain everything breaks until it gets fixed up. So I guess I don't see an advantage in using the cvs branch feature, at this point. Eric From dkotian3 at vsnl.net Tue Apr 15 13:41:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 15 13:41:00 2003 Subject: Is cmos.layout generated? References: Message-ID: <000201c3037a$889a1a80$1a2141db@vsnl.net> But is there some kind of command or utility which creates this file(cmos.layout) ? Regards Deepak ----- Original Message ----- From: "Eric W. Biederman" To: "Shubhangi Jadhav" Cc: "LinuxBIOS" Sent: Tuesday, April 15, 2003 8:14 PM Subject: Re: Is cmos.layout generated? > "Shubhangi Jadhav" writes: > > > Hi, > > > > Is the cmos.layout file a standard file or is it generated? which is the > > > > utility that generates it? > > cmos.layout is not autogenerated. Instead it describes the cmos variables > that a port of LinuxBIOS uses. > > Eric > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Tue Apr 15 13:42:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 15 13:42:00 2003 Subject: Is cmos.layout generated? References: Message-ID: <005d01c3037a$527ca620$1a2141db@vsnl.net> But is there some kind of command or utility which creates this file(cmos.layout) ? Regards Deepak ----- Original Message ----- From: "Eric W. Biederman" To: "Shubhangi Jadhav" Cc: "LinuxBIOS" Sent: Tuesday, April 15, 2003 8:14 PM Subject: Re: Is cmos.layout generated? > "Shubhangi Jadhav" writes: > > > Hi, > > > > Is the cmos.layout file a standard file or is it generated? which is the > > > > utility that generates it? > > cmos.layout is not autogenerated. Instead it describes the cmos variables > that a port of LinuxBIOS uses. > > Eric > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dpilon at atipa.com Tue Apr 15 13:48:00 2003 From: dpilon at atipa.com (Denis Pilon) Date: Tue Apr 15 13:48:00 2003 Subject: linuxbios/etherboot ... Message-ID: <1050430796.17053.2.camel@atipa-dp> OK, I'm having problems working with etherboot 5.1.7. Linuxbios is telling me it can't load the elf image. But when I use 5.0.8 it works fine.(Using the ebi) Are there any changes that I need to make in my config file in linuxbios to work with 5.1.7 Denis -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rminnich at lanl.gov Tue Apr 15 13:59:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 15 13:59:00 2003 Subject: 1.1 development release targets. In-Reply-To: Message-ID: On 15 Apr 2003, Eric W. Biederman wrote: > Steve Gehlbach writes: > > What about the alternative floppy/ide routines? > > Given that there are several types of rom chips we will > need to continue to support several types of loaders, not just > loading from a classic x86 rom chip. I think the floppy/ide > routines may be a bit of unnecessary but there is no real impact > on the rest of the code base. I think there is going to be continued use of the ide/floppy stuff without etherboot. I have requests for that all the time. It's going to have to stay. I support only loading elfboot images from linuxbios. It just makes too much sense. ron From dkotian3 at vsnl.net Tue Apr 15 15:28:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 15 15:28:01 2003 Subject: Development fork on the way References: Message-ID: <031c01c302fe$89fe50c0$233e41db@vsnl.net> Hi, Are the more detailed port guide available or experience then the one listed at http://www.linuxbios.org/developer/portguides/index.html Please let me know. Thanks and Regards Deepak From ebiederman at lnxi.com Tue Apr 15 18:09:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 15 18:09:00 2003 Subject: 1.1 development release targets. In-Reply-To: References: Message-ID: ron minnich writes: > On 15 Apr 2003, Eric W. Biederman wrote: > > > Steve Gehlbach writes: > > > What about the alternative floppy/ide routines? > > > > Given that there are several types of rom chips we will > > need to continue to support several types of loaders, not just > > loading from a classic x86 rom chip. I think the floppy/ide > > routines may be a bit of unnecessary but there is no real impact > > on the rest of the code base. > > I think there is going to be continued use of the ide/floppy stuff without > etherboot. I have requests for that all the time. It's going to have to > stay. That is fine. As long as we can keep the restriction on only one compiled into the code. Beyond the infrastructure level I really don't care. > I support only loading elfboot images from linuxbios. It just makes too > much sense. Eric From shubhangi.jadhav at patni.com Wed Apr 16 07:37:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Wed Apr 16 07:37:00 2003 Subject: support for 82801AA southbridge Message-ID: Hi, Does LinuxBIOS support the 82801AA southbridge? The code in southbridge\82801 directory is for the 82801AB chip. However in one of the posts to the mailing list I read that the 82801AA southbridge is supported. Could someone guide me please. Am I missing something. Thanks, Shubhangi From steve at nexpath.com Wed Apr 16 12:23:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 16 12:23:00 2003 Subject: 1.1 development release targets. In-Reply-To: References: Message-ID: <3E9D8C9A.9050606@nexpath.com> ron minnich wrote: > I think there is going to be continued use of the ide/floppy stuff without > etherboot. I have requests for that all the time. It's going to have to > stay. > > I support only loading elfboot images from linuxbios. It just makes too > much sense. > Does linuxbios in the current version support elfboot images from ide/floppy (directly, without etherboot)? I guess if it does I don't see it. One issue I have with elfboot images is the inability to dynamically pass parameters to the zero page (specifically the vga cursor location). I understand the idea of not locking the boot mechanism into the bios, but there are pros and cons. The final step of booting Linux is pretty independent of the other linuxbios code, though, so it is easy to patch in an alternate technique for someone (like me) that has specific requirements. -Steve From rminnich at lanl.gov Wed Apr 16 12:53:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 16 12:53:01 2003 Subject: 1.1 development release targets. In-Reply-To: <3E9D8C9A.9050606@nexpath.com> Message-ID: On Wed, 16 Apr 2003, Steve Gehlbach wrote: > One issue I have with elfboot images is the inability to dynamically > pass parameters to the zero page (specifically the vga cursor location). > I understand the idea of not locking the boot mechanism into the bios, > but there are pros and cons. that is a good point. I think it can be resolved via the linuxbios tables that Eric has been pushing, and done in a way that will work with other OSes (e.g. freebsd). ron From tulonja at koti.soon.fi Wed Apr 16 13:53:00 2003 From: tulonja at koti.soon.fi (Jarmo Tulonen) Date: Wed Apr 16 13:53:00 2003 Subject: Via VT82C686B southbridge Message-ID: <20030416182452.VMVV639.fep01-app.kolumbus.fi@there> Hello I'm trying to get Linuxbios work on a Wincomm WPC-660 mainboard having VT82C686B southbridge and VT8601A northbridge (both from Via). Basically I've just copied the Via Epia code, replacing the VT8231-southbridge code with the VT82C686-code. So far the only sign of life is flashing of the keyboard lights after reset, so it's not totally dead. The serial port, however, does not work. Has anyone actually used this southbridge code ? Any ideas how to try to get it work ? Jarmo From ebiederman at lnxi.com Wed Apr 16 14:09:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 16 14:09:01 2003 Subject: 1.1 development release targets. In-Reply-To: <3E9D8C9A.9050606@nexpath.com> References: <3E9D8C9A.9050606@nexpath.com> Message-ID: Steve Gehlbach writes: > ron minnich wrote: > > I think there is going to be continued use of the ide/floppy stuff without > > etherboot. I have requests for that all the time. It's going to have to stay. > > I support only loading elfboot images from linuxbios. It just makes too much > > sense. > > > > Does linuxbios in the current version support elfboot images from ide/floppy > (directly, without etherboot)? I guess if it does I don't see it. Yes. The switch on the ``rom loader'' is the same on code, for either the elf loader or the other one. > One issue I have with elfboot images is the inability to dynamically pass > parameters to the zero page (specifically the vga cursor location). I understand > the idea of not locking the boot mechanism into the bios, but there are pros and > cons. As ron mentioned putting enough information in the LinuxBIOS table so the code in mkelfImage can properly position the vga cursor location is straight forward. > The final step of booting Linux is pretty independent of the other linuxbios > code, though, so it is easy to patch in an alternate technique for someone (like > me) that has specific requirements. Right. But before we give up on a common mechanism I would like to see if we can figure out how to build a general purpose solution. Eric From rminnich at lanl.gov Wed Apr 16 14:48:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 16 14:48:01 2003 Subject: Via VT82C686B southbridge In-Reply-To: <20030416182452.VMVV639.fep01-app.kolumbus.fi@there> Message-ID: the existing code used to work. I believe the vt5426 motherboard is what you want to try. ron From jerj at coplanar.net Wed Apr 16 15:45:00 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Wed Apr 16 15:45:00 2003 Subject: CVS issues [was: 1.1 development release targets.] References: <00b001c3036c$f0169930$7c07a8c0@kennet.coplanar.net> Message-ID: <00ae01c30455$96a78c80$7c07a8c0@kennet.coplanar.net> ----- Original Message ----- From: "Eric W. Biederman" To: "Jeremy Jackson" Cc: "LinuxBIOS" Sent: Tuesday, April 15, 2003 2:04 PM Subject: Re: 1.1 development release targets. > "Jeremy Jackson" writes: > > > > To implement this I have started a freebios2 tree at sourceforge > > > and I will gradually populating this tree. I am taking the expedient > > > of copying old code over as needed. Leaving old code for now > > > unneeded functionality to just fade away. It is the only easy > > > > Have you considered using the branch facility of CVS? > > Thought about it. > > I don't much care about the CVS history. And where there are > several trees it is less then correct a number of times. I am about > ready to rip out all of the $Id$ tags from the tree they are a consistent > source of noise when you have two different repositories and you are > comparing the differences. There is the -kk keyword substitution mode. But you have to rid the repository (at least HEAD version) of any -kb (binary) files first. > > In addition. By pulling in source modules in on demand I have a real > nice way of filtering out old dead code and showing which code works > in the new development tree. With my radical changes to hardwaremain > everything breaks until it gets fixed up. There are Exp/Stable version labels, which nobody uses, or (never done this though) you could check in an empty branch, then pull the files in one at a time. > So I guess I don't see an advantage in using the cvs branch feature, > at this point. Probably not. CVS doesn't handle chaos very well (moving files, renaming dirs, etc). But if Ron it to start creating releases at some point, CVS is definitely your friend. With production code, being able to branch an old release and merge *just* the fixes to a bug from the HEAD branch, then tag it and push out a build is smooth, like butta, it's something I've used on a production system and it just works. So not using CVS to branch probably isn't hurting anything at this point, but I think it's a good idea to start thinking about the issues. I think getting rid of binary files in the repository is pretty important. It's ok for .jpg files on a website project, where nobody uses branches (or as most web monkeys operate even release tags), but it's not good for serious projects. Regards, Jeremy From james.mcmechan at navy.mil Wed Apr 16 17:10:01 2003 From: james.mcmechan at navy.mil (McMechan, James W CIV) Date: Wed Apr 16 17:10:01 2003 Subject: Random comments on LinuxBIOS Message-ID: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Here is what I wrote while trying to catch up to list, though there is a FAQ I produced this while working my way through the emails perhaps some of it could be added to the FAQ LinuxBIOS is a project to replace the original BIOS on motherboards with open source. It appears that the two most common uses for LinuxBIOS are for Supercomputer clusters and embedded systems. LinuxBIOS no longer always uses a real OS to load the OS. LinuxBIOS currently can boot several payloads including: 1 Linux which is what started all this, hence the name. 2 Plan9 Ron Minnich seems to like this one. 3 VxWorks Felix Radensky said he had success with a older boot method 4 EtherBoot which despite its name can also boot from IDE hard disks with a patch or the dev branch 5 ADLO is a shim to allow the Boch emulator's BIOS to be used on top of LinuxBIOS to boot some versions of that other OS or to run VGA controller ROMs which seem to assume that a BIOS is present. ? The Supercomputer people appear to want remote control, remote booting and fast boot times without user interaction. For example "Press F2" is a very inconvenient task on a system with 100 computers The embedded people appear to want full control of the boot process and very fast boot times. For example who wants to watch a memory test go by on their VCR/MP3 player or see "Press F2" without a keyboard. LinuxBIOS is installed into the flash ROM in place of the original BIOS. Most flash ROMs are currently 2 Megabits in size, i.e. a 29SF020 which is only 256 Kilobytes of storage, or some might be 4 Mbit ROM which provides 512 Kbytes and so could allow for very squeezed down Linux kernel to also be included in the ROM, though I don't know of a useful kernel that small. It would be nice to add a larger flash ROM but many motherboards do not connect enough address lines to the Flash ROM socket, for even the 4 Mbit part or they solder the chip down. Another popular if grumbled about option is the disk on a chip the DoC Millennium which is a paged flash part that can have LinuxBIOS in the first part and up to 8 Mbytes of other data on other pages of the chip. Very short ~3-10 seconds boot times should be possible with this method. The BIOS Flash can usually be programmed in place by either just reflashing the original BIOS or by the hotswap method also known as the dental floss method, referring to putting dental floss under a PLCC chip to make it easy to remove without requiring metal tools in a powered on system. The hotswap method is not for the faint of heart but not all that difficult either, the system is booted with a working ROM which is then pulled out and the ROM to be programmed inserted in its place, then the ROM flashing program is run, and when done a reboot can be tried, this does not require a programmer as it uses the motherboards ability to update its BIOS as the programmer, if the reboot fails the original ROM is reinserted and the system rebooted again, all this requires, is some spare flash ROMs it is a good idea to make and label a couple backups incase something goes wrong, I have put ROMs in DIP sockets backwards destroying the ROMs, and ESD is a worry even if it is not a common problem. Some people also use a very nice device called the BIOS savior which allows for two BIOS ROMS to be switched by the user with a switch. Compact Flash IDE adaptors appear to be popular as the CF does not need the spin up time of a real hard disk. Very short boot times of ~3-10 seconds are claimed for this configuration also. Each model of motherboard currently needs its own build information due to a lack of standards in how the circuitboards are laid out between models and manufactures. If you have a motherboard that is not listed, a "lspci" will show what chipsets are in use and you can then check to see if they are in the LinuxBIOS tree, if not some research may be needed to get the documentation for that chipset. The kexec patches date back to the original LinuxBIOS with the Linux kernel in the BIOS ROM and allowed for new kernels to be loaded and executed without rebooting back to the BIOS this would allow for a minimal kernel in the BIOS and a full featured kernel to be loaded by the first kernel. Also the BIOS ROMs have a limited programming life and limited space so a small unchanging kernel that then loads the final kernel from something else prolongs the useful life of the ROMs. Ron Minnich at LANL uses similar feature called "two kernel monte" http://sourceforge.net/projects/monte/ LinuxBIOS now has the ability to make use of the CMOS space for fallback/normal boot image selection, serial port baud rates, and boot order for Etherboot, Ron Minnich also used it authentication information for a Plan9 cluster. See the cmos.layout files for information. Steven James had nice comments on Normal/Fallback images: LinuxBIOS for that board (and many others) supports two copies of LinuxBIOS, fallback and primary. The fallback image is intended to be old and stable and it's purpose is to act as a sort of rescue image. The primary image is the one that gets updated to the latest and greatest, and is meant to be more versatile and complete. The fallback image is loaded at the top of the flash, and always receives control at power on or reset. It does very minimal setup (switch to 32bit flat memory protected mode), checks and clears the CMOS boot bit, and if it was set, jumps to the primary image. This way, if the primary fails, hitting reset will boot from the fallback image alone. From ebiederman at lnxi.com Wed Apr 16 17:57:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 16 17:57:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: "McMechan, James W CIV" writes: > Here is what I wrote while trying to catch up to list, though there is a FAQ I > produced this while working my way through the emails perhaps some of it could > be added to the FAQ Would you be interested in maintaining or helping to maintain the FAQ? You seem to have a nack for putting the information together. > LinuxBIOS is a project to replace the original BIOS on motherboards with open > source. > > It appears that the two most common uses for LinuxBIOS are for Supercomputer > clusters and embedded systems. > > > LinuxBIOS no longer always uses a real OS to load the OS. That is not quite correct. LinuxBIOS requires a real OS, or at least a piece of software that can run standalone without any OS or BIOS services on the bare hardware. Which is essentially the definition of a real OS though it is not the definition of a general purpose OS. In the case of ADLO, ADLO can be considered a special purpose OS that mimics the syscall interface of a standard BIOS. > > LinuxBIOS currently can boot several payloads including: > 1 Linux which is what started all this, hence the name. > 2 Plan9 Ron Minnich seems to like this one. > 3 VxWorks Felix Radensky said he had success with a older boot method > 4 EtherBoot which despite its name can also boot from IDE hard disks with a > patch or the dev branch > > 5 ADLO is a shim to allow the Boch emulator's BIOS to be used on top of > LinuxBIOS to boot some versions of that other OS or to run VGA controller ROMs > which seem to assume that a BIOS is present. > > ? > > The Supercomputer people appear to want remote control, remote booting and fast > boot times without user interaction. For example "Press F2" is a very > inconvenient task on a system with 100 computers > > > The embedded people appear to want full control of the boot process and very > fast boot times. For example who wants to watch a memory test go by on their > VCR/MP3 player or see "Press F2" without a keyboard. In addition there is some background work on secure/trusted booting. I don't know if anything has really happened with that yet, but ADLO grow out of the interest in trusted booting. > LinuxBIOS is installed into the flash ROM in place of the original BIOS. > > Most flash ROMs are currently 2 Megabits in size, i.e. a 29SF020 which is only > 256 Kilobytes of storage, or some might be 4 Mbit ROM which provides 512 Kbytes > and so could allow for very squeezed down Linux kernel to also be included in > the ROM, though I don't know of a useful kernel that small. > > It would be nice to add a larger flash ROM but many motherboards do not connect > enough address lines to the Flash ROM socket, for even the 4 Mbit part or they > solder the chip down. For LPC flash parts the address line count is not a real issue, new boards have them and they current are up to 8Mbit in size, but do not have a theoretical limit. > Another popular if grumbled about option is the disk on a chip the DoC > Millennium which is a paged flash part that can have LinuxBIOS in the first part > and up to 8 Mbytes of other data on other pages of the chip. Very short ~3-10 > seconds boot times should be possible with this method. > > > The BIOS Flash can usually be programmed in place by either just reflashing the > original BIOS or by the hotswap method also known as the dental floss method, > referring to putting dental floss under a PLCC chip to make it easy to remove > without requiring metal tools in a powered on system. The hotswap method is not > for the faint of heart but not all that difficult either, the system is booted > with a working ROM which is then pulled out and the ROM to be programmed > inserted in its place, then the ROM flashing program is run, and when done a > reboot can be tried, this does not require a programmer as it uses the > motherboards ability to update its BIOS as the programmer, if the reboot fails > the original ROM is reinserted and the system rebooted again, all this requires, > is some spare flash ROMs it is a good idea to make and label a couple backups > incase something goes wrong, I have put ROMs in DIP sockets backwards destroying > the ROMs, and ESD is a worry even if it is not a commo! > > n problem. > > Some people also use a very nice device called the BIOS savior which allows for > two BIOS ROMS to be switched by the user with a switch. > > > Compact Flash IDE adaptors appear to be popular as the CF does not need the spin > up time of a real hard disk. Very short boot times of ~3-10 seconds are claimed > for this configuration also. > > > Each model of motherboard currently needs its own build information due to a > lack of standards in how the circuitboards are laid out between models and > manufactures. > > > If you have a motherboard that is not listed, a "lspci" will show what chipsets > are in use and you can then check to see if they are in the LinuxBIOS tree, if > not some research may be needed to get the documentation for that chipset. > > > The kexec patches date back to the original LinuxBIOS with the Linux kernel in > the BIOS ROM and allowed for new kernels to be loaded and executed without > rebooting back to the BIOS this would allow for a minimal kernel in the BIOS and > a full featured kernel to be loaded by the first kernel. Also the BIOS ROMs have > a limited programming life and limited space so a small unchanging kernel that > then loads the final kernel from something else prolongs the useful life of the > ROMs. > > > Ron Minnich at LANL uses similar feature called "two kernel monte" > http://sourceforge.net/projects/monte/ > > > LinuxBIOS now has the ability to make use of the CMOS space for fallback/normal > boot image selection, serial port baud rates, and boot order for Etherboot, Ron > Minnich also used it authentication information for a Plan9 cluster. See the > cmos.layout files for information. > > > Steven James had nice comments on Normal/Fallback images: > LinuxBIOS for that board (and many others) supports two copies of > LinuxBIOS, fallback and primary. The fallback image is intended to be old > and stable and it's purpose is to act as a sort of rescue image. The > primary image is the one that gets updated to the latest and greatest, and > is meant to be more versatile and complete. And after the most recent update the ``hair trigger'' problem has been solved. In particular it now requires several boot failures in a row before LinuxBIOS switches into fallback mode. Generally this is 3 successive failures. > The fallback image is loaded at the top of the flash, and always receives > control at power on or reset. It does very minimal setup (switch to 32bit > flat memory protected mode), checks and clears the CMOS boot bit, and if > it was set, jumps to the primary image. This way, if the primary fails, > hitting reset will boot from the fallback image alone. The mechanism has been updated to keep a failure count in the CMOS and only when the failure count crosses a threshold is the boot bit cleared. The counter is reset after LinuxBIOS has successfully loaded an image like etherboot. Etherboot now implements a similar scheme to allow two images on the disk. One at the start one at the end, but installing an image at the end of the disk still requires some work. > >From your config, you were building a primary image. the missing bytes are > for the fallback image to be appended. It didn't work because you didn't > have a fallback image at the top of the flash (the end of the image file). > > Ron Minnich expanded on fallback vs. primary: > > You need to build TWO linuxbioses, a fallback and a primary. The fallback > is 64k. The primary is (romsize-64k). > > To build a romimage, you build the fallback, build the primary, then: > > cat primary/romimage fallback/romimage > final_romimage > > flash_rom final_romimage > > Steve Gehlbach wrote about using BOOT_IDE: > See pcchips787.config in util/config for complete configuration. Note significant pieces of this information are not using the ELF bootloader, and so describes a deprecated feature. But it will not be removed in the 1.0.x stable series. In the development series the code will not be added. > option BOOT_IDE=1 > This enables booting from IDE, the file to use is linux.bin.gz: > > option IDE_BOOT_DRIVE=2 > If you do not use drive 0 (default), then you can set which drive to > boot; (0,1,2,3) are the four standard PC drives: > > option ONE_TRACK=32 > The linux.bin.gz file is put in raw form at partition 1, ie, the first > partition on the disk. This is located just past the partition table. > The partition table size varies, it is "one track" from the beginning of > the disk. "one track" in c/h/s notation is "s" or the number of sectors > per track. ONE_TRACK is in sectors, the software multiplies by 512. > Most disks are 63 sectors per track (the default), but my CF is 32 > sectors per track. eg, the partion table is 63x512 or 32x512 bytes. > > You can partition your disk as you want, but Linux goes raw in partition > 1; just make sure partition 1 is big enough, not a problem on today's > disks. You could put the Linux root file system on partition 2, for > example. In pcchips787.config, I put the Linux root file system on IDE > 0, partition 2 (I was experimenting with Linux in partition 1), but I > eventually put Linux on drive 2 using CF. You are right, copying of > linux.bin.gz raw to the partition is dangerous, and something like "cat > linux.bin.gz > /dev/hda1" will definitely screw the disk if you put the > wrong disk or partition. I recommend a shell script, fingers cannot be > trusted. You can also use "dd" but "cat" works. > > Greg Watsons had some good comments on on the boot process and file layout: > There seem to be two main parts to linuxbios. The first is > arch/{arch}/config/ctr0.base which does the very low level > initialization, like turning on memory, etc. The second is > arch/{arch}/lib/c_start.S which does whatever else is necessary to > call the C function hardwaremain(). hardwaremain() then does whatever > else is necessary to load Linux. > > c_start.S is linked with linuxbios.a, a library containing generic > support routines (those found in the lib directory) and anything > specified using the 'object' directive in a Config file (and other > stuff). The resultant 'executable' is called linuxbios_c. The loader > script used to link linuxbios_c is config/linuxbios_c.ld, and is > configured to be loaded relative to _RAMBASE. > > crt0.base is not linked against anything. Any additional assembly > routines you need must be specified using the 'mainboardinit' > directive in a Config file. This causes the specified assembly file > to be added to "crt0_includes.h" which is in turn included at the > start of crt0.base (or at the end in the case of the ppc version). > The loader script used to link crt0.base is in > arch/{arch}/config/ldscript.base. The resultant 'executable' is > called linuxbios and will be loaded at _ROMBASE. The tricky thing is > that this loader script will also load the linuxbios_c 'executable' > at a location called _payload in this file. The main task of > crt0.base is then to initialize enough hardware so that this payload > can be copied from ROM into ram (which may also involve uncompressing > code). Then control is transferred to _start, which is the first > location in linuxbios_c. > > To get an idea of how crt0.base works, look at the following files. > This is the order of execution specified by the configuration file > for sis735. > > cpu/i386/entry16.inc > cpu/i386/entry32.inc > superio/sis/950/setup_serial.inc > pc80/serial.inc > arch/i386/lib/console.inc > cpu/k7/earlymtrr.inc > northsouthbridge/sis/735/raminit.inc > arch/i386/config/crt0.base > > Next look at c_start.S which will show you what happens once control > is transferred to _start. Finally, look at > arch/{arch}/lib/hardwaremain.c to see what other stuff is done to get > Linux loaded. > > Most other files are specific to particular hardware, so it can be > pretty confusing to just browse the tree. > > Hope this helps, It Looks like a very good summary. Eric From james.mcmechan at navy.mil Wed Apr 16 19:10:01 2003 From: james.mcmechan at navy.mil (McMechan, James W CIV) Date: Wed Apr 16 19:10:01 2003 Subject: Random comments on LinuxBIOS Message-ID: <02E6075B2450E94CA159E22331C1620473EA46@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> > > Here is what I wrote while trying to catch up to list, > though there is a FAQ I > > produced this while working my way through the emails > perhaps some of it could > > be added to the FAQ > > Would you be interested in maintaining or helping to maintain the FAQ? > You seem to have a nack for putting the information together. > I would be willing, I don't have fast CVS access due to firewall issues but I can get a dial up account. Where is the FAQ located? Is it just the html or is it backed by some source format? it is fairly easy to read as html. > > LinuxBIOS no longer always uses a real OS to load the OS. > > That is not quite correct. > > LinuxBIOS requires a real OS, or at least a piece of software > that can run standalone without any OS or BIOS services on > the bare hardware. Which is essentially the definition of > a real OS though it is not the definition of a general purpose > OS. > > In the case of ADLO, ADLO can be considered a special purpose > OS that mimics the syscall interface of a standard BIOS. > I am willing to argue the other way, we no longer "always" need to load a full OS we can load elf payloads that are obviously (to users) not OSs, but merely boot loaders that don't require BIOS. Even the smallest micro-kernels do a lot more than memtest86 for example. ... > In addition there is some background work on secure/trusted > booting. I don't > know if anything has really happened with that yet, but ADLO > grow out of > the interest in trusted booting. > "grew out of" I assume, but still "trusted booting" has not been discussed much recently. Perhaps I was thinking of Ron's built in VGA ROM stuff, I may have also been the first person to suggest bochs as a GPL compatible bios example quite a while back. It appears that plex86 is hibernating though Kevin and some useful bits of the design have made it over to the bochs project recently. ... > > It would be nice to add a larger flash ROM but many > motherboards do not connect > > enough address lines to the Flash ROM socket, for even the > 4 Mbit part or they > > solder the chip down. > > For LPC flash parts the address line count is not a real > issue, new boards > have them and they current are up to 8Mbit in size, but do > not have a theoretical > limit. > LPC parts are usually soldered down right? I have not yet seen one, I thought they were part of Intel's new firmware hubs, and most boards I have seen still have DIP or PLCC which is why I said "but many motherboards" and should add that "new boards with LPC ROMs do not have the pin count limitation." > > Steven James had nice comments on Normal/Fallback images: ... > And after the most recent update the ``hair trigger'' problem has been > solved. In particular it now requires several boot failures in a row > before LinuxBIOS switches into fallback mode. Generally this > is 3 successive > failures. > I was wondering about that I realized that it had gone in but had not seen where people were hitting the hair trigger, is it part of the problem with those watchdog timers that keep resetting on people? ... > > hitting reset will boot from the fallback image alone. > I was copying verbatim his email it seem like a nice explanation > The mechanism has been updated to keep a failure count in the > CMOS and only > when the failure count crosses a threshold is the boot bit > cleared. The > counter is reset after LinuxBIOS has successfully loaded an image like > etherboot. > > Etherboot now implements a similar scheme to allow two images > on the disk. > One at the start one at the end, but installing an image at the end of > the disk still requires some work. > I had not seen any messages about etherboot having this. Spiffy ... > > Steve Gehlbach wrote about using BOOT_IDE: > > See pcchips787.config in util/config for complete configuration. > > Note significant pieces of this information are not using the ELF > bootloader, and so describes a deprecated feature. But it will > not be removed in the 1.0.x stable series. In the development series > the code will not be added. > I was thinking that it was the clearest explanation I have yet seen on it. It could be a elf payload I suppose but currently it is built in. Hum, making it into a elf payload would duplicate some of the LinuxBIOS but that would not be a problem. Also for a elf payload IDE boot I would just suggest Etherboot as the payload. ... > > > > Hope this helps, > > It Looks like a very good summary. > > Eric > > Glad you liked it, I was reading a couple weeks worth of LinuxBIOS email at once and was bothered by the questions that had, what I thought were obvious answers, so I started typing as I was reading then I checked the FAQ and dropped a few comments and wrote some more. From agnew at cs.umd.edu Wed Apr 16 19:23:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 16 19:23:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: Message-ID: <20030416195621.F85635-100000@www.missl.cs.umd.edu> > In addition there is some background work on secure/trusted booting. I don't > know if anything has really happened with that yet, but ADLO grow out of > the interest in trusted booting. The evils of software patents are still in force. It exists, it can't be released.. From ebiederman at lnxi.com Wed Apr 16 19:37:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 16 19:37:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <02E6075B2450E94CA159E22331C1620473EA46@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> References: <02E6075B2450E94CA159E22331C1620473EA46@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: "McMechan, James W CIV" writes: > > > Here is what I wrote while trying to catch up to list, > > though there is a FAQ I > > > produced this while working my way through the emails > > perhaps some of it could > > > be added to the FAQ > > > > Would you be interested in maintaining or helping to maintain the FAQ? > > You seem to have a nack for putting the information together. > > > I would be willing, I don't have fast CVS access due to firewall issues but I > can get a dial up account. > > Where is the FAQ located? Is it just the html or is it backed by some source > format? it is fairly easy to read as html. This is where I expect Ron to chime in... The answers are all in my head but have not found paper yet. > > > LinuxBIOS no longer always uses a real OS to load the OS. > > > > That is not quite correct. > > > > LinuxBIOS requires a real OS, or at least a piece of software > > that can run standalone without any OS or BIOS services on > > the bare hardware. Which is essentially the definition of > > a real OS though it is not the definition of a general purpose > > OS. > > > > In the case of ADLO, ADLO can be considered a special purpose > > OS that mimics the syscall interface of a standard BIOS. > > > I am willing to argue the other way, we no longer "always" need to load a full > OS we can load elf payloads that are obviously (to users) not OSs, but merely > boot loaders that don't require BIOS. Even the smallest micro-kernels do a lot > more than memtest86 for example. True I guess I just want the distinction to be present that the LinuxBIOS core does not support callbacks so whatever you run it must be able to stand up on it's own. Which is quite rare. > > In addition there is some background work on secure/trusted > > booting. I don't > > know if anything has really happened with that yet, but ADLO > > grow out of > > the interest in trusted booting. > > > "grew out of" I assume, but still "trusted booting" has not been discussed much > recently. Perhaps I was thinking of Ron's built in VGA ROM stuff, I may have > also been the first person to suggest bochs as a GPL compatible bios example > quite a while back. It appears that plex86 is hibernating though Kevin and some > useful bits of the design have made it over to the bochs project recently. If you look up the Adams who worked on ADLO trusted booting is their research area. I am totally mystified why we have not seen more direct feedback but I think it is a very interesting area for LinuxBIOS to expand into. The topic came up not too long ago on the etherboot developers list. Still just discussion, but all of the hooks are in the code. > ... > > > It would be nice to add a larger flash ROM but many > > motherboards do not connect > > > enough address lines to the Flash ROM socket, for even the > > 4 Mbit part or they > > > solder the chip down. > > > > For LPC flash parts the address line count is not a real > > issue, new boards > > have them and they current are up to 8Mbit in size, but do > > not have a theoretical > > limit. > > > LPC parts are usually soldered down right? I have usually seen them in an LPC socketed and in a LPC form factor. > I have not yet seen one, I thought > they were part of Intel's new firmware hubs, and most boards I have seen still > have DIP or PLCC which is why I said "but many motherboards" and should add that > "new boards with LPC ROMs do not have the pin count limitation." Newer AMD boards like the AMD760MPX also have it. Intel's firmware hub is a variation on the same them. The difference is which commands are actually used on the LPC bus. The firmware hubs have special commands so it is easier to support multiple devices. The pure LPC devices don't bother and can just be strapped to live at the appropriate address. > > > Steven James had nice comments on Normal/Fallback images: > ... > > And after the most recent update the ``hair trigger'' problem has been > > solved. In particular it now requires several boot failures in a row > > before LinuxBIOS switches into fallback mode. Generally this > > is 3 successive > > failures. > > > I was wondering about that I realized that it had gone in but had not seen where > people were hitting the hair trigger, is it part of the problem with those > watchdog timers that keep resetting on people? A typical problem is when you are installing hardware or other times you are booting a board and you turn it on and then say darn I was not yet ready to boot yet. The result of little issues like that is I was seeing an unacceptably high number of systems running in fallback. It did not cause show stopper problems because my fallback and normal images are not yet asymmetrical. But if you can't trust a machine to only be in fallback when it needs rescue recovery the scheme falls apart. So basically there was a hair trigger to get into fallback mode, and now it is some harder. > ... > > > hitting reset will boot from the fallback image alone. > > > I was copying verbatim his email it seem like a nice explanation Right and that was the way the code worked until I updated it recently. > > The mechanism has been updated to keep a failure count in the > > CMOS and only > > when the failure count crosses a threshold is the boot bit > > cleared. The > > counter is reset after LinuxBIOS has successfully loaded an image like > > etherboot. > > > > Etherboot now implements a similar scheme to allow two images > > on the disk. > > One at the start one at the end, but installing an image at the end of > > the disk still requires some work. > > > I had not seen any messages about etherboot having this. Spiffy On the documentation side it is also worth nothing the boot order options. I can currently specify which class of device to try first second and third. In addition I can specify which device of the first type tried to boot from. A lot of this was done to support Pink and MCR and now that the code works I am willing to talk about it. There was a little mention of it during the design phaze but mostly there was not much feedback. And the discussion really would have occurred on the etherboot list. Which is another reason you might have missed it. > ... > > > Steve Gehlbach wrote about using BOOT_IDE: > > > See pcchips787.config in util/config for complete configuration. > > > > Note significant pieces of this information are not using the ELF > > bootloader, and so describes a deprecated feature. But it will > > not be removed in the 1.0.x stable series. In the development series > > the code will not be added. > > > I was thinking that it was the clearest explanation I have yet seen on it. > It could be a elf payload I suppose but currently it is built in. > Hum, making it into a elf payload would duplicate some of the LinuxBIOS but that > would not be a problem. > > Also for a elf payload IDE boot I would just suggest Etherboot as the payload. The case where IDE_BOOT and an linux kernel comes into play is if you have a system with compact flash and that is the only location large enough to put it. But yes, in most instances I would suggest etherboot as well. > ... > > > > > > Hope this helps, > > > > It Looks like a very good summary. > > > > Eric > > > > > Glad you liked it, I was reading a couple weeks worth of LinuxBIOS email at once > and was bothered by the questions that had, what I thought were obvious answers, > so I started typing as I was reading then I checked the FAQ and dropped a few > comments and wrote some more. Now just to figure out how to get this incorporated into the FAQ. However it gets maintained I am certain from this email that it really should be posted to the list periodically so people just reading their email will see it and we can point out places it needs corrections just by replying... If nothing else it is a good way to draw answers out of my head... Eric From ebiederman at lnxi.com Wed Apr 16 19:41:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 16 19:41:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030416195621.F85635-100000@www.missl.cs.umd.edu> References: <20030416195621.F85635-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > > In addition there is some background work on secure/trusted booting. I don't > > know if anything has really happened with that yet, but ADLO grow out of > > the interest in trusted booting. > > The evils of software patents are still in force. It exists, it can't be > released.. Is this the patent application process or due to conflicting patents? In any case I would love to get a good feel for what problems are being addressed and roughly how that is done. As time goes by this is one of the features I would like to see well integrated if parts of it are not built on rubish ideas. Eric From jerj at coplanar.net Wed Apr 16 21:41:01 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Wed Apr 16 21:41:01 2003 Subject: Random comments on LinuxBIOS References: <20030416195621.F85635-100000@www.missl.cs.umd.edu> Message-ID: <01c101c30487$4eddca40$7c07a8c0@kennet.coplanar.net> If anyone has some actual patent #'s we can all find out more at http://delphion.com Of interest so far I find: US6546489: Disk drive which provides a secure boot of a host computer system from a protected area of a disk -doesn't sound like it uses PKI but it's a start US6175924: Method and apparatus for protecting application data in secure storage areas -uses PKI, but for userspace code. And we could see the really interesting stuff about MS's Palladium (I know, they changed the name...) pending applications for a mere $75 a month. Jeremy ----- Original Message ----- From: "Adam Agnew" To: "Eric W. Biederman" Cc: "McMechan, James W CIV" ; Sent: Wednesday, April 16, 2003 7:58 PM Subject: Re: Random comments on LinuxBIOS > > > In addition there is some background work on secure/trusted booting. I don't > > know if anything has really happened with that yet, but ADLO grow out of > > the interest in trusted booting. > > The evils of software patents are still in force. It exists, it can't be > released.. > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From agnew at cs.umd.edu Wed Apr 16 22:18:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 16 22:18:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <01c101c30487$4eddca40$7c07a8c0@kennet.coplanar.net> Message-ID: <20030416223905.X85912-100000@www.missl.cs.umd.edu> The one in question is 6,185,678 http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) And from my understanding, it's owned by the U of Penn and comes down to their decision when an open source implementation can be released. - Adam A. On Wed, 16 Apr 2003, Jeremy Jackson wrote: > If anyone has some actual patent #'s we can all find out more at > http://delphion.com > > Of interest so far I find: > > US6546489: Disk drive which provides a secure boot of a host computer system > from a protected area of a disk > -doesn't sound like it uses PKI but it's a start > US6175924: Method and apparatus for protecting application data in secure > storage areas > -uses PKI, but for userspace code. > > And we could see the really interesting stuff about MS's Palladium (I know, > they changed the name...) pending applications for a mere $75 a month. > > Jeremy > ----- Original Message ----- > From: "Adam Agnew" > To: "Eric W. Biederman" > Cc: "McMechan, James W CIV" ; > > Sent: Wednesday, April 16, 2003 7:58 PM > Subject: Re: Random comments on LinuxBIOS > > > > > > > In addition there is some background work on secure/trusted booting. I > don't > > > know if anything has really happened with that yet, but ADLO grow out of > > > the interest in trusted booting. > > > > The evils of software patents are still in force. It exists, it can't be > > released.. > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From ollie at sis.com.tw Wed Apr 16 22:54:00 2003 From: ollie at sis.com.tw (ollie lho) Date: Wed Apr 16 22:54:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030416223905.X85912-100000@www.missl.cs.umd.edu> References: <20030416223905.X85912-100000@www.missl.cs.umd.edu> Message-ID: <1050549730.19637.1.camel@ollie> On Thu, 2003-04-17 at 10:53, Adam Agnew wrote: > The one in question is 6,185,678 > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) > > And from my understanding, it's owned by the U of Penn and comes down to > their decision when an open source implementation can be released. > Doesn't the professor (Bill) own a patent on secure boot ? Can you cross license with U of Penn ? -- ollie lho From steve at nexpath.com Wed Apr 16 22:56:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 16 22:56:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: <3E9E232B.7040402@nexpath.com> Eric W. Biederman wrote: > For LPC flash parts the address line count is not a real issue, new boards > have them and they current are up to 8Mbit in size, but do not have a theoretical > limit. This reminds me of something I have intended to mention. Right now, linuxbios starts up in the high 4G and jumps to the start of the assy code, which is located at 0xffff0004 or 0xf0004 depending on the setting of biosbase (which sets _ROMBASE). For biosbase set to 0xffff0000, the jump is a 16-bit relative jump to 0xffff0004 and so it does a kind of wrap around I think (segment register not reloaded). Given the large LPC flash parts are coming, and so large flash storage, it seems we should change the startup to set 32-bit mode first, then jump 32-bit to the start of the assy code. The usual way that I have done this jump relative on reset enough bytes below the reset vector, to setup a flat gdt, and go into 32-bit mode, then jump 32-bit to the flash code start. This allows one to start anywhere in flash, not just the top 64k-bytes. I also think we should default to wp cacheing for the flash region as well. In my view anyway, the current startup jump seems clumsy, and I am not clear on why one would want to jump to 0xf0000 (the default) and run there (for those that aren't familiar 0xf0000 is normally aliased to 0xffff0000 by most if not all chipsets on startup, and doesn't become ram until you enable it). But being in 32-bit mode you could go where you want. Maybe there is a good reason for this but it escapes me. -Steve From ollie at sis.com.tw Wed Apr 16 22:59:01 2003 From: ollie at sis.com.tw (ollie lho) Date: Wed Apr 16 22:59:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030416223905.X85912-100000@www.missl.cs.umd.edu> References: <20030416223905.X85912-100000@www.missl.cs.umd.edu> Message-ID: <1050550023.19637.6.camel@ollie> On Thu, 2003-04-17 at 10:53, Adam Agnew wrote: > The one in question is 6,185,678 > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) > > And from my understanding, it's owned by the U of Penn and comes down to > their decision when an open source implementation can be released. > > - Adam A. Wow !! It is really a patent evil. Prof Arbaugh invented the patent but assigned it to U of Penn such that he and/or his student can not do some things related in U of Maryland ?? -- ollie lho From ollie at sis.com.tw Wed Apr 16 23:16:01 2003 From: ollie at sis.com.tw (ollie lho) Date: Wed Apr 16 23:16:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <3E9E232B.7040402@nexpath.com> References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <3E9E232B.7040402@nexpath.com> Message-ID: <1050551118.19637.14.camel@ollie> On Thu, 2003-04-17 at 11:44, Steve Gehlbach wrote: > > This reminds me of something I have intended to mention. Right now, > linuxbios starts up in the high 4G and jumps to the start of the assy > code, which is located at 0xffff0004 or 0xf0004 depending on the setting > of biosbase (which sets _ROMBASE). For biosbase set to 0xffff0000, the > jump is a 16-bit relative jump to 0xffff0004 and so it does a kind of > wrap around I think (segment register not reloaded). > > Given the large LPC flash parts are coming, and so large flash storage, > it seems we should change the startup to set 32-bit mode first, then > jump 32-bit to the start of the assy code. The usual way that I have > done this jump relative on reset enough bytes below the reset vector, to > setup a flat gdt, and go into 32-bit mode, then jump 32-bit to the flash > code start. This allows one to start anywhere in flash, not just the > top 64k-bytes. I also think we should default to wp cacheing for the > flash region as well. > > In my view anyway, the current startup jump seems clumsy, and I am not > clear on why one would want to jump to 0xf0000 (the default) and run > there (for those that aren't familiar 0xf0000 is normally aliased to > 0xffff0000 by most if not all chipsets on startup, and doesn't become > ram until you enable it). But being in 32-bit mode you could go where > you want. > The very first LinuxBIOS implementation does almost this. Part of the historical reason of 16-bit code is, for DoC with only 512 IPL area, every bit counts. So the asm code for chipsets with DoC support are coded in 16-bit mode instead of 32-bit mode. Now, it seems that DoC is obselete by the non-open attitude of its manufacture and Firmware Hub, we may consider consolidating these asm code to 32-bit in the next development cycle. -- ollie lho From steve at nexpath.com Wed Apr 16 23:20:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 16 23:20:01 2003 Subject: 1.1 development release targets. In-Reply-To: References: <3E9D8C9A.9050606@nexpath.com> Message-ID: <3E9E28DA.6000400@nexpath.com> > >>Does linuxbios in the current version support elfboot images from ide/floppy >>(directly, without etherboot)? I guess if it does I don't see it. > > Yes. The switch on the ``rom loader'' is the same on code, for > either the elf loader or the other one. > Thanks, I guess I never tried it since I needed to fiddle with the zero page parameters. > >>One issue I have with elfboot images is the inability to dynamically pass >>parameters to the zero page (specifically the vga cursor location). I understand >>the idea of not locking the boot mechanism into the bios, but there are pros and >>cons. > > > As ron mentioned putting enough information in the LinuxBIOS table so > the code in mkelfImage can properly position the vga cursor location is > straight forward. > I don't understand. Which linuxbios table are you referring to? Is this a proposal or currently implemented? -Steve From steve at nexpath.com Wed Apr 16 23:58:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 16 23:58:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <1050551118.19637.14.camel@ollie> References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <3E9E232B.7040402@nexpath.com> <1050551118.19637.14.camel@ollie> Message-ID: <3E9E3190.2030604@nexpath.com> ollie lho wrote: > The very first LinuxBIOS implementation does almost this. Part of the > historical reason of 16-bit code is, for DoC with only 512 IPL area, > every bit counts. So the asm code for chipsets with DoC support are > coded in 16-bit mode instead of 32-bit mode. Now, it seems that DoC is > obselete by the non-open attitude of its manufacture and Firmware Hub, > we may consider consolidating these asm code to 32-bit in the next > development cycle. > Thanks, I suspected it had something to do with 16-bit code, but I couldn't understand why. This explains it. Also, CF is much cheaper than DOC, too. -Steve From ebiederman at lnxi.com Thu Apr 17 01:27:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 01:27:01 2003 Subject: 1.1 development release targets. In-Reply-To: <3E9E28DA.6000400@nexpath.com> References: <3E9D8C9A.9050606@nexpath.com> <3E9E28DA.6000400@nexpath.com> Message-ID: Steve Gehlbach writes: > > > >>Does linuxbios in the current version support elfboot images from ide/floppy > >>(directly, without etherboot)? I guess if it does I don't see it. > > Yes. The switch on the ``rom loader'' is the same on code, for > > either the elf loader or the other one. > > > > Thanks, I guess I never tried it since I needed to fiddle with the > zero page parameters. I have found that most of the time it is enough to simply tell the kernel it has a VGA adapter. And the rest it seems to handle properly. That is what I currently do in mkelfImage. > >>One issue I have with elfboot images is the inability to dynamically pass > >>parameters to the zero page (specifically the vga cursor location). I > understand > > >>the idea of not locking the boot mechanism into the bios, but there are pros > and > > >>cons. > > As ron mentioned putting enough information in the LinuxBIOS table so > > the code in mkelfImage can properly position the vga cursor location is > > straight forward. > > > > I don't understand. Which linuxbios table are you referring to? Is > this a proposal or currently implemented? Currently implemented. see src/include/boot/linuxbios_tables.h The LinuxBIOS table on x86 is located near address 0, or around 0xF0000 - 0x100000. It is a binary table of tagged data so it can be grown and expanded while still retaining backwards compatibility. It is at a fixed address so we don't need to pass it explicitly to anyone, but the data in it is position independent in case the table needs to be moved. In addition the table is composed of fixed size data types so it can be used on dual 32/64 bit platforms without surprises. This is how we currently pass memory size information. We have mptables and irq tables setup separately mostly for historical reasons. As we rework things in 1.1. for LinuxBIOS to assign irqs I want to get those in there as well and generate the other tables from the LinuxBIOS table. Any binary loaded under LinuxBIOS can go out look for and find the LinuxBIOS table. We tried passing parameters explicitly but the LinuxBIOS table is much more stable, for large vary amounts of data. Especially when you interpose something like etherboot in between your BIOS and your loaded kernel. Eric From ebiederman at lnxi.com Thu Apr 17 01:42:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 01:42:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <3E9E232B.7040402@nexpath.com> References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <3E9E232B.7040402@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > > For LPC flash parts the address line count is not a real issue, new boards > > have them and they current are up to 8Mbit in size, but do not have a > theoretical > > > limit. > > This reminds me of something I have intended to mention. Right now, linuxbios > starts up in the high 4G and jumps to the start of the assy code, which is > located at 0xffff0004 or 0xf0004 depending on the setting of biosbase (which > sets _ROMBASE). For biosbase set to 0xffff0000, the jump is a 16-bit relative > jump to 0xffff0004 and so it does a kind of wrap around I think (segment > register not reloaded). The CPU starts at: CS = 0xFFFF0000 IP=0xFFF0 So you have an entire 64K to work with. > Given the large LPC flash parts are coming, and so large flash storage, it seems > > we should change the startup to set 32-bit mode first, then jump 32-bit to the > start of the assy code. The usual way that I have done this jump relative on > reset enough bytes below the reset vector, to setup a flat gdt, and go into > 32-bit mode, then jump 32-bit to the flash code start. This allows one to start > anywhere in flash, not just the top 64k-bytes. I also think we should default to > wp cacheing for the flash region as well. WP caching?? We should cache it certainly but note that WP caching != Write Protect in the normal sense. Except for the location the 16->32bit transition this is pretty much what the code currently does. And for our ``normal'' images when we have 2 copies of LinuxBIOS the start up is entirely 32bit. > In my view anyway, the current startup jump seems clumsy, and I am not clear on > why one would want to jump to 0xf0000 (the default) and run there (for those > that aren't familiar 0xf0000 is normally aliased to 0xffff0000 by most if not > all chipsets on startup, and doesn't become ram until you enable it). But being > in 32-bit mode you could go where you want. There are other reasons for keeping the BIOS image small so a 64K restriction has not been a big deal. That is the reason it has not changed up to this point. But thank you for the reminder. This is something that is worth looking at the development branch along with removing the estimates of LinuxBIOS size that we currently use. > Maybe there is a good reason for this but it escapes me. Mostly it is one of those we started with it and have not yet found it to be a real limitation. Eric From mjw99 at ic.ac.uk Thu Apr 17 02:34:00 2003 From: mjw99 at ic.ac.uk (Mark Williamson) Date: Thu Apr 17 02:34:00 2003 Subject: Newbie question on Message-ID: <3E9E525A.8070101@ic.ac.uk> Dear List, I've just been looking through the mail archive for this list at http://www.clustermatic.org/pipermail/linuxbios/ and throught the CVS code for the project. I am sorry to spam the list, but I am still unsure if my motherboard is supported; [root at machine root]# lspci 00:00.0 Host bridge: VIA Technologies, Inc. VT82C693A/694x [Apollo PRO133x] (rev 42) 00:01.0 PCI bridge: VIA Technologies, Inc. VT82C598/694x [Apollo MVP3/Pro133x AGP] 00:07.0 ISA bridge: VIA Technologies, Inc. VT82C596 ISA [Mobile South] (rev 11) 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586B PIPC Bus Master IDE (rev 06) 00:07.2 USB Controller: VIA Technologies, Inc. USB (rev 05) 00:07.3 Host bridge: VIA Technologies, Inc. VT82C596 Power Management (rev 20) 00:0a.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 00:0b.0 Network controller: Standard Microsystems Corp [SMC] SMC2602W EZConnect / Addtron AWA-100 (rev 02) 01:00.0 VGA compatible controller: Matrox Graphics, Inc. MGA G200 AGP (rev 01) If any further details are required, please email and I will put the verbose response(s) on a url somewhere and just link to it, to keep the spam level down :) Best regards, Mark From ebiederman at lnxi.com Thu Apr 17 02:37:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 02:37:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030416223905.X85912-100000@www.missl.cs.umd.edu> References: <20030416223905.X85912-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > The one in question is 6,185,678 > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) That was an interesting read. At least I now have an idea of what Bill was thinking of. Most interesting is that there is not a mechanism for the trust to go both ways. In particular how is the loaded code to know it is running on a trusted system. In addition there are some fundamental things in his description that I would simply not implement as described. Nastily extending DHCP and TFTP when IPsec could be used. And in general I don't think any trust is needed at all of the Network Packets. Just the loaded image needs to carry a signature that can be verified. And the description does not address when the system has exploitable bugs. In particular systems like the X-box can be compromised with buffer overflows and other security standard security holes. Allowing an untrusted application to gain special privileges on the machine. This is not to say that strong/trusted integrity checks of the components of the system are a bad idea. But rather to show that simply loading trusted components does not give a secure system. That requires only trusting bug free software which as an engineering assumption is impossible. > And from my understanding, it's owned by the U of Penn and comes down to > their decision when an open source implementation can be released. The universities playing with ``IP'' and stifling innovation... Sigh. From ebiederman at lnxi.com Thu Apr 17 02:52:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 02:52:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <1050550023.19637.6.camel@ollie> References: <20030416223905.X85912-100000@www.missl.cs.umd.edu> <1050550023.19637.6.camel@ollie> Message-ID: ollie lho writes: > > Wow !! It is really a patent evil. Prof Arbaugh invented the patent > but assigned it to U of Penn such that he and/or his student can not > do some things related in U of Maryland ?? I wonder if there is enough wiggle room in there to walk around the patent in a LinuxBIOS scenario. Our standing assumptions are quite different, and there are some of the pieces suggested that have much cleaner alternatives. As I slowly digest the ideas I have a very hard time with the idea of signing a boot block, and believing that will provide some measure of security and trust. At that point there is software chaining but more than that there is not enough room in a boot block to verify the loaded operating system. Unless someone is a lot more creative with 512 bytes that I am. Quite a few of the pieces I know it is safe to write and deploy while they do not compose a complete solution. Teaching etherboot about IPsec is totally outside the scope of the patent, for example. Though for Bill and his students there may be some good faith with the university system that keeps them from exercising all of their options. Eric From jerj at coplanar.net Thu Apr 17 09:48:00 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Thu Apr 17 09:48:00 2003 Subject: Random comments on LinuxBIOS References: <20030416223905.X85912-100000@www.missl.cs.umd.edu> Message-ID: <003301c304ec$df4dd9c0$7c07a8c0@kennet.coplanar.net> ----- Original Message ----- From: "Eric W. Biederman" To: "Adam Agnew" Cc: "Jeremy Jackson" ; "McMechan, James W CIV" ; Sent: Thursday, April 17, 2003 3:11 AM Subject: Re: Random comments on LinuxBIOS > Adam Agnew writes: > > > The one in question is 6,185,678 > > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtm l/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=Wil liam+AND+Arbaugh&RS=(William+AND+Arbaugh) > > That was an interesting read. At least I now have an idea of what > Bill was thinking of. Most interesting is that there is not a mechanism > for the trust to go both ways. In particular how is the loaded > code to know it is running on a trusted system. I think that is addressed by MS's Palladium (new name escapes me). From agnew at cs.umd.edu Thu Apr 17 09:59:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 17 09:59:00 2003 Subject: Secure boot and patents ( was Re: Random comments on LinuxBIOS) In-Reply-To: Message-ID: <20030417103144.G86488-100000@www.missl.cs.umd.edu> [The content of this email was forwarded on request from Bill Arbaugh] All, Adam Agnew asked me to respond to the issues involving UPenn's patent based on my dissertation work. If you've done any technical work at University here in the US, then you know that the University where you are going to school (or working) owns your IP in most circumstances (not all fortunately). As a result, the patent on my dissertation research is owned by UPenn and not myself. When we started the secure boot work at Maryland, I had assurances from UPenn that we could release the code to the open source community. I've been working with the lawyers at UPenn for two years now to make that happen. A slow and painful process. I'm told that the paperwork to do that is now awaiting approval. I have not seen the final paperwork, nor even an initial draft. So I have no idea, what it looks like. Once it is possible to release the secure boot code, we will. The ADLO work (minus the security hooks) is a different matter completely. While it came out of the secure boot work, it is not encumbered by the UPenn patent and can be released. Bill On 17 Apr 2003, Eric W. Biederman wrote: > Adam Agnew writes: > > > The one in question is 6,185,678 > > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) > > That was an interesting read. At least I now have an idea of what > Bill was thinking of. Most interesting is that there is not a mechanism > for the trust to go both ways. In particular how is the loaded > code to know it is running on a trusted system. > > In addition there are some fundamental things in his description > that I would simply not implement as described. Nastily extending > DHCP and TFTP when IPsec could be used. And in general I don't think > any trust is needed at all of the Network Packets. Just the loaded > image needs to carry a signature that can be verified. > > And the description does not address when the system has exploitable > bugs. In particular systems like the X-box can be compromised with buffer > overflows and other security standard security holes. Allowing an untrusted > application to gain special privileges on the machine. > > This is not to say that strong/trusted integrity checks of the components > of the system are a bad idea. But rather to show that simply loading > trusted components does not give a secure system. That requires only > trusting bug free software which as an engineering assumption is impossible. > > > And from my understanding, it's owned by the U of Penn and comes down to > > their decision when an open source implementation can be released. > > The universities playing with ``IP'' and stifling innovation... > Sigh. > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rminnich at lanl.gov Thu Apr 17 10:22:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 10:22:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: nice job. The one thing I'd add is it can be useful to make tags then vi -t hardwaremain (or emacs equivalent) and walk the source. Is there someplace on the web page you'd like to see this go? ron From agnew at cs.umd.edu Thu Apr 17 10:23:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 17 10:23:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: Message-ID: <20030417104102.Y86488-100000@www.missl.cs.umd.edu> On 17 Apr 2003, Eric W. Biederman wrote: > Adam Agnew writes: > > > The one in question is 6,185,678 > > http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/search-adv.htm&r=1&f=G&l=50&d=PTXT&p=1&p=1&S1=(William+AND+Arbaugh)&OS=William+AND+Arbaugh&RS=(William+AND+Arbaugh) > > That was an interesting read. At least I now have an idea of what > Bill was thinking of. Most interesting is that there is not a mechanism > for the trust to go both ways. In particular how is the loaded > code to know it is running on a trusted system. > > In addition there are some fundamental things in his description > that I would simply not implement as described. Nastily extending > DHCP and TFTP when IPsec could be used. And in general I don't think > any trust is needed at all of the Network Packets. Just the loaded > image needs to carry a signature that can be verified. > A lot of things like that in the description weren't implemented by choice. All we've done is to check elf image signatures so far. I think you'd agree that's the only part that was really necessary towards getting an operating system up in a trusted state. As long as you confine execution to elf images, you can continue to chain along.. As far as etherboot goes, same rules. No need to worry about individual packets. I'm just going to check the signature once it all arrives anyway. > > And the description does not address when the system has exploitable > bugs. In particular systems like the X-box can be compromised with buffer > overflows and other security standard security holes. Allowing an untrusted > application to gain special privileges on the machine. Right, we trust that the components are signed and therefor an authority intended to grant the component permission to run on the machine. It does not mean that we trust the new component not to break the chain of trust, or be bug ridden. Why, we even thought about "trusting" certain closed operating systems! From agnew at cs.umd.edu Thu Apr 17 10:29:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Thu Apr 17 10:29:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: Message-ID: <20030417105930.D86488-100000@www.missl.cs.umd.edu> On 17 Apr 2003, Eric W. Biederman wrote: > ollie lho writes: > > > > Wow !! It is really a patent evil. Prof Arbaugh invented the patent > > but assigned it to U of Penn such that he and/or his student can not > > do some things related in U of Maryland ?? > > I wonder if there is enough wiggle room in there to walk around the > patent in a LinuxBIOS scenario. Our standing assumptions are quite > different, and there are some of the pieces suggested that have much > cleaner alternatives. > > As I slowly digest the ideas I have a very hard time with the idea > of signing a boot block, and believing that will provide some measure > of security and trust. At that point there is software chaining but > more than that there is not enough room in a boot block to verify > the loaded operating system. Unless someone is a lot more creative > with 512 bytes that I am. I can't think of a way to sign a boot block either. That's why its nice to just load the linux kernel as an elf image. When you start booting expansion roms, loading bootloaders from boot sectors, booting off of floppies, loading Windows.. it gets a lot fuzzier.. > > Quite a few of the pieces I know it is safe to write and deploy > while they do not compose a complete solution. Teaching etherboot > about IPsec is totally outside the scope of the patent, for example. > > Though for Bill and his students there may be some good faith with > the university system that keeps them from exercising all of their > options. Yeah. From rminnich at lanl.gov Thu Apr 17 10:50:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 10:50:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: Message-ID: On 16 Apr 2003, Eric W. Biederman wrote: > This is where I expect Ron to chime in... The answers are all in my > head but have not found paper yet. it's on the web page and out of data. I have a ton of email to add faq entries. Our web server went away and it's been much less easy to edit the web page since that happened. ron From ebiederman at lnxi.com Thu Apr 17 11:00:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 11:00:01 2003 Subject: Stable storage of BIOS parameters... Message-ID: Currently I have implemented code to store BIOS options in the cmos, and the layout of the options is reported by the LinuxBIOS table. Going forward there are going to be options that we cannot store in the cmos. Either because of size or because of unreliability of the cmos. I think I want to setup a small jffs2 filesystem in the rom chip, for maintaining and manipulating those options. The concern I have is that this sucks up a lot of flash blocks (at least two), because of the size of the information. The one piece of information I do not feel comfortable storing there is the cmos boot count and normal/failsafe settings. Because LinuxBIOS must set those and _reliably_ writing to a flash chip appears to be a non-trivial exercise. By using jffs2 other people can manipulate the stable store as well. My big question is how should I encode the options that LinuxBIOS utilizes. This is a piece I have not implemented yet, but that looks increasing useful especially as a store of various bits of persistent data. Especially when things like cryptographic keys get involved. Eric From rminnich at lanl.gov Thu Apr 17 11:04:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 11:04:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <3E9E232B.7040402@nexpath.com> Message-ID: On Wed, 16 Apr 2003, Steve Gehlbach wrote: > Given the large LPC flash parts are coming, and so large flash storage, > it seems we should change the startup to set 32-bit mode first, then > jump 32-bit to the start of the assy code. The usual way that I have > done this jump relative on reset enough bytes below the reset vector, to > setup a flat gdt, and go into 32-bit mode, then jump 32-bit to the flash > code start. This allows one to start anywhere in flash, not just the > top 64k-bytes. I also think we should default to wp cacheing for the > flash region as well. I would rather not. There are chipset reasons for just assuming that only the top 64K of flash is usable, which is why we have left it as is. The use of the top 64K is not just due to real mode issues. ron From rminnich at lanl.gov Thu Apr 17 11:10:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 11:10:01 2003 Subject: Newbie question on In-Reply-To: <3E9E525A.8070101@ic.ac.uk> Message-ID: I don't think we support the apollo pro. ron From ebiederman at lnxi.com Thu Apr 17 11:11:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 11:11:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030417104102.Y86488-100000@www.missl.cs.umd.edu> References: <20030417104102.Y86488-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > A lot of things like that in the description weren't implemented by > choice. All we've done is to check elf image signatures so far. I think > you'd agree that's the only part that was really necessary towards > getting an operating system up in a trusted state. As long as you > confine execution to elf images, you can continue to chain along.. As far > as etherboot goes, same rules. No need to worry about individual packets. > I'm just going to check the signature once it all arrives anyway. Ok interesting. The implemented code base is quite different from the description in the patent :) In the etherboot case there is a bit of vulnerability with respect to replay attacks where an attacker could trick the software into loading and old but buggy but previously signed image. > > And the description does not address when the system has exploitable > > bugs. In particular systems like the X-box can be compromised with buffer > > overflows and other security standard security holes. Allowing an untrusted > > application to gain special privileges on the machine. > > Right, we trust that the components are signed and therefor an authority > intended to grant the component permission to run on the machine. It does > not mean that we trust the new component not to break the chain of trust, > or be bug ridden. Why, we even thought about "trusting" certain closed > operating systems! Yep, my main goal here was to understand the limitations. A lot of the noise you hear from other systems talking about trusted booting seems to promise the world. Eric From rminnich at lanl.gov Thu Apr 17 11:16:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 11:16:00 2003 Subject: Stable storage of BIOS parameters... In-Reply-To: Message-ID: On 17 Apr 2003, Eric W. Biederman wrote: > > By using jffs2 other people can manipulate the stable store as well. > My big question is how should I encode the options that LinuxBIOS > utilizes. I really am liking S-expressions for this type of thing nowadays. Trivial to parse, we have a small fast library for this (see http://sourceforge.net/project/showfiles.php?group_id=521&release_id=111528) ron From aip at cwlinux.com Thu Apr 17 11:19:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Thu Apr 17 11:19:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <3E9E3190.2030604@nexpath.com>; from Steve Gehlbach on Wed, Apr 16, 2003 at 09:46:08PM -0700 References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <3E9E232B.7040402@nexpath.com> <1050551118.19637.14.camel@ollie> <3E9E3190.2030604@nexpath.com> Message-ID: <20030417235116.A2310@mail.cwlinux.com> > Also, CF is much cheaper than DOC, too. FYI, snapgear finished the DOC Millennium plus support on Linux, and it is under GPL. It could be a new option for LinuxBIOS. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From steve at nexpath.com Thu Apr 17 11:43:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 17 11:43:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <3E9E232B.7040402@nexpath.com> Message-ID: <3E9ED4DC.2000802@nexpath.com> Eric W. Biederman wrote: >>I also think we should default to >>wp cacheing for the flash region as well. > > > WP caching?? We should cache it certainly but note > that WP caching != Write Protect in the normal sense. > > Except for the location the 16->32bit transition this is pretty much > what the code currently does. And for our ``normal'' images when > we have 2 copies of LinuxBIOS the start up is entirely 32bit. I was referring to memory type=5, which Intel calls WP (vol 3, 10-23). It is not by default done for the upper 4G regions in linuxbios AFAIK (biosbase=0xffff0000). One has to set XIP_ROM_BASE, XIP_ROM_SIZE, and then an mtrr is set of type 5 for this region, I am pretty sure. It made a big difference in the pcchips787, as I could not use compression otherwise, took too long to decompress, and the watchdog reset kicked off. I have not studied in detail what WP means precisely but I know that others use this type for flash regions. -Steve From steve at nexpath.com Thu Apr 17 11:56:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 17 11:56:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: References: Message-ID: <3E9ED7BC.6090801@nexpath.com> ron minnich wrote: > On Wed, 16 Apr 2003, Steve Gehlbach wrote: > > >>Given the large LPC flash parts are coming, and so large flash storage, >>it seems we should change the startup to set 32-bit mode first, then >>jump 32-bit to the start of the assy code. The usual way that I have >>done this jump relative on reset enough bytes below the reset vector, to >>setup a flat gdt, and go into 32-bit mode, then jump 32-bit to the flash >>code start. This allows one to start anywhere in flash, not just the >>top 64k-bytes. I also think we should default to wp cacheing for the >>flash region as well. > > > I would rather not. There are chipset reasons for just assuming that only > the top 64K of flash is usable, which is why we have left it as is. The > use of the top 64K is not just due to real mode issues. > Maybe this is too minor of an issue to take up everyones time on. It just seems to me that as the flash parts get larger, limiting ourselves to 64K of code space is being trapped in the past. But we do now have compression which helps a lot. -Steve From stuge-linuxbios at cdy.org Thu Apr 17 12:02:33 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Thu Apr 17 12:02:33 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <20030417104102.Y86488-100000@www.missl.cs.umd.edu> References: <20030417104102.Y86488-100000@www.missl.cs.umd.edu> Message-ID: <20030417162319.GC29510@foo.birdnet.se> On Thu, Apr 17, 2003 at 10:59:07AM -0400, Adam Agnew wrote: > A lot of things like that in the description weren't implemented by > choice. All we've done is to check elf image signatures so far. I think > you'd agree that's the only part that was really necessary towards > getting an operating system up in a trusted state. As long as you > confine execution to elf images, you can continue to chain along.. As far > as etherboot goes, same rules. No need to worry about individual packets. > I'm just going to check the signature once it all arrives anyway. I may be missing something (please tell me!) but is there any real point in checking signatures when the checking and the public key isn't done in hardware, like in the X-Box? If the OS allows a complete write of the memory area where data, signatures or public keys are stored, this allows the administrator to distribute updated versions of the code using only software. It will allow the hacker to do the exact same thing however. If the OS doesn't allow a complete write of that memory, no automated lowlevel firmware updates will be possible either for root or r00t. How does it become secure? (Yeah, I should read the patent. Will it answer the question?) //Peter From rminnich at lanl.gov Thu Apr 17 12:51:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 12:51:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <3E9ED7BC.6090801@nexpath.com> Message-ID: On Thu, 17 Apr 2003, Steve Gehlbach wrote: > Maybe this is too minor of an issue to take up everyones time on. It > just seems to me that as the flash parts get larger, limiting ourselves > to 64K of code space is being trapped in the past. But we do now have > compression which helps a lot. you're not totally limited to 64k, it's just the code that makes you not limited to 64k is limited to 64k, if you get my drift. ron From jerj at coplanar.net Thu Apr 17 12:55:01 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Thu Apr 17 12:55:01 2003 Subject: Random comments on LinuxBIOS References: <20030416223905.X85912-100000@www.missl.cs.umd.edu><1050550023.19637.6.camel@ollie> Message-ID: <009301c30507$09757b30$7c07a8c0@kennet.coplanar.net> ----- Original Message ----- From: "Eric W. Biederman" To: "ollie lho" Cc: "Adam Agnew" ; "Jeremy Jackson" ; "McMechan, James W CIV" ; "LinuxBIOS Mailing List" Sent: Thursday, April 17, 2003 3:26 AM Subject: Re: Random comments on LinuxBIOS > ollie lho writes: > > > > Wow !! It is really a patent evil. Prof Arbaugh invented the patent > > but assigned it to U of Penn such that he and/or his student can not > > do some things related in U of Maryland ?? > > I wonder if there is enough wiggle room in there to walk around the > patent in a LinuxBIOS scenario. Our standing assumptions are quite > different, and there are some of the pieces suggested that have much > cleaner alternatives. That was my thought as well. To be air tight would require getting (ie purchasing a copy) of the full text of the patent, and comparing etherboot, etc. against the claims. It specifically states (as do most patents) that the examples given (ie the freely available part you get on the website) do not limit the scope of the patent, and that the claims state what is covered by the patent. Regards, Jeremy From james.mcmechan at navy.mil Thu Apr 17 13:05:01 2003 From: james.mcmechan at navy.mil (McMechan, James W CIV) Date: Thu Apr 17 13:05:01 2003 Subject: Random comments on LinuxBIOS Message-ID: <02E6075B2450E94CA159E22331C1620473EAD1@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> > From: ron minnich [mailto:rminnich at lanl.gov] ... > On 16 Apr 2003, Eric W. Biederman wrote: > > > This is where I expect Ron to chime in... The answers are all in my > > head but have not found paper yet. > > it's on the web page and out of data. I have a ton of email > to add faq > entries. Our web server went away and it's been much less > easy to edit the > web page since that happened. > > ron > > Do you have a bundle of these emails? In what way is it hard to edit? Are the web pages just plain html not generated from some other source format? Would you like patches to the html or just a new version of the document? From dkotian3 at vsnl.net Thu Apr 17 13:47:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 17 13:47:00 2003 Subject: Random comments on LinuxBIOS References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: <017f01c3050d$c52732a0$5c2141db@vsnl.net> I think, the paper at this link is also good http://www.lysator.liu.se/upplysning/fa/linuxbios.pdf A newbie would like to read this first .. May like to add it or update more on this and put it somewhere in linuxbios site. Regards Deepak ----- Original Message ----- From: "McMechan, James W CIV" To: Sent: Thursday, April 17, 2003 3:11 AM Subject: Random comments on LinuxBIOS > Here is what I wrote while trying to catch up to list, though there is a FAQ I produced this while working my way through the emails perhaps some of it could be added to the FAQ > > LinuxBIOS is a project to replace the original BIOS on motherboards with open source. > It appears that the two most common uses for LinuxBIOS are for Supercomputer clusters and embedded systems. > > LinuxBIOS no longer always uses a real OS to load the OS. > > LinuxBIOS currently can boot several payloads including: > 1 Linux which is what started all this, hence the name. > 2 Plan9 Ron Minnich seems to like this one. > 3 VxWorks Felix Radensky said he had success with a older boot method > 4 EtherBoot which despite its name can also boot from IDE hard disks with a patch or the dev branch > 5 ADLO is a shim to allow the Boch emulator's BIOS to be used on top of LinuxBIOS to boot some versions of that other OS or to run VGA controller ROMs which seem to assume that a BIOS is present. > ? > > The Supercomputer people appear to want remote control, remote booting and fast boot times without user interaction. For example "Press F2" is a very inconvenient task on a system with 100 computers > > The embedded people appear to want full control of the boot process and very fast boot times. For example who wants to watch a memory test go by on their VCR/MP3 player or see "Press F2" without a keyboard. > > LinuxBIOS is installed into the flash ROM in place of the original BIOS. > > Most flash ROMs are currently 2 Megabits in size, i.e. a 29SF020 which is only 256 Kilobytes of storage, or some might be 4 Mbit ROM which provides 512 Kbytes and so could allow for very squeezed down Linux kernel to also be included in the ROM, though I don't know of a useful kernel that small. > It would be nice to add a larger flash ROM but many motherboards do not connect enough address lines to the Flash ROM socket, for even the 4 Mbit part or they solder the chip down. > > Another popular if grumbled about option is the disk on a chip the DoC Millennium which is a paged flash part that can have LinuxBIOS in the first part and up to 8 Mbytes of other data on other pages of the chip. Very short ~3-10 seconds boot times should be possible with this method. > > The BIOS Flash can usually be programmed in place by either just reflashing the original BIOS or by the hotswap method also known as the dental floss method, referring to putting dental floss under a PLCC chip to make it easy to remove without requiring metal tools in a powered on system. The hotswap method is not for the faint of heart but not all that difficult either, the system is booted with a working ROM which is then pulled out and the ROM to be programmed inserted in its place, then the ROM flashing program is run, and when done a reboot can be tried, this does not require a programmer as it uses the motherboards ability to update its BIOS as the programmer, if the reboot fails the original ROM is reinserted and the system rebooted again, all this requires, is some spare flash ROMs it is a good idea to make and label a couple backups incase something goes wrong, I have put ROMs in DIP sockets backwards destroying the ROMs, and ESD is a worry even if it is not a commo! > n problem. > > Some people also use a very nice device called the BIOS savior which allows for two BIOS ROMS to be switched by the user with a switch. > > Compact Flash IDE adaptors appear to be popular as the CF does not need the spin up time of a real hard disk. Very short boot times of ~3-10 seconds are claimed for this configuration also. > > Each model of motherboard currently needs its own build information due to a lack of standards in how the circuitboards are laid out between models and manufactures. > > If you have a motherboard that is not listed, a "lspci" will show what chipsets are in use and you can then check to see if they are in the LinuxBIOS tree, if not some research may be needed to get the documentation for that chipset. > > The kexec patches date back to the original LinuxBIOS with the Linux kernel in the BIOS ROM and allowed for new kernels to be loaded and executed without rebooting back to the BIOS this would allow for a minimal kernel in the BIOS and a full featured kernel to be loaded by the first kernel. Also the BIOS ROMs have a limited programming life and limited space so a small unchanging kernel that then loads the final kernel from something else prolongs the useful life of the ROMs. > > Ron Minnich at LANL uses similar feature called "two kernel monte" http://sourceforge.net/projects/monte/ > > LinuxBIOS now has the ability to make use of the CMOS space for fallback/normal boot image selection, serial port baud rates, and boot order for Etherboot, Ron Minnich also used it authentication information for a Plan9 cluster. See the cmos.layout files for information. > > Steven James had nice comments on Normal/Fallback images: > LinuxBIOS for that board (and many others) supports two copies of > LinuxBIOS, fallback and primary. The fallback image is intended to be old > and stable and it's purpose is to act as a sort of rescue image. The > primary image is the one that gets updated to the latest and greatest, and > is meant to be more versatile and complete. > > The fallback image is loaded at the top of the flash, and always receives > control at power on or reset. It does very minimal setup (switch to 32bit > flat memory protected mode), checks and clears the CMOS boot bit, and if > it was set, jumps to the primary image. This way, if the primary fails, > hitting reset will boot from the fallback image alone. > > From your config, you were building a primary image. the missing bytes are > for the fallback image to be appended. It didn't work because you didn't > have a fallback image at the top of the flash (the end of the image file). > > Ron Minnich expanded on fallback vs. primary: > > You need to build TWO linuxbioses, a fallback and a primary. The fallback > is 64k. The primary is (romsize-64k). > > To build a romimage, you build the fallback, build the primary, then: > > cat primary/romimage fallback/romimage > final_romimage > > flash_rom final_romimage > > Steve Gehlbach wrote about using BOOT_IDE: > See pcchips787.config in util/config for complete configuration. > > option BOOT_IDE=1 > This enables booting from IDE, the file to use is linux.bin.gz: > > option IDE_BOOT_DRIVE=2 > If you do not use drive 0 (default), then you can set which drive to > boot; (0,1,2,3) are the four standard PC drives: > > option ONE_TRACK=32 > The linux.bin.gz file is put in raw form at partition 1, ie, the first > partition on the disk. This is located just past the partition table. > The partition table size varies, it is "one track" from the beginning of > the disk. "one track" in c/h/s notation is "s" or the number of sectors > per track. ONE_TRACK is in sectors, the software multiplies by 512. > Most disks are 63 sectors per track (the default), but my CF is 32 > sectors per track. eg, the partion table is 63x512 or 32x512 bytes. > > You can partition your disk as you want, but Linux goes raw in partition > 1; just make sure partition 1 is big enough, not a problem on today's > disks. You could put the Linux root file system on partition 2, for > example. In pcchips787.config, I put the Linux root file system on IDE > 0, partition 2 (I was experimenting with Linux in partition 1), but I > eventually put Linux on drive 2 using CF. You are right, copying of > linux.bin.gz raw to the partition is dangerous, and something like "cat > linux.bin.gz > /dev/hda1" will definitely screw the disk if you put the > wrong disk or partition. I recommend a shell script, fingers cannot be > trusted. You can also use "dd" but "cat" works. > > Greg Watsons had some good comments on on the boot process and file layout: > There seem to be two main parts to linuxbios. The first is > arch/{arch}/config/ctr0.base which does the very low level > initialization, like turning on memory, etc. The second is > arch/{arch}/lib/c_start.S which does whatever else is necessary to > call the C function hardwaremain(). hardwaremain() then does whatever > else is necessary to load Linux. > > c_start.S is linked with linuxbios.a, a library containing generic > support routines (those found in the lib directory) and anything > specified using the 'object' directive in a Config file (and other > stuff). The resultant 'executable' is called linuxbios_c. The loader > script used to link linuxbios_c is config/linuxbios_c.ld, and is > configured to be loaded relative to _RAMBASE. > > crt0.base is not linked against anything. Any additional assembly > routines you need must be specified using the 'mainboardinit' > directive in a Config file. This causes the specified assembly file > to be added to "crt0_includes.h" which is in turn included at the > start of crt0.base (or at the end in the case of the ppc version). > The loader script used to link crt0.base is in > arch/{arch}/config/ldscript.base. The resultant 'executable' is > called linuxbios and will be loaded at _ROMBASE. The tricky thing is > that this loader script will also load the linuxbios_c 'executable' > at a location called _payload in this file. The main task of > crt0.base is then to initialize enough hardware so that this payload > can be copied from ROM into ram (which may also involve uncompressing > code). Then control is transferred to _start, which is the first > location in linuxbios_c. > > To get an idea of how crt0.base works, look at the following files. > This is the order of execution specified by the configuration file > for sis735. > > cpu/i386/entry16.inc > cpu/i386/entry32.inc > superio/sis/950/setup_serial.inc > pc80/serial.inc > arch/i386/lib/console.inc > cpu/k7/earlymtrr.inc > northsouthbridge/sis/735/raminit.inc > arch/i386/config/crt0.base > > Next look at c_start.S which will show you what happens once control > is transferred to _start. Finally, look at > arch/{arch}/lib/hardwaremain.c to see what other stuff is done to get > Linux loaded. > > Most other files are specific to particular hardware, so it can be > pretty confusing to just browse the tree. > > Hope this helps, > > Greg > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From dkotian3 at vsnl.net Thu Apr 17 14:01:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 17 14:01:01 2003 Subject: Random comments on LinuxBIOS References: <02E6075B2450E94CA159E22331C1620473EA20@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: <01b601c3050f$c0d2e1c0$5c2141db@vsnl.net> > Compact Flash IDE adaptors appear to be popular as the CF does not need the spin up time of a real hard disk. Very short boot times of ~3-10 seconds are claimed for this configuration also. > Is there some kind of patch available for the kernel which can be applied to achieve the above point. I am using 2.4.18-3 kernel, please let me know. Regards Deepak From rminnich at lanl.gov Thu Apr 17 14:15:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 14:15:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <02E6075B2450E94CA159E22331C1620473EAD1@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: On Thu, 17 Apr 2003, McMechan, James W CIV wrote: > Do you have a bundle of these emails? I will send you the inbox. You will need to work out the questions and answers from the text, but it is pretty obvious. > In what way is it hard to edit? mbox format. > Are the web pages just plain html not generated from some other source format? hand-edited html. > Would you like patches to the html or just a new version of the document? whatever you prefer. ron From ebiederman at lnxi.com Thu Apr 17 14:47:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 14:47:00 2003 Subject: Stable storage of BIOS parameters... In-Reply-To: References: Message-ID: ron minnich writes: > On 17 Apr 2003, Eric W. Biederman wrote: > > > > > By using jffs2 other people can manipulate the stable store as well. > > My big question is how should I encode the options that LinuxBIOS > > utilizes. > > I really am liking S-expressions for this type of thing nowadays. Trivial > to parse, we have a small fast library for this (see > http://sourceforge.net/project/showfiles.php?group_id=521&release_id=111528) For data format I don't have a real problem with that, but there are circumstances I don't think it is appropriate. For settings that can either be in cmos or in a rom chip I definitely want something simpler, so I can use the same data format both places. But regardless the basic question is does putting a jffs2 filesystem on the flash start to sound like a good idea? And doing things like hard coding filenames or something into the LinuxBIOS code that looks up information. I'm just playing with ideas at this point. And the real question is how to put the information on the flash. Not what should be the content of the individual files. So S-expr is hopefully irrelevant to this bit of conversation. I have yet to be convinced a self describing format does not have unfortunate size overheads. Eric From steve at nexpath.com Thu Apr 17 15:36:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 17 15:36:00 2003 Subject: 1.1 development release targets. In-Reply-To: References: <3E9D8C9A.9050606@nexpath.com> <3E9E28DA.6000400@nexpath.com> Message-ID: <3E9F0B7C.7050803@nexpath.com> > see src/include/boot/linuxbios_tables.h > > Eric Thanks, I'll check it out. -Steve From rminnich at lanl.gov Thu Apr 17 16:05:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 16:05:01 2003 Subject: Stable storage of BIOS parameters... In-Reply-To: Message-ID: On 17 Apr 2003, Eric W. Biederman wrote: > But regardless the basic question is does putting a jffs2 filesystem on > the flash start to sound like a good idea? And doing things like > hard coding filenames or something into the LinuxBIOS code that looks > up information. seems ok at first glance. > I'm just playing with ideas at this point. And the real question is > how to put the information on the flash. Not what should be the content > of the individual files. So S-expr is hopefully irrelevant to this bit > of conversation. I have yet to be convinced a self describing format > does not have unfortunate size overheads. Yes, but: isn't a file system in some sense an attempt to get a self-describing format? ron From rminnich at lanl.gov Thu Apr 17 16:05:19 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 17 16:05:19 2003 Subject: Stable storage of BIOS parameters... In-Reply-To: Message-ID: On 17 Apr 2003, Eric W. Biederman wrote: > I'm just playing with ideas at this point. And the real question is > how to put the information on the flash. Not what should be the content > of the individual files. So S-expr is hopefully irrelevant to this bit > of conversation. I have yet to be convinced a self describing format > does not have unfortunate size overheads. Also, one other comment, note the silliness with the _MP_ versions would probably not exist with a self-describing format. ron From ebiederman at lnxi.com Thu Apr 17 16:17:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 17 16:17:00 2003 Subject: Stable storage of BIOS parameters... In-Reply-To: References: Message-ID: ron minnich writes: > On 17 Apr 2003, Eric W. Biederman wrote: > > > But regardless the basic question is does putting a jffs2 filesystem on > > the flash start to sound like a good idea? And doing things like > > hard coding filenames or something into the LinuxBIOS code that looks > > up information. > > seems ok at first glance. > > > I'm just playing with ideas at this point. And the real question is > > how to put the information on the flash. Not what should be the content > > of the individual files. So S-expr is hopefully irrelevant to this bit > > of conversation. I have yet to be convinced a self describing format > > does not have unfortunate size overheads. > > Yes, but: isn't a file system in some sense an attempt to get a > self-describing format? O.k. Lets try describing this another way: - A filesystem is needed so the data is 1) To handle the strangeness of flash chips. 2) Self describing at least on a course level. For the CMOS I keep the data described in the LinuxBIOS table, with the actual data being stored out of line. In some instances a binary only representation is better because the value can be directly used. So if we are already self describing at the fs level do we need S-exprs? In any event it is important that someplace you define what all of the kinds of data are and what they mean. And beyond that I think size and simplicity are important. Not so important that we paint ourselves into a corner. But important enough that we keep a close eye on what we are doing. Eric From james.mcmechan at navy.mil Thu Apr 17 16:45:01 2003 From: james.mcmechan at navy.mil (McMechan, James W CIV) Date: Thu Apr 17 16:45:01 2003 Subject: Random comments on LinuxBIOS Message-ID: <02E6075B2450E94CA159E22331C1620473EB2D@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> ... > > Compact Flash IDE adaptors appear to be popular as the CF > does not need > the spin up time of a real hard disk. Very short boot times > of ~3-10 seconds > are claimed for this configuration also. > > > > Is there some kind of patch available for the kernel which > can be applied to > achieve the above point. > I am using 2.4.18-3 kernel, please let me know. > > Regards > Deepak > I am not quite clear on what you are asking. The Compact Flash in a CF->IDE adaptor should look exactly like a very very fast hard disk that does not require spinup time. Using any one of the IDE boot methods should just plain work. What were you wanting to patch? The boot process after the kernel is loaded and running can be speeded up with better boot scripts "serel" from www.fastboot.org has been mentioned as helping RedHat style startup scripts. From stuge-linuxbios at cdy.org Thu Apr 17 19:20:01 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Thu Apr 17 19:20:01 2003 Subject: Random comments on LinuxBIOS In-Reply-To: References: <02E6075B2450E94CA159E22331C1620473EAD1@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: <20030417234231.GA10065@foo.birdnet.se> On Thu, Apr 17, 2003 at 12:46:01PM -0600, ron minnich wrote: > > In what way is it hard to edit? > mbox format. > > > Are the web pages just plain html not generated from some other source format? > hand-edited html. This is where my wiki suggestion comes in to place. Just some $0.02. :) //Peter From spirit at reactor.ru Fri Apr 18 01:39:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Fri Apr 18 01:39:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <02E6075B2450E94CA159E22331C1620473EB2D@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> References: <02E6075B2450E94CA159E22331C1620473EB2D@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> Message-ID: <138989760159.20030418101046@reactor.ru> Hello James, Friday, April 18, 2003, 1:16:54 AM, you wrote: MJWC> ... >> > Compact Flash IDE adaptors appear to be popular as the CF >> does not need the spin up time of a real hard disk. >> Is there some kind of patch available for the kernel which >> can be applied to >> achieve the above point. MJWC> Using any one of the IDE boot methods should just plain work. MJWC> What were you wanting to patch? I think Deepak wants a patch for disabling the LinuxBIOS hard-coded IDE spinup delay. I issued a patch for that some time ago. You may want to search the archive. With best regards, Alexander mailto:spirit at reactor.ru From ebiederman at lnxi.com Fri Apr 18 02:10:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 02:10:00 2003 Subject: Status report.... Message-ID: I am slowly putting together a working LinuxBIOS development tree. I have romcc up and working and initializing memory on my target hammer board. The memory configuration completely hard coded but it works. I am using gcc's C preprocessor because I need the ability to stringize a macro to print out the LinuxBIOS version string, and I don't feel like implementing macros with arguments and all of the little quirks right now. One of my first steps is making hardwaremain generic and in just starting that and looking at it's dependencies I have found a huge mess of code that needs to be cleaned up. Just a small taste. Why are the I/O primitives in cpu/p5/io.h and not directly in arch/io.h? And why to we have piles of code in that header that are ifdefed out. Most of the code left over from the earliest days of LinuxBIOS is similarly nasty. There is a list that goes on and on, and tomorrow I will get serious about cleaning it out. With a little luck I am maybe a week out from having my hammer code being self hosting. That is a week out from gluing the LinuxBIOS tree back together enough that I can load etherboot and have normal and fallback images. At which point I should have enough bandwidth to get the last of the little things I care about resolved in on the stable branch. I would worry about that more but my primary responsibility at this point is to get the Hammer port up and going. Eric From ebiederman at lnxi.com Fri Apr 18 02:36:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 02:36:00 2003 Subject: Random comments on LinuxBIOS In-Reply-To: <138989760159.20030418101046@reactor.ru> References: <02E6075B2450E94CA159E22331C1620473EB2D@NAWECHLKEX02VA.nadsuswe.nads.navy.mil> <138989760159.20030418101046@reactor.ru> Message-ID: Alexander Amelkin writes: > Hello James, > > Friday, April 18, 2003, 1:16:54 AM, you wrote: > > MJWC> ... > >> > Compact Flash IDE adaptors appear to be popular as the CF > >> does not need the spin up time of a real hard disk. > > >> Is there some kind of patch available for the kernel which > >> can be applied to > >> achieve the above point. > > MJWC> Using any one of the IDE boot methods should just plain work. > MJWC> What were you wanting to patch? > > I think Deepak wants a patch for disabling the LinuxBIOS hard-coded IDE spinup > delay. I issued a patch for that some time ago. You may want to search > the archive. I have not seen that but there is a very clean core in the etherboot driver that you might want to snag. It gets the IDE spinup correct. At least for all devices that implement the BSY bit. So it is less likely to be annoying. Eric From rminnich at lanl.gov Fri Apr 18 09:49:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 09:49:00 2003 Subject: Status report.... In-Reply-To: Message-ID: thanks, eric, once you have that initial thing up I'll try populating my stuff and see how it goes. I'll document what I do so we have an upgrade guide people can use. Folks, if you don't take responsibility for a mainboard, it won't move forward, and nobody has written me to do this yet. We really need some takers. ron From steve at nexpath.com Fri Apr 18 10:46:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 18 10:46:01 2003 Subject: Status report.... In-Reply-To: References: Message-ID: <3EA018F4.9040003@nexpath.com> ron minnich wrote: > Folks, if you don't take responsibility for a mainboard, it won't move > forward, and nobody has written me to do this yet. We really need some > takers. I'm in for the stpc/consumer2 and the pcchips/m787cl+. I hope I end up with a way to boot from IDE without etherboot; I tried USE_ELF_BOOT + BOOT_IDE and all I got was checksum errors. Wonder if anyone has tried that. -Steve From gwatson at lanl.gov Fri Apr 18 11:04:00 2003 From: gwatson at lanl.gov (Greg Watson) Date: Fri Apr 18 11:04:00 2003 Subject: Status report.... In-Reply-To: References: Message-ID: Needless to say, I'm happy to maintain motorola/sandpoint. Greg At 8:19 AM -0600 18/4/03, ron minnich wrote: >thanks, eric, once you have that initial thing up I'll try populating my >stuff and see how it goes. > >I'll document what I do so we have an upgrade guide people can use. > >Folks, if you don't take responsibility for a mainboard, it won't move >forward, and nobody has written me to do this yet. We really need some >takers. > >ron > >_______________________________________________ >Linuxbios mailing list >Linuxbios at clustermatic.org >http://www.clustermatic.org/mailman/listinfo/linuxbios From pyro at linuxlabs.com Fri Apr 18 11:38:00 2003 From: pyro at linuxlabs.com (steven james) Date: Fri Apr 18 11:38:00 2003 Subject: Status report.... In-Reply-To: Message-ID: Greetings, I've got the Clearwaters (Intel) and the S2723 (Tyan Tiger i7501). Also tyan TigerMPX and guiness unless Eric wants guiness back. G'day, sjames On Fri, 18 Apr 2003, Greg Watson wrote: > Needless to say, I'm happy to maintain motorola/sandpoint. > > Greg > > > At 8:19 AM -0600 18/4/03, ron minnich wrote: > >thanks, eric, once you have that initial thing up I'll try populating my > >stuff and see how it goes. > > > >I'll document what I do so we have an upgrade guide people can use. > > > >Folks, if you don't take responsibility for a mainboard, it won't move > >forward, and nobody has written me to do this yet. We really need some > >takers. > > > >ron > > > >_______________________________________________ > >Linuxbios mailing list > >Linuxbios at clustermatic.org > >http://www.clustermatic.org/mailman/listinfo/linuxbios > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Fri Apr 18 11:39:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 11:39:00 2003 Subject: Status report.... In-Reply-To: Message-ID: if you are a committer go ahead and take ownership of your boards via the STATUS file, and then if you can pull down freebios2 (when Eric is done) see what it takes. Any experiences you write up, send to me. ron From dpilon at atipa.com Fri Apr 18 15:05:01 2003 From: dpilon at atipa.com (Denis Pilon) Date: Fri Apr 18 15:05:01 2003 Subject: Error ... Message-ID: <1050694673.1208.21.camel@atipa-dp> I've got linuxbios somewhat working with the SuperMicro X5DPE-G2, but I have discovered one problem. If I run lm_sensors and initialize the chips, the node will not reboot. It gives me a no memory error at boot. Has anyone seen this? How do I start debuggin ? Denis -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From rminnich at lanl.gov Fri Apr 18 15:11:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 15:11:00 2003 Subject: Error ... In-Reply-To: <1050694673.1208.21.camel@atipa-dp> Message-ID: On 18 Apr 2003, Denis Pilon wrote: > If I run lm_sensors and initialize the chips, the node will not reboot. > It gives me a no memory error at boot. consider getting rid of lm_sensors first. > Has anyone seen this? How do I start debuggin ? lm_sensors is messing up the SPD bus for linuxbios. I think we're going to need to fix linuxbios so that it will always undo whatever lm_sensors is doing. See if you can see what lm_sensors is doing to the southbridge to change the I2C bus around. ron From garlick at llnl.gov Fri Apr 18 16:16:01 2003 From: garlick at llnl.gov (Jim Garlick) Date: Fri Apr 18 16:16:01 2003 Subject: Error ... In-Reply-To: <1050694673.1208.21.camel@atipa-dp> Message-ID: Sounds very similar to a problem observed on the SuperMicro P4DPR/DPE which has the same winbond sensor chip. The solution was to tell lm_sensors to access the Winbond via ISA rather than I2C. I think there is also some code in the p4dpr/p4dpe linuxbios port to notice the memory initialization problem and work around it (e.g. wait a while and reset?) The details have faded but you might check the code. Jim On Fri, 18 Apr 2003, Denis Pilon wrote: > I've got linuxbios somewhat working with the SuperMicro X5DPE-G2, but I > have discovered one problem. > > If I run lm_sensors and initialize the chips, the node will not reboot. > It gives me a no memory error at boot. > > Has anyone seen this? How do I start debuggin ? > > Denis > > From rminnich at lanl.gov Fri Apr 18 16:22:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 16:22:01 2003 Subject: Error ... In-Reply-To: Message-ID: On Fri, 18 Apr 2003, Jim Garlick wrote: > Sounds very similar to a problem observed on the SuperMicro P4DPR/DPE which > has the same winbond sensor chip. The solution was to tell lm_sensors to > access the Winbond via ISA rather than I2C. Or do what I did: chuck lm_sensors completely and write a simple little driver using supermon, with much help from Steve James. Way better than that lm_sensors stuff. ron From spirit at reactor.ru Fri Apr 18 16:28:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Fri Apr 18 16:28:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: <501043128599.20030419010014@reactor.ru> Hello ron, Friday, April 18, 2003, 11:42:29 PM, you wrote: rm> On 18 Apr 2003, Denis Pilon wrote: >> If I run lm_sensors and initialize the chips, the node will not reboot. >> It gives me a no memory error at boot. rm> consider getting rid of lm_sensors first. Personally, I'd prefer LinuxBIOS to be compatible with lm_sensors. It's very pity to miss such a nice part of hardware as monitoring. What's the problem with supporting it? With best regards, Alexander mailto:spirit at reactor.ru From marc.miller at amd.com Fri Apr 18 16:32:01 2003 From: marc.miller at amd.com (marc.miller at amd.com) Date: Fri Apr 18 16:32:01 2003 Subject: Error ... Message-ID: <858788618A93D111B45900805F85267A07FCDF98@caexmta3.amd.com> Forgive my ignorance, but can we simply ask the lm_sensors developers to fix lm_sensors? -----Original Message----- From: Alexander Amelkin [mailto:spirit at reactor.ru] Sent: Friday, April 18, 2003 2:00 PM To: Ron Minnich Subject: Re[2]: Error ... Hello ron, Friday, April 18, 2003, 11:42:29 PM, you wrote: rm> On 18 Apr 2003, Denis Pilon wrote: >> If I run lm_sensors and initialize the chips, the node will not reboot. >> It gives me a no memory error at boot. rm> consider getting rid of lm_sensors first. Personally, I'd prefer LinuxBIOS to be compatible with lm_sensors. It's very pity to miss such a nice part of hardware as monitoring. What's the problem with supporting it? With best regards, Alexander mailto:spirit at reactor.ru _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Fri Apr 18 16:33:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 16:33:00 2003 Subject: Error ... In-Reply-To: <501043128599.20030419010014@reactor.ru> Message-ID: On Sat, 19 Apr 2003, Alexander Amelkin wrote: > Personally, I'd prefer LinuxBIOS to be compatible with lm_sensors. > It's very pity to miss such a nice part of hardware as monitoring. > What's the problem with supporting it? monitoring is good. Especially if it is good monitoring. Lm_sensors is not good enough for us. I think the supermon framework for hardware monitors is lots better. linuxbios is compatible with lm_sensors, but on this one platform, lm_sensors is breaking linuxbios somehow. It will get fixed. ron From rminnich at lanl.gov Fri Apr 18 16:35:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 16:35:01 2003 Subject: Error ... In-Reply-To: <858788618A93D111B45900805F85267A07FCDF98@caexmta3.amd.com> Message-ID: On Fri, 18 Apr 2003 marc.miller at amd.com wrote: > Forgive my ignorance, but can we simply ask the lm_sensors developers to > fix lm_sensors? that's a great idea, and we should, but we should also make linuxbios able to handle this kind of thing. We can fix lm_sensors but linuxbios should tolerate arbitrary (well, almost) misbehavior from software. thanks ron From spirit at reactor.ru Fri Apr 18 16:42:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Fri Apr 18 16:42:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: <351043931133.20030419011337@reactor.ru> Hello ron, Saturday, April 19, 2003, 1:04:20 AM, you wrote: rm> On Sat, 19 Apr 2003, Alexander Amelkin wrote: >> Personally, I'd prefer LinuxBIOS to be compatible with lm_sensors. >> It's very pity to miss such a nice part of hardware as monitoring. >> What's the problem with supporting it? rm> monitoring is good. Especially if it is good monitoring. Lm_sensors is not rm> good enough for us. I think the supermon framework for hardware monitors rm> is lots better. Thanks for the hint. I'll try the supermon. rm> linuxbios is compatible with lm_sensors, but on this one platform, rm> lm_sensors is breaking linuxbios somehow. It will get fixed. Well... on sis630 (pcchips 787cl+) lm_sensors don't show anything with linuxbios, while with AMIBIOS they do show lots of things. A little off-topic... I was quite amazed when I found that on EPIA ESP5000 there is a monitoring chip detected by lm_sensors, but no actual sensors present. :) Anyone knows if that is really so, or is that just a bug of lm_sensors? With best regards, Alexander mailto:spirit at reactor.ru From dpilon at atipa.com Fri Apr 18 16:43:00 2003 From: dpilon at atipa.com (Denis Pilon) Date: Fri Apr 18 16:43:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: <1050700550.1206.32.camel@atipa-dp> Thanks Jim. Removing the i2c prevented the problem. But I will try to see where lm_sensors is messing up the bus. Denis On Fri, 2003-04-18 at 15:48, Jim Garlick wrote: > Sounds very similar to a problem observed on the SuperMicro P4DPR/DPE which > has the same winbond sensor chip. The solution was to tell lm_sensors to > access the Winbond via ISA rather than I2C. > > I think there is also some code in the p4dpr/p4dpe linuxbios port to notice > the memory initialization problem and work around it (e.g. wait a while > and reset?) The details have faded but you might check the code. > > Jim > > On Fri, 18 Apr 2003, Denis Pilon wrote: > > > I've got linuxbios somewhat working with the SuperMicro X5DPE-G2, but I > > have discovered one problem. > > > > If I run lm_sensors and initialize the chips, the node will not reboot. > > It gives me a no memory error at boot. > > > > Has anyone seen this? How do I start debuggin ? > > > > Denis > > > > -- =================================================== Denis E. Pilon Systems Integration Specialist Atipa Technologies http://www.atipa.com/ dpilon at atipa.com (785)841-9513 x3111 =================================================== -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part URL: From ebiederman at lnxi.com Fri Apr 18 17:21:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 17:21:01 2003 Subject: Error ... In-Reply-To: References: Message-ID: Jim Garlick writes: > Sounds very similar to a problem observed on the SuperMicro P4DPR/DPE which > has the same winbond sensor chip. The solution was to tell lm_sensors to > access the Winbond via ISA rather than I2C. > > I think there is also some code in the p4dpr/p4dpe linuxbios port to notice > the memory initialization problem and work around it (e.g. wait a while > and reset?) The details have faded but you might check the code. So the details. On Intels ICH3 southbridge there is an smbus/i2c controller that when it sees invalid on the bus it locks up. But it only sees invalid data when you are actually using the bus. Using lm_sensors just increases the probability that the smbus controller will lock up. At which point a reboot will fail because SPD information cannot be read over the smbus, leading the LinuxBIOS to think you have no RAM. To the best of my knowledge there is no fix. This is just something that needs to be avoided. Eric From pyro at linuxlabs.com Fri Apr 18 18:06:01 2003 From: pyro at linuxlabs.com (steven james) Date: Fri Apr 18 18:06:01 2003 Subject: Error ... In-Reply-To: Message-ID: Greetings, If you're using the linuxbios_reset (kernel linuxbios patch), you can have reset write 0x0e to 0xcf9. That will do a 3 second poweroff instead of a simple reset. That should avoid the issue at least. G'day, sjames On 18 Apr 2003, Eric W. Biederman wrote: > Jim Garlick writes: > > > Sounds very similar to a problem observed on the SuperMicro P4DPR/DPE which > > has the same winbond sensor chip. The solution was to tell lm_sensors to > > access the Winbond via ISA rather than I2C. > > > > I think there is also some code in the p4dpr/p4dpe linuxbios port to notice > > the memory initialization problem and work around it (e.g. wait a while > > and reset?) The details have faded but you might check the code. > > So the details. On Intels ICH3 southbridge there is an smbus/i2c controller > that when it sees invalid on the bus it locks up. But it only sees invalid > data when you are actually using the bus. Using lm_sensors just increases > the probability that the smbus controller will lock up. > > At which point a reboot will fail because SPD information cannot be > read over the smbus, leading the LinuxBIOS to think you have no RAM. > > To the best of my knowledge there is no fix. This is just something > that needs to be avoided. > > Eric > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From ebiederman at lnxi.com Fri Apr 18 19:31:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 19:31:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: steven james writes: > Greetings, > > If you're using the linuxbios_reset (kernel linuxbios patch), you can have > reset write 0x0e to 0xcf9. That will do a 3 second poweroff instead of a > simple reset. That should avoid the issue at least. A) lm_sensors would still not be reliable. B) It would quite possibly undo the fix for the PCIH2 that causes the boards to lock up while scanning the PCI bus. And without a cluster like MCR to test on I don't feel comfortable doing that. Eric From ebiederman at lnxi.com Fri Apr 18 19:34:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 19:34:01 2003 Subject: Error ... In-Reply-To: References: Message-ID: ron minnich writes: > On Fri, 18 Apr 2003 marc.miller at amd.com wrote: > > > Forgive my ignorance, but can we simply ask the lm_sensors developers to > > fix lm_sensors? > > that's a great idea, and we should, but we should also make linuxbios able > to handle this kind of thing. We can fix lm_sensors but linuxbios should > tolerate arbitrary (well, almost) misbehavior from software. Currently LinuxBIOS walks a very fine line between 3 hardware bugs. I think it is quite possibly working as well as the hardware allows. Unless someone knows of a magic software fix to remove all noise from an i2c line. Eric From ebiederman at lnxi.com Fri Apr 18 19:37:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 19:37:00 2003 Subject: Status report.... In-Reply-To: <3EA018F4.9040003@nexpath.com> References: <3EA018F4.9040003@nexpath.com> Message-ID: Steve Gehlbach writes: > ron minnich wrote: > > > Folks, if you don't take responsibility for a mainboard, it won't move > > forward, and nobody has written me to do this yet. We really need some takers. > > > I'm in for the stpc/consumer2 and the pcchips/m787cl+. I hope I end up with a > way to boot from IDE without etherboot; I tried USE_ELF_BOOT + BOOT_IDE and all > I got was checksum errors. Wonder if anyone has tried that. Hmm. This increases my feeling that the IDE driver needs to be backported from etherboot. Though without an error log I can't really guess what was going wrong. Eric From ebiederman at lnxi.com Fri Apr 18 19:42:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 18 19:42:01 2003 Subject: Status report.... In-Reply-To: References: Message-ID: ron minnich writes: > if you are a committer go ahead and take ownership of your boards via the > STATUS file, and then if you can pull down freebios2 (when Eric is done) > see what it takes. > > Any experiences you write up, send to me. Ron I guess put me down for something like: Tyan s2466 Supermicro p4dpr, p4dpe, p4dpeg2, x5dpr (not yet working) I don't think I have the oomph to go through the stable tree even for such a simple thing as create the STATUS files. Eric From pyro at linuxlabs.com Fri Apr 18 20:19:00 2003 From: pyro at linuxlabs.com (steven james) Date: Fri Apr 18 20:19:00 2003 Subject: Error ... In-Reply-To: Message-ID: Greetings, I've been doing that for some time now. I haven't had a great problem, though the sample size is smaller than MCR. How is the OS triggering a power off followed by a cold start going to screw up PCIH2? G'day, sjames On 18 Apr 2003, Eric W. Biederman wrote: > steven james writes: > > > Greetings, > > > > If you're using the linuxbios_reset (kernel linuxbios patch), you can have > > reset write 0x0e to 0xcf9. That will do a 3 second poweroff instead of a > > simple reset. That should avoid the issue at least. > > A) lm_sensors would still not be reliable. > B) It would quite possibly undo the fix for the PCIH2 that causes the > boards to lock up while scanning the PCI bus. > And without a cluster like MCR to test on I don't feel comfortable doing that. > > Eric > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Fri Apr 18 22:04:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 22:04:00 2003 Subject: Error ... In-Reply-To: <351043931133.20030419011337@reactor.ru> Message-ID: On Sat, 19 Apr 2003, Alexander Amelkin wrote: > > Well... on sis630 (pcchips 787cl+) lm_sensors don't show anything with > linuxbios, while with AMIBIOS they do show lots of things. that's interesting. we need to figure this one out. > A little off-topic... I was quite amazed when I found that on EPIA > ESP5000 there is a monitoring chip detected by lm_sensors, but no actual > sensors present. :) Anyone knows if that is really so, or is that just > a bug of lm_sensors? > we've seen our share of bugs with lm_sensors. ron From chris.schafer at pnl.gov Fri Apr 18 22:09:01 2003 From: chris.schafer at pnl.gov (Schafer, Christopher L) Date: Fri Apr 18 22:09:01 2003 Subject: Support for supermicro 7501 based boards and issues with 3ware raid cards Message-ID: Unfortunatally it is the only way to build raid arrays on the controllers. You can recover arrays from the os but not build them. I will contact 3ware and see what they think... Thanks -----Original Message----- From: steven james [mailto:pyro at linuxlabs.com] Sent: Wed 3/26/2003 3:20 PM To: Schafer, Christopher L Cc: linuxbios at clustermatic.org Subject: Re: Support for supermicro 7501 based boards and issues with 3ware raid cards Greetings, The 7501 chipset is supported, but the setup will probably need some tweaking for the Supermicro. Currently, it's for Intel Clearwater (se7501sw2). The 2Ware BIOS tools will likely not work under LinuxBIOS. Are they actually necessary, or just a convieniance (I've not worked with it)? G'day, sjames On Fri, 14 Mar 2003, Schafer, Christopher L wrote: > Linux BIOS Team, > > I was wondering if there is any active work on the above Supermicro > boards and are there issues with accessing the 3ware raid controllers > bios tools if one is using linux BIOS? > > Thanks, > > Chris Schafer > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From chris.schafer at pnl.gov Fri Apr 18 22:09:05 2003 From: chris.schafer at pnl.gov (Schafer, Christopher L) Date: Fri Apr 18 22:09:05 2003 Subject: Support for supermicro 7501 based boards and issues with 3ware raid cards Message-ID: The device is hardware raid upto 12 drives. Presents a SCSI device to the driver/kernel. From David.Barr at student.oc.edu Fri Apr 18 22:09:09 2003 From: David.Barr at student.oc.edu (David Barr) Date: Fri Apr 18 22:09:09 2003 Subject: Intel l440lx Message-ID: Greetings, I notice the Intel l440bx and l440gx boards are or have been supported. I have a 440lx, so I was wondering if this would be my big chance to do a port. Here's my query: 1) How might I determine if one's close enough for a n00b like me? 2) Which should I start on (grab the closer of gx||bx, start kludging)? 3) Where would I find out enough about the lx to figure the changes? If it's an obvious answer, just let me dangle on my own rope for a few days, and I'm sure I'll get impatient and start hacking blindly. But a little guidance couldn't hurt, either. :D Thanks, -dBarr. From davidmurdie at yahoo.co.uk Fri Apr 18 22:09:13 2003 From: davidmurdie at yahoo.co.uk (=?iso-8859-1?q?David=20Murdie?=) Date: Fri Apr 18 22:09:13 2003 Subject: EPOX 8KTA3+ Pro Message-ID: <20030402123002.53448.qmail@web40304.mail.yahoo.com> Anybody tried to get any of the EPoX boards to go? Thinking in particular of the 8KTA3+ Pro (http://www.epox.org/products/mainboards/8kta3pluspro/) If you look at the picture, you'll notice that it has a built in debug card. (2 seven segment displays - a few of the other EPoX boards appear to have them) -- David Murdie __________________________________________________ Yahoo! Plus For a better Internet experience http://www.yahoo.co.uk/btoffer From kewalbk at hotmail.com Fri Apr 18 22:09:18 2003 From: kewalbk at hotmail.com (kewal karavinkoppa) Date: Fri Apr 18 22:09:18 2003 Subject: help for LinuxBIOS!!!!! Message-ID: hi friends, i am an undergraduate student, studing computer science, from INDIA. i along with my friends are doing a project on LinuxBIOS, so we need some help with this reguard... currently we are facing the problem to chose the motherboard to select.. so can u please help... Thank you....... _________________________________________________________________ Catch the Oscar fever. See winners & losers. http://server1.msn.co.in/MSNSpecials/oscar2003/index.asp Right here From feldbauerr at daisydata.com Fri Apr 18 22:09:22 2003 From: feldbauerr at daisydata.com (Robert Feldbauer) Date: Fri Apr 18 22:09:22 2003 Subject: Aaeon GENE-6310 support? Message-ID: <200304031606.30140.feldbauerr@daisydata.com> Wondering if anyone has worked on support for Aaeon's GENE-6310? I tried using the form on the site, but got an error - "The requested URL /cgi-bin/linuxbios/help.cgi was not found on this server." lspci output for the GENE-6310: 00:00.0 Host bridge: VIA Technologies, Inc. VT8605 [ProSavage PM133] 00:01.0 PCI bridge: VIA Technologies, Inc. VT8605 [PM133 AGP] 00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super South] (rev 40) 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586/B/686A/B PIPC Bus Master IDE (rev 06) 00:07.2 USB Controller: VIA Technologies, Inc. USB (rev 1a) 00:07.3 USB Controller: VIA Technologies, Inc. USB (rev 1a) 00:07.4 Bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI] (rev 40) 00:07.5 Multimedia audio controller: VIA Technologies, Inc. VT82C686 AC97 Audio Controller (rev 50) 00:0d.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 01:00.0 VGA compatible controller: S3 Inc. 86C380 [ProSavageDDR K4M266] (rev 02) Bob Feldbauer feldbauerr at daisydata.com From seandembrosky at hotmail.com Fri Apr 18 22:09:26 2003 From: seandembrosky at hotmail.com (Sean Dembrosky) Date: Fri Apr 18 22:09:26 2003 Subject: Via Epia 5000 research. Message-ID: (perhaps replies/forwards can be cc: to seandembrosky at hotmail.com ? thanks!) I have only recently begun researching Linux and BIOS modifications, and your group is doing exactly what I would love to be involved with and researching myself, but my interests are on the other end of the spectrum - flex/mini ATX formfactors for specific, small uses (router, DVD player, MP3 player, etc., all loading from a modded BIOS that loads the OS via TFTP/NFS, all diskless) I believe some flavor of your LinuxBIOS could be paired wonderfully with the via EPIA line of mainboards, specifically with the 5000 product... http://www.viavpsd.com/product/epia_mini_itx_spec.jsp?motherboardId=21 Do you have plans to research implementing your genius on this mainboard? If so, how can I help! If not, do you have any suggestions or ideas of where I should go in order to research my own experimentations? And good luck with this amazing idea, I feel quite confident that you will have tremendous success with this, and you deserve it for all the hard work it must have taken! _________________________________________________________________ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus From rminnich at lanl.gov Fri Apr 18 22:16:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 18 22:16:01 2003 Subject: Status report.... In-Reply-To: Message-ID: On 18 Apr 2003, Eric W. Biederman wrote: > Hmm. This increases my feeling that the IDE driver needs to be backported > from etherboot. absolutely. ron From steve at nexpath.com Fri Apr 18 23:48:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 18 23:48:00 2003 Subject: Status report.... In-Reply-To: References: Message-ID: <3EA0D25A.7090505@nexpath.com> ron minnich wrote: > On 18 Apr 2003, Eric W. Biederman wrote: > >>Hmm. This increases my feeling that the IDE driver needs to be backported >>from etherboot. > > absolutely. > > ron > Maybe, but I have used the IDE driver plenty on non-elf images without problems. I could have made the elf image wrong (not sure I have the latest mkelfImage). This is the first time I have used it so it might be pilot error. I'll get a debug output and post it. -Steve From hansolofalcon at worldnet.att.net Sat Apr 19 01:15:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sat Apr 19 01:15:01 2003 Subject: Using LinuxBIOS to boot other kernels Message-ID: <001701c30637$3d2e0560$239efea9@who5> Hello again from Gregg C Levine Regarding the OSkit kernels. Has anyone attempted to use an appropriate Linux BIOS image to boot an OSkit kernel? I know Eric looked into that kernel project a while ago, but has anyone else looked into it? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) From vincent at ulyssis.org Sat Apr 19 05:33:00 2003 From: vincent at ulyssis.org (Vincent Touquet) Date: Sat Apr 19 05:33:00 2003 Subject: Support for supermicro 7501 based boards and issues with 3ware raid cards In-Reply-To: References: Message-ID: <20030419100527.GA10130@lea.ulyssis.org> Note that they also have a web based tool called 3DM, in which you can do everything you want to (the same and more options as in the BIOS). So you could do all your RAID setup on your nodes in early user space, so after you loaded the kernel ? Its probably possible to do it before you load the kernel. I'm quite optimistic that you could get some support from 3ware on this front as they are in general very good citizens :) best regards, Vincent On Wed, Mar 26, 2003 at 08:31:09PM -0800, Schafer, Christopher L wrote: >Unfortunatally it is the only way to build raid arrays on the controllers. You can recover arrays from the os but not build them. > >I will contact 3ware and see what they think... > >Thanks > >-----Original Message----- >From: steven james [mailto:pyro at linuxlabs.com] >Sent: Wed 3/26/2003 3:20 PM >To: Schafer, Christopher L >Cc: linuxbios at clustermatic.org >Subject: Re: Support for supermicro 7501 based boards and issues with 3ware raid cards > >Greetings, > >The 7501 chipset is supported, but the setup will probably need some >tweaking for the Supermicro. Currently, it's for Intel Clearwater >(se7501sw2). > >The 2Ware BIOS tools will likely not work under LinuxBIOS. Are they >actually necessary, or just a convieniance (I've not worked with it)? > >G'day, >sjames > > > >On Fri, 14 Mar 2003, Schafer, Christopher L wrote: > >> Linux BIOS Team, >> >> I was wondering if there is any active work on the above Supermicro >> boards and are there issues with accessing the 3ware raid controllers >> bios tools if one is using linux BIOS? >> >> Thanks, >> >> Chris Schafer >> _______________________________________________ >> Linuxbios mailing list >> Linuxbios at clustermatic.org >> http://www.clustermatic.org/mailman/listinfo/linuxbios >> > >-- >-------------------------steven james, director of research, linux labs >... ........ ..... .... 230 peachtree st nw ste 2701 >the original linux labs atlanta.ga.us 30303 > -since 1995 http://www.linuxlabs.com > office 404.577.7747 fax 404.577.7743 >----------------------------------------------------------------------- > > > > > >_______________________________________________ >Linuxbios mailing list >Linuxbios at clustermatic.org >http://www.clustermatic.org/mailman/listinfo/linuxbios From pyro at linuxlabs.com Sat Apr 19 07:48:00 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 19 07:48:00 2003 Subject: Error ... In-Reply-To: Message-ID: Greetings, The sensors in the sis950 (superio) have to be turned on. Apparently, the AMIBIOS does that, so lm_sensors doesn't try to initialize it. The standalone supermon driver I wrote for that does the initialization itself. G'day, sjames On Fri, 18 Apr 2003, ron minnich wrote: > On Sat, 19 Apr 2003, Alexander Amelkin wrote: > > > > > Well... on sis630 (pcchips 787cl+) lm_sensors don't show anything with > > linuxbios, while with AMIBIOS they do show lots of things. > > that's interesting. we need to figure this one out. > > > > A little off-topic... I was quite amazed when I found that on EPIA > > ESP5000 there is a monitoring chip detected by lm_sensors, but no actual > > sensors present. :) Anyone knows if that is really so, or is that just > > a bug of lm_sensors? > > > > we've seen our share of bugs with lm_sensors. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From justin at street-vision.com Sat Apr 19 09:20:01 2003 From: justin at street-vision.com (Justin Cormack) Date: Sat Apr 19 09:20:01 2003 Subject: Support for supermicro 7501 based boards and issues with 3ware raid cards In-Reply-To: <20030419100527.GA10130@lea.ulyssis.org> References: <20030419100527.GA10130@lea.ulyssis.org> Message-ID: <1050760323.1652.28.camel@lotte> On Sat, 2003-04-19 at 11:05, Vincent Touquet wrote: > Note that they also have a web based tool called 3DM, > in which you can do everything you want to (the same > and more options as in the BIOS). > > So you could do all your RAID setup on your nodes > in early user space, so after you loaded the kernel ? > > Its probably possible to do it before you load the > kernel. I'm quite optimistic that you could get > some support from 3ware on this front as they > are in general very good citizens :) No, you cannot create arrays from the web tool. And there is no source to 3dm. I dont think you will find them that cooperative either. Try running the BIOS under wmulation I guess. Justin From hcyun at etri.re.kr Sat Apr 19 10:35:01 2003 From: hcyun at etri.re.kr (hcyun at etri.re.kr) Date: Sat Apr 19 10:35:01 2003 Subject: serial debug patch for ADLO Message-ID: <8470181DABD5D511B3E700D0B7A8AC4A9CD721@cms3> Hello, I made a patch which make all bochs debug message can be seen from serial terminal. I think it might be useful if you want to apply ADLO to your own target. you can apply this patch againt util/ADLO/bochs/bios/rombios.c by # patch -p0 < rombios-serial.patch Regards, -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: rombios-serial.patch Type: application/octet-stream Size: 1552 bytes Desc: not available URL: From adam at cfar.umd.edu Sat Apr 19 10:55:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sat Apr 19 10:55:00 2003 Subject: serial debug patch for ADLO In-Reply-To: <8470181DABD5D511B3E700D0B7A8AC4A9CD721@cms3> Message-ID: <20030419112230.L89507-100000@www.missl.cs.umd.edu> to clarify. He's been working on getting ADLO running on EPIA-M. He's been pretty succesfull so far. Using serial console he's got both lilo and linux to boot using ADLO, so that you can use them pretty much "out of box" as long as you don't need video. However from what I heard you can get video once boot is done and you run some video program (TinyX Xvesa here). As far as video goes, on cold-reboot he's getting : Then, screen is blinkering with various color, but I can't see any text. As far as video goes, on warm-reboot (after booting with original PC BIOS) he's getting : If I first booted from original bios and reset with linuxbios + adlo, I can see text messages of original bios remained. ideas on how to attack the issue? Overall I would say that anyone who want to try get ADLO to work should start with serial console and leave VIDEO to the end given difficulty debugging binary-only module. Another good idea is to try warm-boot from PC BIOS into ADLO at beginning of development/work on ADLO. It leaves you with somewhat functioning video-bios (or fully-functioning as it was for me). On Sun, 20 Apr 2003 hcyun at etri.re.kr wrote: > Hello, > > I made a patch which make all bochs debug message can be seen from serial > terminal. > I think it might be useful if you want to apply ADLO to your own target. > > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > # patch -p0 < rombios-serial.patch > > Regards, > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From hansolofalcon at worldnet.att.net Sat Apr 19 11:28:01 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Sat Apr 19 11:28:01 2003 Subject: serial debug patch for ADLO In-Reply-To: <20030419112230.L89507-100000@www.missl.cs.umd.edu> Message-ID: <000801c3068c$ebb372a0$fbc3580c@who5> Hello from Gregg C Levine So I gathered from his earlier posts on this subject, and about that board specifically. But Adam, isn't ADLO board agnostic? By that I mean, that it does not need to know what motherboard it is being used on, unlike basic Linux BIOS. But, I am indeed watching this with interest. ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of Adam Sulmicki > Sent: Saturday, April 19, 2003 11:32 AM > To: hcyun at etri.re.kr > Cc: linuxbios at clustermatic.org > Subject: Re: serial debug patch for ADLO > > > to clarify. He's been working on getting ADLO running on EPIA-M. > > He's been pretty succesfull so far. > > Using serial console he's got both lilo and linux to boot using ADLO, so > that you can use them pretty much "out of box" as long as you don't need > video. However from what I heard you can get video once boot is done and > you run some video program (TinyX Xvesa here). > > As far as video goes, on cold-reboot he's getting : > > Then, screen is blinkering with various color, but I can't see any > text. > > As far as video goes, on warm-reboot (after booting with original PC BIOS) > he's getting : > > If I first booted from original bios and reset with linuxbios + > adlo, I can see text messages of original bios remained. > > ideas on how to attack the issue? > > Overall I would say that anyone who want to try get ADLO to work should > start with serial console and leave VIDEO to the end given difficulty > debugging binary-only module. > > Another good idea is to try warm-boot from PC BIOS into ADLO at beginning > of development/work on ADLO. It leaves you with somewhat functioning > video-bios (or fully-functioning as it was for me). > > On Sun, 20 Apr 2003 hcyun at etri.re.kr wrote: > > > Hello, > > > > I made a patch which make all bochs debug message can be seen from serial > > terminal. > > I think it might be useful if you want to apply ADLO to your own target. > > > > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > > # patch -p0 < rombios-serial.patch > > > > Regards, > > > > > > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From adam at cfar.umd.edu Sat Apr 19 11:39:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sat Apr 19 11:39:00 2003 Subject: serial debug patch for ADLO In-Reply-To: <000801c3068c$ebb372a0$fbc3580c@who5> Message-ID: <20030419121101.A89507-100000@www.missl.cs.umd.edu> mostly yes, this is why he was able to get serial boot in just several hours of work. the major quirks were that he needed to code up the patch for serial output for bochs to see what's going on. then reconfigure lilo for serial output. Then to the same with linux. another issue was that while VIA EPIA-M has 128mb ram, upper 32 mb are revered for video. thus E820 map had to be modifed so that it shows 96mb of ram instead of 128mb to linux. i think that's about it. On Sat, 19 Apr 2003, Gregg C Levine wrote: > Hello from Gregg C Levine > So I gathered from his earlier posts on this subject, and about that > board specifically. But Adam, isn't ADLO board agnostic? By that I > mean, that it does not need to know what motherboard it is being used > on, unlike basic Linux BIOS. But, I am indeed watching this with > interest. > ------------------- > Gregg C Levine hansolofalcon at worldnet.att.net > ------------------------------------------------------------ > "The Force will be with you...Always." Obi-Wan Kenobi > "Use the Force, Luke."? Obi-Wan Kenobi > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > (This company dedicates this E-Mail to Master Yoda ) > > > > > -----Original Message----- > > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > > admin at clustermatic.org] On Behalf Of Adam Sulmicki > > Sent: Saturday, April 19, 2003 11:32 AM > > To: hcyun at etri.re.kr > > Cc: linuxbios at clustermatic.org > > Subject: Re: serial debug patch for ADLO > > > > > > to clarify. He's been working on getting ADLO running on EPIA-M. > > > > He's been pretty succesfull so far. > > > > Using serial console he's got both lilo and linux to boot using > ADLO, so > > that you can use them pretty much "out of box" as long as you don't > need > > video. However from what I heard you can get video once boot is done > and > > you run some video program (TinyX Xvesa here). > > > > As far as video goes, on cold-reboot he's getting : > > > > Then, screen is blinkering with various color, but I can't see > any > > text. > > > > As far as video goes, on warm-reboot (after booting with original PC > BIOS) > > he's getting : > > > > If I first booted from original bios and reset with linuxbios > + > > adlo, I can see text messages of original bios remained. > > > > ideas on how to attack the issue? > > > > Overall I would say that anyone who want to try get ADLO to work > should > > start with serial console and leave VIDEO to the end given > difficulty > > debugging binary-only module. > > > > Another good idea is to try warm-boot from PC BIOS into ADLO at > beginning > > of development/work on ADLO. It leaves you with somewhat functioning > > video-bios (or fully-functioning as it was for me). > > > > On Sun, 20 Apr 2003 hcyun at etri.re.kr wrote: > > > > > Hello, > > > > > > I made a patch which make all bochs debug message can be seen from > serial > > > terminal. > > > I think it might be useful if you want to apply ADLO to your own > target. > > > > > > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > > > # patch -p0 < rombios-serial.patch > > > > > > Regards, > > > > > > > > > > > > -- > > Adam Sulmicki > > http://www.eax.com The Supreme Headquarters of the 32 bit > registers > > > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From spirit at reactor.ru Sat Apr 19 12:50:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Sat Apr 19 12:50:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: <61116427147.20030419212153@reactor.ru> Hello steven, Saturday, April 19, 2003, 4:20:32 PM, you wrote: sj> Greetings, sj> The sensors in the sis950 (superio) have to be turned on. Apparently, the sj> AMIBIOS does that, so lm_sensors doesn't try to initialize it. The sj> standalone supermon driver I wrote for that does the initialization sj> itself. Supermon is described as 'high-speed cluster monitoring' thing. What if I don't run any clusters? Just a standalone embedded device? The word 'cluster' scares me a little. :) Is supermon small enough to fit in my 16Mb Linux configuration instead of lm_sensors? I think I don't need anything 'high-speed'. I just want to check temperature, fan and voltage parameters from time to time. With best regards, Alexander mailto:spirit at reactor.ru From pyro at linuxlabs.com Sat Apr 19 13:57:01 2003 From: pyro at linuxlabs.com (steven james) Date: Sat Apr 19 13:57:01 2003 Subject: Error ... In-Reply-To: <61116427147.20030419212153@reactor.ru> Message-ID: Greetings, Supermon was developed for clustering, but really is fairly general purpose (I'm looking into using it on a variety of servers). It consists of a few parts. Supermon itself and mon are daemons. mon gathers data from the kernel and passes it up to supermon. supermon simply aggregates the data and serves it to various clients. The other two parts are supermon_proc and a sensor module. The sensor module just presents the sensor data in /proc/sys/supermon_sensors in the form of an S expression (a Lisp like expression, easily parsable and human readable as well). G'day, sjames On Sat, 19 Apr 2003, Alexander Amelkin wrote: > Hello steven, > > Saturday, April 19, 2003, 4:20:32 PM, you wrote: > > sj> Greetings, > > sj> The sensors in the sis950 (superio) have to be turned on. Apparently, the > sj> AMIBIOS does that, so lm_sensors doesn't try to initialize it. The > sj> standalone supermon driver I wrote for that does the initialization > sj> itself. > > Supermon is described as 'high-speed cluster monitoring' thing. What > if I don't run any clusters? Just a standalone embedded device? The > word 'cluster' scares me a little. :) Is supermon small enough to fit > in my 16Mb Linux configuration instead of lm_sensors? I think I don't > need anything 'high-speed'. I just want to check temperature, fan and > voltage parameters from time to time. > > With best regards, > Alexander mailto:spirit at reactor.ru > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Sat Apr 19 15:35:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sat Apr 19 15:35:01 2003 Subject: Error ... In-Reply-To: <61116427147.20030419212153@reactor.ru> Message-ID: supermon drivers are smaller and far less complex than lm_sensors drivers, and they run lots faster. ron From bendany at mistdl.com Sun Apr 20 02:40:01 2003 From: bendany at mistdl.com (bendany) Date: Sun Apr 20 02:40:01 2003 Subject: serial debug patch for ADLO References: <20030419121101.A89507-100000@www.missl.cs.umd.edu> Message-ID: <002001c3078a$3e1990d0$2a00a8c0@ben> hi,all. I have got the serial patch, and test it. but still have problem. my mainboard is k7sem ver3.0c here is some output message: --------------------------------------------------- Searching for server (DHCP)... ...Me: 192.168.0.40, Server: 192.168.0.38, Gateway 192.168.0.250 Loading 192.168.0.38:/var/tftpboot/kernel ...(ELF)... ............................................................................ ................done rombios.c,v 1.1 2002/11/25 02:07:53 rminnich Exp $ FATAL: Keyboard RESET error:31 FATAL: Keyboard RESET error:994 FATAL: Keyboard RESET error:995 FATAL: Keyboard RESET error:996 *** int 15h function AX=F401, BX=0005 not yet supported! Bochs BIOS, 1 cpu, $Revision: 1.1 $ $Date: 2002/11/25 02:07:53 $ [BOCHS BIOS VER:1.79] [COMPILE DATE:Apr 20 2003 TIME:15:00:19] DEVICE:0 ata0 master: PCHS=16383/16/63 translation=lba LCHS=1024/255/63 ata0 master: IBM-DTLA-307030 ATA-5 Hard-Disk (29314 MBytes) PCI BIOS function 0x01 --------------------------------------------------- and the screen can show some strange character, sometimes are part of original BIOS message. I think the bochs BIOS has run. any tips? thanks! ----- Original Message ----- From: "Adam Sulmicki" To: "Gregg C Levine" Cc: "Linuxbios" Sent: Saturday, April 19, 2003 9:15 AM Subject: RE: serial debug patch for ADLO > > mostly yes, this is why he was able to get serial boot in just several > hours of work. > > the major quirks were that he needed to code up the patch for serial > output for bochs to see what's going on. then reconfigure lilo for serial > output. Then to the same with linux. > > another issue was that while VIA EPIA-M has 128mb ram, upper 32 mb are > revered for video. thus E820 map had to be modifed so that it shows 96mb > of ram instead of 128mb to linux. > > i think that's about it. > > On Sat, 19 Apr 2003, Gregg C Levine wrote: > > > Hello from Gregg C Levine > > So I gathered from his earlier posts on this subject, and about that > > board specifically. But Adam, isn't ADLO board agnostic? By that I > > mean, that it does not need to know what motherboard it is being used > > on, unlike basic Linux BIOS. But, I am indeed watching this with > > interest. > > ------------------- > > Gregg C Levine hansolofalcon at worldnet.att.net > > ------------------------------------------------------------ > > "The Force will be with you...Always." Obi-Wan Kenobi > > "Use the Force, Luke."?Obi-Wan Kenobi > > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > > (This company dedicates this E-Mail to Master Yoda ) > > > > > > > > > -----Original Message----- > > > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > > > admin at clustermatic.org] On Behalf Of Adam Sulmicki > > > Sent: Saturday, April 19, 2003 11:32 AM > > > To: hcyun at etri.re.kr > > > Cc: linuxbios at clustermatic.org > > > Subject: Re: serial debug patch for ADLO > > > > > > > > > to clarify. He's been working on getting ADLO running on EPIA-M. > > > > > > He's been pretty succesfull so far. > > > > > > Using serial console he's got both lilo and linux to boot using > > ADLO, so > > > that you can use them pretty much "out of box" as long as you don't > > need > > > video. However from what I heard you can get video once boot is done > > and > > > you run some video program (TinyX Xvesa here). > > > > > > As far as video goes, on cold-reboot he's getting : > > > > > > Then, screen is blinkering with various color, but I can't see > > any > > > text. > > > > > > As far as video goes, on warm-reboot (after booting with original PC > > BIOS) > > > he's getting : > > > > > > If I first booted from original bios and reset with linuxbios > > + > > > adlo, I can see text messages of original bios remained. > > > > > > ideas on how to attack the issue? > > > > > > Overall I would say that anyone who want to try get ADLO to work > > should > > > start with serial console and leave VIDEO to the end given > > difficulty > > > debugging binary-only module. > > > > > > Another good idea is to try warm-boot from PC BIOS into ADLO at > > beginning > > > of development/work on ADLO. It leaves you with somewhat functioning > > > video-bios (or fully-functioning as it was for me). > > > > > > On Sun, 20 Apr 2003 hcyun at etri.re.kr wrote: > > > > > > > Hello, > > > > > > > > I made a patch which make all bochs debug message can be seen from > > serial > > > > terminal. > > > > I think it might be useful if you want to apply ADLO to your own > > target. > > > > > > > > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > > > > # patch -p0 < rombios-serial.patch > > > > > > > > Regards, > > > > > > > > > > > > > > > > > -- > > > Adam Sulmicki > > > http://www.eax.com The Supreme Headquarters of the 32 bit > > registers > > > > > > > > > > > > _______________________________________________ > > > Linuxbios mailing list > > > Linuxbios at clustermatic.org > > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From spirit at reactor.ru Sun Apr 20 03:06:00 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Sun Apr 20 03:06:00 2003 Subject: Error ... In-Reply-To: References: Message-ID: <1771167831212.20030420113837@reactor.ru> Hello ron, Sunday, April 20, 2003, 12:06:57 AM, you wrote: rm> supermon drivers are smaller and far less complex than lm_sensors drivers, rm> and they run lots faster. Right, but I found that they just won't create anything in the /proc filesystem of my router with a 2.4.20 kernel. Symdm won't create anything as well. :( I think, though, that I must move with this to the supermon list. Or just stick to lm_sensors, which work w/o problems for me. With best regards, Alexander mailto:spirit at reactor.ru From adam at cfar.umd.edu Sun Apr 20 09:09:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sun Apr 20 09:09:00 2003 Subject: serial debug patch for ADLO In-Reply-To: <002001c3078a$3e1990d0$2a00a8c0@ben> Message-ID: <20030420093518.S238-100000@www.missl.cs.umd.edu> hmm that does not look right. in general bochs bios should have tried to: - initalize vga module - try to boot from floppy (and fail) - try to load MBR from /dev/hda I think it did init vga bios as it is probably the first thing it does (before the bochs banner which we can see below). Though u might want to double check on this. I would expect to see the "floppy boot fail" (or similar) message instead of that "PCI BIOS function" which seems to be the last message here. as far as vga bios goes, I would try doing warm boot from pc bios and see what happen. as far as bochs bios itself goes, u will have to try enable some debugging information and see where exactly it stop. in particular I would try to enable interupt 13, since it might actually be reading MBR (that's the only place that I can think of hand that would probe for PCI bios). On Sun, 20 Apr 2003, bendany wrote: > hi,all. > I have got the serial patch, and test it. but still have problem. > my mainboard is k7sem ver3.0c > here is some output message: > --------------------------------------------------- > Searching for server (DHCP)... > ...Me: 192.168.0.40, Server: 192.168.0.38, Gateway 192.168.0.250 > Loading 192.168.0.38:/var/tftpboot/kernel ...(ELF)... > ............................................................................ > ................done > rombios.c,v 1.1 2002/11/25 02:07:53 rminnich Exp $ > FATAL: Keyboard RESET error:31 > FATAL: Keyboard RESET error:994 > FATAL: Keyboard RESET error:995 > FATAL: Keyboard RESET error:996 > *** int 15h function AX=F401, BX=0005 not yet supported! > Bochs BIOS, 1 cpu, $Revision: 1.1 $ $Date: 2002/11/25 02:07:53 $ > [BOCHS BIOS VER:1.79] > [COMPILE DATE:Apr 20 2003 TIME:15:00:19] > > DEVICE:0 > ata0 master: PCHS=16383/16/63 translation=lba LCHS=1024/255/63 > ata0 master: IBM-DTLA-307030 ATA-5 Hard-Disk (29314 MBytes) > > PCI BIOS function 0x01 > --------------------------------------------------- > and the screen can show some strange character, sometimes are part of > original BIOS message. > I think the bochs BIOS has run. > any tips? thanks! > > ----- Original Message ----- > From: "Adam Sulmicki" > To: "Gregg C Levine" > Cc: "Linuxbios" > Sent: Saturday, April 19, 2003 9:15 AM > Subject: RE: serial debug patch for ADLO > > > > > > mostly yes, this is why he was able to get serial boot in just several > > hours of work. > > > > the major quirks were that he needed to code up the patch for serial > > output for bochs to see what's going on. then reconfigure lilo for serial > > output. Then to the same with linux. > > > > another issue was that while VIA EPIA-M has 128mb ram, upper 32 mb are > > revered for video. thus E820 map had to be modifed so that it shows 96mb > > of ram instead of 128mb to linux. > > > > i think that's about it. > > > > On Sat, 19 Apr 2003, Gregg C Levine wrote: > > > > > Hello from Gregg C Levine > > > So I gathered from his earlier posts on this subject, and about that > > > board specifically. But Adam, isn't ADLO board agnostic? By that I > > > mean, that it does not need to know what motherboard it is being used > > > on, unlike basic Linux BIOS. But, I am indeed watching this with > > > interest. > > > ------------------- > > > Gregg C Levine hansolofalcon at worldnet.att.net > > > ------------------------------------------------------------ > > > "The Force will be with you...Always." Obi-Wan Kenobi > > > "Use the Force, Luke."?Obi-Wan Kenobi > > > (This company dedicates this E-Mail to General Obi-Wan Kenobi ) > > > (This company dedicates this E-Mail to Master Yoda ) > > > > > > > > > > > > > -----Original Message----- > > > > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > > > > admin at clustermatic.org] On Behalf Of Adam Sulmicki > > > > Sent: Saturday, April 19, 2003 11:32 AM > > > > To: hcyun at etri.re.kr > > > > Cc: linuxbios at clustermatic.org > > > > Subject: Re: serial debug patch for ADLO > > > > > > > > > > > > to clarify. He's been working on getting ADLO running on EPIA-M. > > > > > > > > He's been pretty succesfull so far. > > > > > > > > Using serial console he's got both lilo and linux to boot using > > > ADLO, so > > > > that you can use them pretty much "out of box" as long as you don't > > > need > > > > video. However from what I heard you can get video once boot is done > > > and > > > > you run some video program (TinyX Xvesa here). > > > > > > > > As far as video goes, on cold-reboot he's getting : > > > > > > > > Then, screen is blinkering with various color, but I can't see > > > any > > > > text. > > > > > > > > As far as video goes, on warm-reboot (after booting with original PC > > > BIOS) > > > > he's getting : > > > > > > > > If I first booted from original bios and reset with linuxbios > > > + > > > > adlo, I can see text messages of original bios remained. > > > > > > > > ideas on how to attack the issue? > > > > > > > > Overall I would say that anyone who want to try get ADLO to work > > > should > > > > start with serial console and leave VIDEO to the end given > > > difficulty > > > > debugging binary-only module. > > > > > > > > Another good idea is to try warm-boot from PC BIOS into ADLO at > > > beginning > > > > of development/work on ADLO. It leaves you with somewhat functioning > > > > video-bios (or fully-functioning as it was for me). > > > > > > > > On Sun, 20 Apr 2003 hcyun at etri.re.kr wrote: > > > > > > > > > Hello, > > > > > > > > > > I made a patch which make all bochs debug message can be seen from > > > serial > > > > > terminal. > > > > > I think it might be useful if you want to apply ADLO to your own > > > target. > > > > > > > > > > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > > > > > # patch -p0 < rombios-serial.patch > > > > > > > > > > Regards, > > > > > > > > > > > > > > > > > > > > > > -- > > > > Adam Sulmicki > > > > http://www.eax.com The Supreme Headquarters of the 32 bit > > > registers > > > > > > > > > > > > > > > > _______________________________________________ > > > > Linuxbios mailing list > > > > Linuxbios at clustermatic.org > > > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > > > > > > > > -- > > Adam Sulmicki > > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rminnich at lanl.gov Sun Apr 20 19:08:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sun Apr 20 19:08:01 2003 Subject: Error ... In-Reply-To: <1771167831212.20030420113837@reactor.ru> Message-ID: On Sun, 20 Apr 2003, Alexander Amelkin wrote: > Right, but I found that they just won't create anything in the /proc > filesystem of my router with a 2.4.20 kernel. Symdm won't create > anything as well. :( I think, though, that I must move with this to > the supermon list. Or just stick to lm_sensors, which work w/o > problems for me. ask the supermon list, although we have not seen this problem. Or stick with lm_sensors :-) use what works. ron From rminnich at lanl.gov Mon Apr 21 00:57:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 00:57:00 2003 Subject: mainboard status on the web page Message-ID: I have updated it from the latest CVS. Will owners please take a look and tell me what's missing. I know I have to get some of you who don't do commits in there and I will get to that this week. I'm finally back on the job of keeping the web page up to date after we moved it so things should be better now. ron From jmrus at interactivity.es Mon Apr 21 02:04:00 2003 From: jmrus at interactivity.es (Jose M. Rus) Date: Mon Apr 21 02:04:00 2003 Subject: Booting other OS from Linux Message-ID: <3EA391BA.4080903@interactivity.es> Hi, I'm working on a project that could be viewed as an extended boot loader. Lilo and Grub are fine boot loaders, but a Linux application can do a lot more things, like restoring image partitions for example. For this I need to shutdown Linux cleanly and start another OS (partition) instantly without the BIOS ever knowing it, but I'm not looking to replace the BIOS code like you do, I left to the user the choice of using her motherboard BIOS code of your own Linuxbios code before the first boot. "kexec" and "Two kernel Monte" can load another linux kernels, I need to boot Linux, windows and others OSes from Linux. Maybe you have already done what I'm looking for or you have worked on part of it, so I like to hear your advice. Regards, Jose M. Rus From dkotian3 at vsnl.net Mon Apr 21 09:27:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Mon Apr 21 09:27:01 2003 Subject: entry16.inc code doubt Message-ID: <20030421143258.48CD24FF89@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From dkotian3 at vsnl.net Mon Apr 21 09:47:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Mon Apr 21 09:47:01 2003 Subject: Assembly instruction reference for i386 Message-ID: <20030421145329.6D9E74FDFC@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From adam at cfar.umd.edu Mon Apr 21 09:51:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Mon Apr 21 09:51:00 2003 Subject: Assembly instruction reference for i386 In-Reply-To: <20030421145329.6D9E74FDFC@bom6.vsnl.net.in> Message-ID: <20030421102846.I2515-100000@www.missl.cs.umd.edu> > Can some give a good assembly instruction reference for i386 > architechture which can be used to follow the code in LinuxBIOS. the the last 3 entries at this url: http://www.intel.com/design/pentiumii/manuals/ -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From dkotian3 at vsnl.net Mon Apr 21 11:22:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 11:22:00 2003 Subject: Assembly instruction reference for i386 References: <20030421102846.I2515-100000@www.missl.cs.umd.edu> Message-ID: <001301c3081e$338fa980$2b3e41db@vsnl.net> Hi, Thnaks. Would this one hold OK for VIA C3 processor with Intel chipset as well. Please let me know. Regards Deepak ----- Original Message ----- From: "Adam Sulmicki" To: Cc: Sent: Monday, April 21, 2003 7:59 PM Subject: Re: Assembly instruction reference for i386 > > Can some give a good assembly instruction reference for i386 > > architechture which can be used to follow the code in LinuxBIOS. > > the the last 3 entries at this url: > > http://www.intel.com/design/pentiumii/manuals/ > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > From rminnich at lanl.gov Mon Apr 21 11:32:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 11:32:01 2003 Subject: entry16.inc code doubt In-Reply-To: <20030421143258.48CD24FF89@bom6.vsnl.net.in> Message-ID: On Mon, 21 Apr 2003 dkotian3 at vsnl.net wrote: > ***Extract*** > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > orl $0x60000001, %eax /* CD, NW, PE = 1 */ you really need to get a pengium manual. Then it will all make sense. ron From rminnich at lanl.gov Mon Apr 21 11:33:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 11:33:00 2003 Subject: Assembly instruction reference for i386 In-Reply-To: <20030421145329.6D9E74FDFC@bom6.vsnl.net.in> Message-ID: On Mon, 21 Apr 2003 dkotian3 at vsnl.net wrote: > Can some give a good assembly instruction reference for i386 > architechture which can be used to follow the code in LinuxBIOS. just use google, there are tons of good manuals online. ron From rminnich at lanl.gov Mon Apr 21 12:02:01 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 12:02:01 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken Message-ID: I'm working on it but ... linuxbios_c.o: In function `displayinit': linuxbios_c.o(.text+0x3d6): undefined reference to `get_option' linuxbios_c.o: In function `remove_logical_cpus': linuxbios_c.o(.text+0xc57): undefined reference to `get_option' linuxbios_c.o: In function `sizeram': linuxbios_c.o(.text+0x20ae): undefined reference to `get_option' linuxbios_c.o: In function `i786_cpufixup': linuxbios_c.o(.text+0x2442): undefined reference to `get_option' linuxbios_c.o: In function `mainboard_fixup': linuxbios_c.o(.text+0x25f6): undefined reference to `get_option' linuxbios_c.o(.text+0x2621): more undefined references to `get_option' follow linuxbios_c.o: In function `ich3_rtc_init': linuxbios_c.o(.text+0x5cb8): undefined reference to `rtc_init' collect2: ld returned 1 exit status make: *** [linuxbios_c] Error 1 I'm looking. ron From rminnich at lanl.gov Mon Apr 21 12:16:01 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 12:16:01 2003 Subject: the p4dpe-g2 build problem Message-ID: get_option is defined for the moto clock chip. That is not included in the p4dpe-g2 build any more. I'll try to get something going however. ron From pyro at linuxlabs.com Mon Apr 21 12:23:00 2003 From: pyro at linuxlabs.com (steven james) Date: Mon Apr 21 12:23:00 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: Greetings, You need: option CONFIG_MC146818RTC=1 to fix that. G'day, sjames On Mon, 21 Apr 2003, Ronald G. Minnich wrote: > > I'm working on it but ... > > linuxbios_c.o: In function `displayinit': > linuxbios_c.o(.text+0x3d6): undefined reference to `get_option' > linuxbios_c.o: In function `remove_logical_cpus': > linuxbios_c.o(.text+0xc57): undefined reference to `get_option' > linuxbios_c.o: In function `sizeram': > linuxbios_c.o(.text+0x20ae): undefined reference to `get_option' > linuxbios_c.o: In function `i786_cpufixup': > linuxbios_c.o(.text+0x2442): undefined reference to `get_option' > linuxbios_c.o: In function `mainboard_fixup': > linuxbios_c.o(.text+0x25f6): undefined reference to `get_option' > linuxbios_c.o(.text+0x2621): more undefined references to `get_option' > follow > linuxbios_c.o: In function `ich3_rtc_init': > linuxbios_c.o(.text+0x5cb8): undefined reference to `rtc_init' > collect2: ld returned 1 exit status > make: *** [linuxbios_c] Error 1 > > > I'm looking. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From rminnich at lanl.gov Mon Apr 21 12:28:59 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 12:28:59 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: On Mon, 21 Apr 2003, steven james wrote: > option CONFIG_MC146818RTC=1 > to fix that. yeah, that did it. But what changed that broke the build, which was working until now? ron From adam at cfar.umd.edu Mon Apr 21 12:44:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Mon Apr 21 12:44:01 2003 Subject: Assembly instruction reference for i386 In-Reply-To: <001301c3081e$338fa980$2b3e41db@vsnl.net> Message-ID: <20030421131758.S2515-100000@www.missl.cs.umd.edu> VIA C3 actually works with intel chipsets? Though It did work only with VIA Apollo CLE266 On Mon, 21 Apr 2003, Deepak Kotian wrote: > Hi, > > Thnaks. > Would this one hold OK for VIA C3 processor with Intel chipset as well. > Please let me know. > > Regards > Deepak > ----- Original Message ----- > From: "Adam Sulmicki" > To: > Cc: > Sent: Monday, April 21, 2003 7:59 PM > Subject: Re: Assembly instruction reference for i386 > > > > > Can some give a good assembly instruction reference for i386 > > > architechture which can be used to follow the code in LinuxBIOS. > > > > the the last 3 entries at this url: > > > > http://www.intel.com/design/pentiumii/manuals/ > > > > -- > > Adam Sulmicki > > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > > > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From gwatson at lanl.gov Mon Apr 21 12:50:00 2003 From: gwatson at lanl.gov (Greg Watson) Date: Mon Apr 21 12:50:00 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: Probably removing i386isms for the PPC port. Sorry. Greg At 11:00 AM -0600 21/4/03, Ronald G. Minnich wrote: >On Mon, 21 Apr 2003, steven james wrote: > >> option CONFIG_MC146818RTC=1 >> to fix that. > >yeah, that did it. But what changed that broke the build, which was >working until now? > >ron > >_______________________________________________ >Linuxbios mailing list >Linuxbios at clustermatic.org >http://www.clustermatic.org/mailman/listinfo/linuxbios From ebiederman at lnxi.com Mon Apr 21 12:56:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 12:56:01 2003 Subject: Status report.... In-Reply-To: <3EA0D25A.7090505@nexpath.com> References: <3EA0D25A.7090505@nexpath.com> Message-ID: Steve Gehlbach writes: > ron minnich wrote: > > On 18 Apr 2003, Eric W. Biederman wrote: > >>Hmm. This increases my feeling that the IDE driver needs to be backported > >>from etherboot. > > absolutely. ron > > > > Maybe, but I have used the IDE driver plenty on non-elf images without problems. > > I could have made the elf image wrong (not sure I have the latest mkelfImage). > This is the first time I have used it so it might be pilot error. > > I'll get a debug output and post it. Both the ide and floppy loaders were developed using the ELF bootloader. So if they don't work now something has seriously regressed. Eric From ebiederman at lnxi.com Mon Apr 21 13:02:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 13:02:00 2003 Subject: romcc compiles to p5, right? In-Reply-To: References: Message-ID: ron minnich writes: > also, should .c files for romcc really be .rc, or not? make rules might be > easier. Don't know, just wondering. .c There only needs to be one make rule as everything romcc compiles needs to be compiled all at once. So it is one file with a bunch of includes. Additionally some of the code I hope to allow to be compiled either way. Eric From rminnich at lanl.gov Mon Apr 21 13:07:43 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 13:07:43 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: On Mon, 21 Apr 2003, Greg Watson wrote: > Probably removing i386isms for the PPC port. Sorry. can you undo this? let's talk about how. ron From dkotian3 at vsnl.net Mon Apr 21 13:14:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 13:14:01 2003 Subject: entry16.inc code doubt References: Message-ID: <001601c3082d$bd5942c0$2f3e41db@vsnl.net> > you really need to get a pengium manual. Where do I get this one. Is it some Linux Man Page. Could you please give me the link where I can look for it. Thanks and Regards Deepak ----- Original Message ----- From: "ron minnich" To: Cc: Sent: Monday, April 21, 2003 9:34 PM Subject: Re: entry16.inc code doubt > On Mon, 21 Apr 2003 dkotian3 at vsnl.net wrote: > > > ***Extract*** > > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > > orl $0x60000001, %eax /* CD, NW, PE = 1 */ > > you really need to get a pengium manual. > > > Then it will all make sense. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From dkotian3 at vsnl.net Mon Apr 21 13:20:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 13:20:00 2003 Subject: Assembly instruction reference for i386 References: <20030421131758.S2515-100000@www.missl.cs.umd.edu> Message-ID: <002901c3082d$ffeedaa0$2f3e41db@vsnl.net> Thanks, will the same site http://www.intel.com/design/pentiumii/manuals/ hold good for VIA C3 processor as well. Please let me know. ----- Original Message ----- From: "Adam Sulmicki" To: "Deepak Kotian" Cc: Sent: Monday, April 21, 2003 10:51 PM Subject: Re: Assembly instruction reference for i386 > > VIA C3 actually works with intel chipsets? Though It did work only with > VIA Apollo CLE266 > > On Mon, 21 Apr 2003, Deepak Kotian wrote: > > > Hi, > > > > Thnaks. > > Would this one hold OK for VIA C3 processor with Intel chipset as well. > > Please let me know. > > > > Regards > > Deepak > > ----- Original Message ----- > > From: "Adam Sulmicki" > > To: > > Cc: > > Sent: Monday, April 21, 2003 7:59 PM > > Subject: Re: Assembly instruction reference for i386 > > > > > > > > Can some give a good assembly instruction reference for i386 > > > > architechture which can be used to follow the code in LinuxBIOS. > > > > > > the the last 3 entries at this url: > > > > > > http://www.intel.com/design/pentiumii/manuals/ > > > > > > -- > > > Adam Sulmicki > > > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > > > > > > > > > > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Mon Apr 21 13:26:45 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 13:26:45 2003 Subject: entry16.inc code doubt In-Reply-To: <001601c3082d$bd5942c0$2f3e41db@vsnl.net> Message-ID: On Mon, 21 Apr 2003, Deepak Kotian wrote: > > you really need to get a pengium manual. pentium sorry ron From adam at cfar.umd.edu Mon Apr 21 13:33:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Mon Apr 21 13:33:00 2003 Subject: Assembly instruction reference for i386 In-Reply-To: <001301c3081e$338fa980$2b3e41db@vsnl.net> Message-ID: <20030421140621.J2515-100000@www.missl.cs.umd.edu> > Would this one hold OK for VIA C3 processor umm.. yeah, mostly. most non-intel x86 manufactuers assume you use this particular reference as primary reference with their own smaller reference to see how's it different from the "standard" cpu. so get this three data scheets, then google for via c3 datascheet and use it as supplementary reference to the intel's manuals. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rminnich at lanl.gov Mon Apr 21 13:38:45 2003 From: rminnich at lanl.gov (Ronald G. Minnich) Date: Mon Apr 21 13:38:45 2003 Subject: IDE back port Message-ID: I'm going to have a go at back-porting the IDE driver from etherboot to linuxbios, turns out I need this. ron From ebiederman at lnxi.com Mon Apr 21 13:44:28 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 13:44:28 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: "Ronald G. Minnich" writes: > On Mon, 21 Apr 2003, Greg Watson wrote: > > > Probably removing i386isms for the PPC port. Sorry. > > can you undo this? let's talk about how. A couple of thoughts the option can be enabled by default, to the x86 build. And the fact that get_option is so tightly tied to that particular piece of hardware may be a bug. My question is why was the option needed? That general piece of code was initially taken from the alpha kernel tree, so it is for more than just x86, and the code should compile on multiple platforms. And if nothing referenced the code it should not have been linked in. So I don't see where the previous state of affairs should be a problem. Eric From dkotian3 at vsnl.net Mon Apr 21 13:51:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 13:51:00 2003 Subject: entry16.inc code doubt References: Message-ID: <003e01c3082f$aba04b80$2f3e41db@vsnl.net> OK, would this hold true for VIA C3 processor, or I need to change something on it for it. Basically, would entry16, entry32 would change for VIA C3. I have intel chipset with this processor. On the linuxbios tree, I just find a c3 folder with just invd instruction in one of the files. Is this enough Please let me know. Thanks and Regards Deepak ----- Original Message ----- From: "Ronald G. Minnich" To: "Deepak Kotian" Cc: Sent: Monday, April 21, 2003 11:23 PM Subject: Re: entry16.inc code doubt > On Mon, 21 Apr 2003, Deepak Kotian wrote: > > > > you really need to get a pengium manual. > pentium > > sorry > > ron > > > From dkotian3 at vsnl.net Mon Apr 21 13:59:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 13:59:00 2003 Subject: About 82810e north bridge Message-ID: <014c01c30834$307680a0$2f3e41db@vsnl.net> Hi, There seems to be two devices in Intel 82810e north bridge, one is Memory controller Hub and other is Graphical controller HUB. Is it OK to just do the initialization for Memory controller HUB and forget about Graphical controller HUB. Also, I saw 82815ep on linuxbios tree, which has support for MCH only and no AGP. This is very much similar to 82810e with some address changed. What would happen if I support MCH for 82810e , will LinuxBIOS still come up. Could someone please give an opinon on this. Thanks and Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Mon Apr 21 16:53:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 16:53:00 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: On 21 Apr 2003, Eric W. Biederman wrote: > A couple of thoughts the option can be enabled by default, > to the x86 build. I was wondering if we shouldn't enable it for all builds, as this hardware is so common, and individual mainboards can turn it off. The other option is a quick patch to almost all mainboard files to enable it. Comments? ron From rminnich at lanl.gov Mon Apr 21 16:56:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 16:56:00 2003 Subject: About 82810e north bridge In-Reply-To: <014c01c30834$307680a0$2f3e41db@vsnl.net> Message-ID: On Tue, 22 Apr 2003, Deepak Kotian wrote: > There seems to be two devices in Intel 82810e north bridge, one is > Memory controller > Hub and other is Graphical controller HUB. > Is it OK to just do the initialization for Memory controller HUB > and forget about Graphical controller HUB. I think so. > > Also, I saw 82815ep on linuxbios tree, which has support for > MCH only and no AGP. This is very much similar to 82810e with > some address changed. I don't think this hardware is at all well tested, so if you can fix it, we would be grateful. ron From gwatson at lanl.gov Mon Apr 21 17:09:01 2003 From: gwatson at lanl.gov (Greg Watson) Date: Mon Apr 21 17:09:01 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: I prefer Eric's approach of only enabling it for x86 builds, since it is really x86 specific. Greg At 3:24 PM -0600 21/4/03, ron minnich wrote: >On 21 Apr 2003, Eric W. Biederman wrote: > >> A couple of thoughts the option can be enabled by default, >> to the x86 build. > >I was wondering if we shouldn't enable it for all builds, as this hardware >is so common, and individual mainboards can turn it off. The other option >is a quick patch to almost all mainboard files to enable it. > >Comments? > >ron From rsmith at bitworks.com Mon Apr 21 17:14:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 21 17:14:01 2003 Subject: SDRAM row page size wierdness In-Reply-To: <003e01c3082f$aba04b80$2f3e41db@vsnl.net> References: <003e01c3082f$aba04b80$2f3e41db@vsnl.net> Message-ID: <3EA4669B.9090101@bitworks.com> I've finally got LinuxBIOS up and booted to on our Intel 440BX board. I know the 440BX code was listed as working but after going through a lot of the raminit.inc code I don't see how it would have ever worked. Frirst the DRB register was not getting set properly for bank 1. It always ended up as 00. The original code set the PGPOL, DRB and RPS registers all in the same loop and was very confusing. Looking through some of the code for the other northbridges I saw that these functions were spilt up and much easier to follow. So I wholesale replaced that multi-purpose loop with 3 seperate routines that I scrounged from the 440gx and Erik's general purpose spd code. Interesting to note that the return value from the 440bx smbus_read_byte is exactly opposite from all the other smbus_read_byte implementations. That fixed the DRB problem. The next problem has me very confused. The RPS code sets the RPS for a page size of 4kb. A few debug statements in the RPS routine show that 4kb is indeed the correct value according to whats in the SPD. However, It dosen't work. Doing some compareing with a COTS bios shows that register set up for a page size of 8kb. Hardcode that value in after all the spd configure code is done and up it comes. Any ideas on why I have to feed it a bogus page size to work? Its not registered memory. I'm trying to gather up a few other sticks and see if it only with this brand of memory or in general. -- Richard A. Smith rsmith at bitworks.com From ebiederman at lnxi.com Mon Apr 21 17:19:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 17:19:00 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: Greg Watson writes: > I prefer Eric's approach of only enabling it for x86 builds, since it is really > x86 specific. Greg where did you run into problems with that code? I just want to understand. Because that is code like the serial code. It started on x86 but has been spotted on other platforms. Eric From rminnich at lanl.gov Mon Apr 21 17:20:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 17:20:01 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: On Mon, 21 Apr 2003, Greg Watson wrote: > I prefer Eric's approach of only enabling it for x86 builds, since it > is really x86 specific. but it's not. It was on the alpha. I think it is mainboard specific; should we enable in the mainboard file? ron From rminnich at lanl.gov Mon Apr 21 17:21:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 17:21:00 2003 Subject: SDRAM row page size wierdness In-Reply-To: <3EA4669B.9090101@bitworks.com> Message-ID: Is it registered dram? Then the rules change just a bit. ron From ebiederman at lnxi.com Mon Apr 21 17:29:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 17:29:01 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: ron minnich writes: > On Mon, 21 Apr 2003, Greg Watson wrote: > > > I prefer Eric's approach of only enabling it for x86 builds, since it > > is really x86 specific. > > but it's not. It was on the alpha. > > I think it is mainboard specific; should we enable in the mainboard file? But the code is in a library so unless it is referenced it should compile but not be used. Which is why it really confused me that greg had problems with it. Eric From rminnich at lanl.gov Mon Apr 21 17:30:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 17:30:00 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: Message-ID: On 21 Apr 2003, Eric W. Biederman wrote: > But the code is in a library so unless it is referenced it should > compile but not be used. > > Which is why it really confused me that greg had problems with it. good point. Greg, was the issue that the code would not compile or you just did not want it in there on PPC? ron From rsmith at bitworks.com Mon Apr 21 17:34:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 21 17:34:01 2003 Subject: SDRAM row page size wierdness In-Reply-To: References: Message-ID: <3EA46B60.6080503@bitworks.com> ron minnich wrote: > Is it registered dram? > > Then the rules change just a bit. No.. I saw all that funnyness in your RPS routine from the 440gx and that was the first thing I looked at. At least I don't think it registered DRAM. It's not taking the registered dram code path for sure. -- Richard A. Smith rsmith at bitworks.com From gwatson at lanl.gov Mon Apr 21 17:47:01 2003 From: gwatson at lanl.gov (Greg Watson) Date: Mon Apr 21 17:47:01 2003 Subject: p4dpe/g2 and p4dpeg2 builds broken In-Reply-To: References: Message-ID: At 4:02 PM -0600 21/4/03, ron minnich wrote: >On 21 Apr 2003, Eric W. Biederman wrote: > >> But the code is in a library so unless it is referenced it should >> compile but not be used. >> >> Which is why it really confused me that greg had problems with it. > >good point. Greg, was the issue that the code would not compile or you >just did not want it in there on PPC? > >ron I only removed code that wouldn't compile for some reason. However I just tried it and it seems to compile ok now (apart from a warning about the unused variable i on line 139 :-), so put it back in if you want. Greg From rsmith at bitworks.com Mon Apr 21 21:19:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 21 21:19:01 2003 Subject: SDRAM row page size wierdness In-Reply-To: <3EA46B60.6080503@bitworks.com> References: <3EA46B60.6080503@bitworks.com> Message-ID: <3EA49FEE.5030709@bitworks.com> Richard Smith wrote: Ok... I got the page size detection worked out. I'm still confused but at least it works. Going back over the original 440BX code I noticed that it used the number of _column_ address rather than the number of row adresses. At first I thought this was a typo (spd loc 4 rather than 3). This was further reenforced by the 440gx code and Eriks general code which use the number of row adress bits. Sounds logical... The number of row adress bits indicating row page size but... After testing several DIMMs using the row address bits failed in about 50% of the cases so I started looking back at the column address bits. I don't understand why but if you use the number of column address bits as a index for what page size to use it works great. Basically just subtract 8 from the value to shift it down to zero. 8 column bits = 2k page 9 = 4k 10 = 8k The 440bx dosen't do beyond 8k pages. I tested as many different DIMMs as I could find here and they all work. Any ideas on why this would be the case? -- Richard A. Smith rsmith at bitworks.com From ebiederman at lnxi.com Mon Apr 21 22:27:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 21 22:27:01 2003 Subject: SDRAM row page size wierdness In-Reply-To: <3EA49FEE.5030709@bitworks.com> References: <3EA46B60.6080503@bitworks.com> <3EA49FEE.5030709@bitworks.com> Message-ID: Richard Smith writes: > Richard Smith wrote: > > Ok... I got the page size detection worked out. I'm still confused but at > least it works. > > Going back over the original 440BX code I noticed that it used the number of > _column_ address rather than the number of row adresses. At first I thought > this was a typo (spd loc 4 rather than 3). This was further reenforced by the > 440gx code and Eriks general code which use the number of row adress bits. My code was wrong. > Sounds logical... The number of row adress bits indicating row page size but... The number of column address bits determine the size of a row. the number of row address bits determine the number of rows you have. > > After testing several DIMMs using the row address bits failed in about 50% of > the cases so I started looking back at the column address bits. I don't > understand why but if you use the number of column address bits as a index for > what page size to use it works great. > > Basically just subtract 8 from the value to shift it down to zero. > 8 column bits = 2k page > 9 = 4k > 10 = 8k > > The 440bx dosen't do beyond 8k pages. > > I tested as many different DIMMs as I could find here and they all work. > > Any ideas on why this would be the case? See above. I think I spotted this a long time ago, but forget to do anything about it as I don't have a 440 board. Eric From dkotian3 at vsnl.net Mon Apr 21 23:47:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 23:47:00 2003 Subject: About 82810e north bridge References: Message-ID: <000401c30886$5993bc80$782041db@vsnl.net> OK, One more question, would the messages come up on the console if I do not the initialize Graphical Controller HUB. There is an option in the LinuxBIOS for VGA_CONSOLE or SERIAL_CONSOLE to display messages on the screen. Would VGA_CONSOLE be OK without the above initialization. Could one please elaborate on this. ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: Sent: Tuesday, April 22, 2003 2:58 AM Subject: Re: About 82810e north bridge > On Tue, 22 Apr 2003, Deepak Kotian wrote: > > > There seems to be two devices in Intel 82810e north bridge, one is > > Memory controller > > Hub and other is Graphical controller HUB. > > Is it OK to just do the initialization for Memory controller HUB > > and forget about Graphical controller HUB. > > I think so. > > > > > Also, I saw 82815ep on linuxbios tree, which has support for > > MCH only and no AGP. This is very much similar to 82810e with > > some address changed. > > I don't think this hardware is at all well tested, so if you can fix it, > we would be grateful. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Mon Apr 21 23:48:13 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 21 23:48:13 2003 Subject: About 82810e north bridge References: Message-ID: <000501c30886$5a10dbc0$782041db@vsnl.net> OK, One more question, would the messages come up on the console if I do not the initialize Graphical Controller HUB. There is an option in the LinuxBIOS for VGA_CONSOLE or SERIAL_CONSOLE to display messages on the screen. Would VGA_CONSOLE be OK without the above initialization. Could one please elaborate on this. Regards Deepak ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: Sent: Tuesday, April 22, 2003 2:58 AM Subject: Re: About 82810e north bridge > On Tue, 22 Apr 2003, Deepak Kotian wrote: > > > There seems to be two devices in Intel 82810e north bridge, one is > > Memory controller > > Hub and other is Graphical controller HUB. > > Is it OK to just do the initialization for Memory controller HUB > > and forget about Graphical controller HUB. > > I think so. > > > > > Also, I saw 82815ep on linuxbios tree, which has support for > > MCH only and no AGP. This is very much similar to 82810e with > > some address changed. > > I don't think this hardware is at all well tested, so if you can fix it, > we would be grateful. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rminnich at lanl.gov Mon Apr 21 23:53:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 21 23:53:00 2003 Subject: About 82810e north bridge In-Reply-To: <000401c30886$5993bc80$782041db@vsnl.net> Message-ID: On Tue, 22 Apr 2003, Deepak Kotian wrote: > One more question, would the messages come up on the console if I > do not the initialize Graphical Controller HUB. no :-( > There is an option in the LinuxBIOS for VGA_CONSOLE or SERIAL_CONSOLE > to display messages on the screen. > Would VGA_CONSOLE be OK without the above initialization. no :-( ron From shubhangi.jadhav at patni.com Tue Apr 22 00:22:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 22 00:22:00 2003 Subject: _ROMBASE Message-ID: Hi, Does _ROMBASE=0xf0000 hold good for all i386 processors. I'm using a VIA C3 processor, is this value of _ROMBASE correct. Thanks, Shubhangi From rminnich at lanl.gov Tue Apr 22 00:37:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 00:37:00 2003 Subject: _ROMBASE In-Reply-To: Message-ID: On Tue, 22 Apr 2003, Shubhangi Jadhav wrote: > Does _ROMBASE=0xf0000 hold good for all i386 processors. I'm using a VIA C3 > processor, is this value of _ROMBASE correct. That should be fine. ron From dkotian3 at vsnl.net Tue Apr 22 02:11:00 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Tue Apr 22 02:11:00 2003 Subject: entry16.inc code doubt Message-ID: <20030422071708.8B3124FE77@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From shubhangi.jadhav at patni.com Tue Apr 22 03:28:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 22 03:28:01 2003 Subject: Is reset32.inc required ? Message-ID: Hi, I'm trying to port Linuxbios to a motherboard which has a VIA C3 processor. For this I'm using l440bx mainboard config file as a reference. In the config file I found reference to the following includes : cpu/i386/entry16.inc cpu/i386/entry32.inc cpu/i386/reset16.inc However there's no reference to reset32.inc. Is it required to add reset32.inc, please let me know. Regards, Shubhangi From klos1980 at sina.com Tue Apr 22 03:49:01 2003 From: klos1980 at sina.com (klos) Date: Tue Apr 22 03:49:01 2003 Subject: serial debug patch for ADLO In-Reply-To: <8470181DABD5D511B3E700D0B7A8AC4A9CD721@cms3> Message-ID: <000101c308a3$2c7b5ec0$7ca8a8c0@sh.corp.tyan.com> Hi, It seems to have no responds in my serial port; I use etherboot to load the ADLO. Loading 192.168.1.1: payload ... (ELF)............................................................................... ................done No more information. Some one gives me some suggestions? Baud rate? Best regards klos -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org] On Behalf Of hcyun at etri.re.kr Sent: 2003?4?19? 23:07 To: adam at cfar.umd.edu; linuxbios at clustermatic.org Subject: serial debug patch for ADLO Hello, I made a patch which make all bochs debug message can be seen from serial terminal. I think it might be useful if you want to apply ADLO to your own target. you can apply this patch againt util/ADLO/bochs/bios/rombios.c by # patch -p0 < rombios-serial.patch Regards, From hcyun at etri.re.kr Tue Apr 22 04:30:01 2003 From: hcyun at etri.re.kr (hcyun at etri.re.kr) Date: Tue Apr 22 04:30:01 2003 Subject: serial debug patch for ADLO Message-ID: <8470181DABD5D511B3E700D0B7A8AC4A9CD728@cms3> The patch doesn't initialize serial port because it assume serial was already initialized by linuxbios. I think your problem is you didn't enabled debugs flags. you must enable DEBUG_ROMBIOS and also appropriate debug flags you want to see such as DEBUG_INT15, DEBUG_INT16 and etc. Heechul > -----Original Message----- > From: klos [mailto:klos1980 at sina.com] > Sent: Tuesday, April 22, 2003 4:45 PM > To: hcyun at etri.re.kr; adam at cfar.umd.edu; linuxbios at clustermatic.org > Subject: RE: serial debug patch for ADLO > > > Hi, > It seems to have no responds in my serial port; I use > etherboot to load the ADLO. > Loading 192.168.1.1: payload > ... > (ELF)......................................................... > ...................... > ................done > No more information. > > Some one gives me some suggestions? > > Baud rate? > > > Best regards > klos > -----Original Message----- > From: linuxbios-admin at clustermatic.org > [mailto:linuxbios-admin at clustermatic.org] On Behalf Of > hcyun at etri.re.kr > Sent: 2003?4?19? 23:07 > To: adam at cfar.umd.edu; linuxbios at clustermatic.org > Subject: serial debug patch for ADLO > > Hello, > I made a patch which make all bochs debug message can be seen > from serial terminal. > I think it might be useful if you want to apply ADLO to your > own target. > you can apply this patch againt util/ADLO/bochs/bios/rombios.c by > # patch -p0 < rombios-serial.patch > Regards, > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stuge-linuxbios at cdy.org Tue Apr 22 05:16:01 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Tue Apr 22 05:16:01 2003 Subject: entry16.inc code doubt In-Reply-To: <20030422071708.8B3124FE77@bom6.vsnl.net.in> References: <20030422071708.8B3124FE77@bom6.vsnl.net.in> Message-ID: <20030422093825.GB12989@foo.birdnet.se> On Tue, Apr 22, 2003 at 12:17:08PM +0500, dkotian3 at vsnl.net wrote: > I checked the pentium manuals, could not find any reference > to these statements as listed below. > > > > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > > > orl $0x60000001, %eax /* CD, NW, PE = 1 */ > > The syntax for and is reverse > and > Whereas the code reads > andl > > Could someone please help me on this or redirect to > exact reference for the above statements. Hi. The two instructions you quote are taken out of context and are by themselves completely meaningless. They are mere bit operations on registers: andl performs a bitwise AND of a register (the eax register, in this case) and the 0x7ffaffd1 immediate value, and stores the result back into the register. orl performs a bitwise OR of a register (again eax) and the 0x60000001 immediate value, and stores the result back into eax. Before these two instructions I'm guessing that there is a line reading mov %eax, cr0 and following them, there is a line reading mov cr0, %eax (cr0 above may also be %cr0, I'm more familiar with intel assembly syntax.) Now, this puts the bit mangling in a totally different perspective. cr0 is the Configuration Register 0, which among other things contains the PE bit, as indicated by the comment following the or instruction. PE controls whether the CPU is in protected mode or not. To connect the dots for you, your two instructions combined with my two instructions will set a number of different flags in cr0, among other things switching the CPU to protected mode if not already in it. Hope this helps, and gives a few pointers. //Peter From pyro at linuxlabs.com Tue Apr 22 07:19:00 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 22 07:19:00 2003 Subject: entry16.inc code doubt In-Reply-To: <20030422071708.8B3124FE77@bom6.vsnl.net.in> Message-ID: Greetings, What you're seeing is the difference between Intel syntax (used in their assemblers) and AT&T syntax used in the GNU assembler. The big differences are that the operands are reversed, and the mnemonic (opcode) explicitly refers to the operand size by appending one of b, w, or l for byte, word, or long (1, 2, or 4 bytes). G'day, sjames On Tue, 22 Apr 2003 dkotian3 at vsnl.net wrote: > Hi, > > I checked the pentium manuals, could not find any reference > to these statements as listed below. > > > > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > > > orl $0x60000001, %eax /* CD, NW, PE = 1 */ > > The syntax for and is reverse > and > Whereas the code reads > andl > > Could someone please help me on this or redirect to > exact reference for the above statements. > > Regards > Deepak > > > > > > you really need to get a pengium manual. > > > Where do I get this one. Is it some Linux Man Page. > Could you please give me the link where I can look for > it. > > Thanks and Regards > Deepak > > > ----- Original Message ----- > From: "ron minnich" > To: > Cc: > Sent: Monday, April 21, 2003 9:34 PM > Subject: Re: entry16.inc code doubt > > > > On Mon, 21 Apr 2003 dkotian3 at vsnl.net wrote: > > > > > ***Extract*** > > > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > > > orl $0x60000001, %eax /* CD, NW, PE = 1 */ > > > > you really need to get a pengium manual. > > > > > > Then it will all make sense. > > > > ron > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From zhushisongzhu at yahoo.com Tue Apr 22 09:15:00 2003 From: zhushisongzhu at yahoo.com (zhu shi song) Date: Tue Apr 22 09:15:00 2003 Subject: (no subject) Message-ID: <20030422134749.14146.qmail@web13201.mail.yahoo.com> i have boot linux-2.4.20 using linuxbios cvs 20030422 under winfast6300 successfully. But i met two problems: (1) acpi error in kernel (2) I can't warm reboot linux using ctrl-alt-del. thankszhu --------------------------------- Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. -------------- next part -------------- An HTML attachment was scrubbed... URL: From nathanael at gnat.ca Tue Apr 22 09:33:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 09:33:00 2003 Subject: super I/O Message-ID: <9AB03DEA-74CB-11D7-9E6E-0003931B4D6A@gnat.ca> Hello, I've been working on getting linuxbios working on a sis530 board on and off for the last little while, using a datasheet I have for the sis530 and the actual SBC I have. I finally have it all compiling. Though I know I'm using some sourcefiles that are incorrect for my board since they have been taken from the other sis based mainboards. The super IO for this board is an ITE 8661F and the only one in the tree is the ITE8671F. Obviously they aren't exactly the same, so I'm wondering how I proceed to get that chip working. I don't have any special hardware to do this with, (logic analyzer etc..). Would there be a spec sheet somewhere I could get? Ideas on ways to proceed? -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From bari at onelabs.com Tue Apr 22 09:43:01 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 09:43:01 2003 Subject: super I/O References: <9AB03DEA-74CB-11D7-9E6E-0003931B4D6A@gnat.ca> Message-ID: <3EA54EB0.6020603@onelabs.com> Nathanael Noblet wrote: > The super IO for this board is an ITE 8661F and the only one in the > tree is the ITE8671F. Obviously they aren't exactly the same, so I'm > wondering how I proceed to get that chip working. I don't have any > special hardware to do this with, (logic analyzer etc..). Would there > be a spec sheet somewhere I could get? Ideas on ways to proceed? > http://www.iteusa.com/productInfo/Download.html#IT8661F for the data sheets. --Bari From rminnich at lanl.gov Tue Apr 22 09:45:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 09:45:01 2003 Subject: entry16.inc code doubt In-Reply-To: <20030422071708.8B3124FE77@bom6.vsnl.net.in> Message-ID: On Tue, 22 Apr 2003 dkotian3 at vsnl.net wrote: > I checked the pentium manuals, could not find any reference > to these statements as listed below. > > > > andl $0x7FFAFFD1, %eax /* PG,AM,WP,NE,TS,EM,MP = 0 */ > > > orl $0x60000001, %eax /* CD, NW, PE = 1 */ You need to learn how to read Gnu Assembler. This is GNU assembly code. ron From rminnich at lanl.gov Tue Apr 22 09:51:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 09:51:01 2003 Subject: (no subject) In-Reply-To: <20030422134749.14146.qmail@web13201.mail.yahoo.com> Message-ID: On Tue, 22 Apr 2003, zhu shi song wrote: > i have boot linux-2.4.20 using linuxbios cvs 20030422 under winfast6300 > successfully. good! > But i met two problems: (1) acpi error in kernel (2) I > can't warm reboot linux using ctrl-alt-del. thankszhu you're not going to get acpi, and that is fine by me, as ACPI is a really BAD standard. For the reboot case you probably need my reboot patch, see the kernel_patches directory. ron From rminnich at lanl.gov Tue Apr 22 09:55:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 09:55:00 2003 Subject: super I/O In-Reply-To: <9AB03DEA-74CB-11D7-9E6E-0003931B4D6A@gnat.ca> Message-ID: On Tue, 22 Apr 2003, Nathanael Noblet wrote: > Hello, > I've been working on getting linuxbios working on a sis530 board on > and off for the last little while, using a datasheet I have for the > sis530 and the actual SBC I have. I finally have it all compiling. > Though I know I'm using some sourcefiles that are incorrect for my > board since they have been taken from the other sis based mainboards. > The super IO for this board is an ITE 8661F and the only one in the > tree is the ITE8671F. Obviously they aren't exactly the same, so I'm > wondering how I proceed to get that chip working. I don't have any > special hardware to do this with, (logic analyzer etc..). Would there > be a spec sheet somewhere I could get? Ideas on ways to proceed? you need to get a spec sheet from the ITE web site. create superio/ITE/8661F and clone the files from the 8671F into it. Then modify these files as needed to make it work. ron From nathanael at gnat.ca Tue Apr 22 10:17:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 10:17:01 2003 Subject: super I/O In-Reply-To: Message-ID: On Tuesday, April 22, 2003, at 08:26 AM, ron minnich wrote: > you need to get a spec sheet from the ITE web site. create > superio/ITE/8661F and clone the files from the 8671F into it. Then > modify > these files as needed to make it work. Based on Bari's link, I got the spec sheet, I must say the ite website barely works. In anycase I've done as you suggested, there were only a couple changes to the existing code to make it enter the MB PnP config. Then the code makes some calls to enable the floppy serial etc... I'm not sure where they got the parameters they're using. So I tried to get the spec sheet for the ite 8671f to make comparison against. Well ite's site is so disfunctional (most links don't work), I can't find / get it as well. Does anyone have the ite 8671f spec sheet or a direct link to it? -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From nathanael at gnat.ca Tue Apr 22 10:24:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 10:24:01 2003 Subject: super I/O In-Reply-To: Message-ID: Ignore my last post. I think I have it all figured out (ite 8661f). I might actually be attempting some boots today, this could be fun. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Tue Apr 22 10:30:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 10:30:00 2003 Subject: super I/O In-Reply-To: Message-ID: On Tue, 22 Apr 2003, Nathanael Noblet wrote: > Ignore my last post. I think I have it all figured out (ite 8661f). I > might actually be attempting some boots today, this could be fun. be sure and send me diffs :-) congratulations. ron From bari at onelabs.com Tue Apr 22 10:40:01 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 10:40:01 2003 Subject: super I/O References: Message-ID: <3EA55C04.3020506@onelabs.com> Nathanael Noblet wrote: > Does anyone have the ite 8671f spec sheet or a direct link to it? > IIRC the 8671 is no longer in production. All the ITE data sheets currently are at: http://www.iteusa.com/productInfo/Download.html --Bari From nathanael at gnat.ca Tue Apr 22 10:48:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 10:48:00 2003 Subject: super I/O In-Reply-To: <3EA55C04.3020506@onelabs.com> Message-ID: <1957E3DB-74D6-11D7-9E6E-0003931B4D6A@gnat.ca> On Tuesday, April 22, 2003, at 09:13 AM, Bari Ari wrote: > Nathanael Noblet wrote: > >> Does anyone have the ite 8671f spec sheet or a direct link to it? >> > IIRC the 8671 is no longer in production. All the ITE data sheets > currently are at: > http://www.iteusa.com/productInfo/Download.html Yeah I was looking there, and downloaded a spec sheet for a 8671 but it seemed fairly different in purpose from the 8661 so I was a little confused. I did end up though figuring out what the code for the 8671 was doing with respect to the 8661. There is only one line in the code I am unsure of from setup_serial.inc: SIO_WRITE($0x00,$0x24,$0x00) which says it is selecting the 24MHz CLKIN and clears software suspend mode. I don't know if it is correct for the board I have, I really hope so...;) -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From bari at onelabs.com Tue Apr 22 10:53:00 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 10:53:00 2003 Subject: super I/O References: <1957E3DB-74D6-11D7-9E6E-0003931B4D6A@gnat.ca> Message-ID: <3EA55F21.2030901@onelabs.com> Nathanael Noblet wrote: > > On Tuesday, April 22, 2003, at 09:13 AM, Bari Ari wrote: > >> Nathanael Noblet wrote: >> >>> Does anyone have the ite 8671f spec sheet or a direct link to it? >>> >> IIRC the 8671 is no longer in production. All the ITE data sheets >> currently are at: >> http://www.iteusa.com/productInfo/Download.html > > > Yeah I was looking there, and downloaded a spec sheet for a 8671 but > it seemed fairly different in purpose from the 8661 so I was a little > confused. I did end up though figuring out what the code for the 8671 > was doing with respect to the 8661. There is only one line in the code > I am unsure of > from setup_serial.inc: SIO_WRITE($0x00,$0x24,$0x00) > which says it is selecting the 24MHz CLKIN and clears software suspend > mode. I don't know if it is correct for the board I have, I really > hope so...;) > To make things more confusing ITE made both a 8671 and 8761. The 8671 is gone but the 8761 came out a couple years later and is in production. I think you may be looking at the 8761 data sheet. --Bari From nathanael at gnat.ca Tue Apr 22 11:08:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 11:08:00 2003 Subject: pci_init_table Message-ID: Hello again, Working on an sis530 based board. I've taken most code from other boards and have been modifying the code to what I think it should be for my board. Since my board has a DoC socket (and I have a DoC) I would like to use it. So I have altered the ipl.S file where it needed altering (the sis950_init_table is now the it8661f_init_table and the registers have been changed to suit). I've come across the mainboard dll.inc file which is commented out in the ipl.S file and has similar clock control statements. I've looked at my spec sheet and I have no where near that many registers for clock control. I'm wondering what they all do. What the purpose is, this way I may be able to find the information in the sis530 spec sheet. As well it would be nice to know whether they are being set to their default values or something else, because all I know is what to set the default values to. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Tue Apr 22 11:10:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 11:10:01 2003 Subject: intel firmware hub parts Message-ID: has anyone found a distributor for these parts? this is really amazing, but nobody I have talked to sells them. They get them via hook and crook from intel, but actually buy them? never happens. ron From bari at onelabs.com Tue Apr 22 11:23:01 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 11:23:01 2003 Subject: intel firmware hub parts References: Message-ID: <3EA56623.5040504@onelabs.com> ron minnich wrote: >has anyone found a distributor for these parts? this is really amazing, >but nobody I have talked to sells them. They get them via hook and crook >from intel, but actually buy them? never happens. > > > Do you have a part number? --Bari From rminnich at lanl.gov Tue Apr 22 11:25:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 11:25:00 2003 Subject: intel firmware hub parts In-Reply-To: <3EA56623.5040504@onelabs.com> Message-ID: On Tue, 22 Apr 2003, Bari Ari wrote: > ron minnich wrote: > > >has anyone found a distributor for these parts? this is really amazing, > >but nobody I have talked to sells them. They get them via hook and crook > >from intel, but actually buy them? never happens. > > > > > > > Do you have a part number? 82802ab, 82802ac ron From rminnich at lanl.gov Tue Apr 22 11:37:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 11:37:00 2003 Subject: new build failure on p4dpe-g2 Message-ID: cp /home/rminnich/src/bios/freebios/src/arch/i386/config/crt0.base crt0.S gcc -x assembler-with-cpp -DASSEMBLY -E ... crt0.S > crt0.s cpp0: warning: changing search order for system directory "/usr/lib/gcc-lib/i386-redhat-linux/3.2/include" cpp0: warning: as it has already been specified as a non-system directory In file included from crt0_includes.h:15, from crt0.S:41: /home/rminnich/src/bios/freebios/src/southbridge/intel/82801ca/cmos_failover.inc:42:2: #error "MAX_REBOOT_CNT not defined" I'll set it but how does it get set automagically? ron From garlick at llnl.gov Tue Apr 22 12:06:00 2003 From: garlick at llnl.gov (Jim Garlick) Date: Tue Apr 22 12:06:00 2003 Subject: intel firmware hub parts In-Reply-To: Message-ID: I bought some from Arrow - contact was K. Messner (303) 600-1287. Jim On Tue, 22 Apr 2003, ron minnich wrote: > > has anyone found a distributor for these parts? this is really amazing, > but nobody I have talked to sells them. They get them via hook and crook > from intel, but actually buy them? never happens. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rminnich at lanl.gov Tue Apr 22 12:17:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 12:17:00 2003 Subject: intel firmware hub parts In-Reply-To: Message-ID: On Tue, 22 Apr 2003, Jim Garlick wrote: > I bought some from Arrow - contact was K. Messner (303) 600-1287. I just talked to her. They are "on allocation" which I think is bad :-) Thanks for the tip Jim. This is why I love this list. ron From bari at onelabs.com Tue Apr 22 12:26:01 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 12:26:01 2003 Subject: intel firmware hub parts References: Message-ID: <3EA574E0.60501@onelabs.com> ron minnich wrote: >>>has anyone found a distributor for these parts? this is really amazing, >>>but nobody I have talked to sells them. >>> >>> >>> Do you have a part number? >>> >>> >82802ab, 82802ac > > > Arrow 1-877-237-8621 http://www.arrow.com/ Avnet 1-800-408-8353 http://www.avnet.com/em/ --Bari From rminnich at lanl.gov Tue Apr 22 12:35:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 12:35:01 2003 Subject: intel firmware hub parts In-Reply-To: <3EA574E0.60501@onelabs.com> Message-ID: On Tue, 22 Apr 2003, Bari Ari wrote: > >82802ab, 82802ac and I just learned: N82802AC8 works in their data base much better :-) ron From bari at onelabs.com Tue Apr 22 12:56:01 2003 From: bari at onelabs.com (Bari Ari) Date: Tue Apr 22 12:56:01 2003 Subject: intel firmware hub parts References: Message-ID: <3EA57C11.8070704@onelabs.com> ron minnich wrote: >On Tue, 22 Apr 2003, Bari Ari wrote: > > > >>>82802ab, 82802ac >>> >>> > >and I just learned: > >N82802AC8 > >works in their data base much better :-) > > > That's the fun of Intel. Rarely a full part number in the data sheet. --Bari From ebiederman at lnxi.com Tue Apr 22 13:04:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 22 13:04:00 2003 Subject: Hammer Documentation is now Available at AMD's website Message-ID: So far I have just skimmed the Documentation but it appears that AMD has now publicly published everything needed to do Hammer LinuxBIOS port. No great surprise but it is nice to actually see it ! Yeah! The chaos with the Dual Athlons has finally paid off :) Eric From nathanael at gnat.ca Tue Apr 22 13:09:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 13:09:01 2003 Subject: flashing the bios Message-ID: Ok, so again here I am working on a sis 530 based board. I'd like to copy the bios to disk as a precaution, I would also like to flash linuxbios to the flash chip. I also need to know how to get the kernel in the right place on the DoC. Are there any docs on how to do that? I have looked at the flash_and_burn utils, which work only for the 630/950 sis stuff. Where do I go from now? -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Tue Apr 22 13:17:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 13:17:00 2003 Subject: flashing the bios In-Reply-To: Message-ID: On Tue, 22 Apr 2003, Nathanael Noblet wrote: > Ok, so again here I am working on a sis 530 based board. I'd like to > copy the bios to disk as a precaution, I would also like to flash > linuxbios to the flash chip. I also need to know how to get the kernel > in the right place on the DoC. Are there any docs on how to do that? I > have looked at the flash_and_burn utils, which work only for the > 630/950 sis stuff. Where do I go from now? if you are using doc, use MTD. It has worked well for DoC. use dumpdevmem to save your current flash to disk. you can fix flash_rom to burn flash on the 550, I doubt there is much difference. ron From nathanael at gnat.ca Tue Apr 22 13:21:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 13:21:00 2003 Subject: flashing the bios In-Reply-To: Message-ID: <61543456-74EB-11D7-8C5D-0003931B4D6A@gnat.ca> On Tuesday, April 22, 2003, at 11:48 AM, ron minnich wrote: > On Tue, 22 Apr 2003, Nathanael Noblet wrote: > >> Ok, so again here I am working on a sis 530 based board. I'd like to >> copy the bios to disk as a precaution, I would also like to flash >> linuxbios to the flash chip. I also need to know how to get the kernel >> in the right place on the DoC. Are there any docs on how to do that? I >> have looked at the flash_and_burn utils, which work only for the >> 630/950 sis stuff. Where do I go from now? > > if you are using doc, use MTD. It has worked well for DoC. ok, so I use MTD (you mean the mtd stuff in the kernel) to put both linuxbios AND the kernel on the DoC? If so, are there docs out there about how to do this, doesn't linuxbios need to be on the flash chip? > use dumpdevmem to save your current flash to disk. ok > you can fix flash_rom to burn flash on the 550, I doubt there is much > difference. the less work the better... ;) -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Tue Apr 22 13:59:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 13:59:01 2003 Subject: flashing the bios In-Reply-To: <61543456-74EB-11D7-8C5D-0003931B4D6A@gnat.ca> Message-ID: On Tue, 22 Apr 2003, Nathanael Noblet wrote: > ok, so I use MTD (you mean the mtd stuff in the kernel) to put both > linuxbios AND the kernel on the DoC? If so, are there docs out there > about how to do this, doesn't linuxbios need to be on the flash chip? I thought the 630 HOWTO did some of this. anyway, linuxbios can be on flash, and the kernel on DoC. ron From ebiederman at lnxi.com Tue Apr 22 14:55:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 22 14:55:00 2003 Subject: Status freebios2 Message-ID: There is now a port that builds in the freebios2 tree. Currently it includes the latest romcc code. The build will now fail if you don't include exactly one stream driver to load your kernel. Drivers for the console code are now have an ELF section they are selected from. Cpu configuration still has to be rethought. The current mess where there are multiple CPUs selected is just wrong. I think sharing parts between cpus is pretty much the same as sharing parts between motherboards as far as solutions. This still needs to be looked at. In addition I need to see how to fit cpu configuration into the the generic pci setup code. There need to be hooks to enumerate cpus just like everything else. The only board in the tree so far is the amd solo. A prototype Athlon64 board. Enumerating a Hypertransport chain is a different problem from Enumerating a pci bus so that code still needs to be written. And to make it fun the HT chain needs to be setup before we can talk to the smbus controller on the 8111. Anyway this is starting point for a clean interface. The code is still in flux and but this is a snapshot so you can get the feel of it. And so work on the solo port can be coordinated. Romcc needs to mature more. There is a bug in constant propagation which keeps some C files from compiling. And romcc cannot handle setting manipulating mtrrs. So there are pieces that are still done in assembly. But it is a start. Eric From agnew at cs.umd.edu Tue Apr 22 15:15:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Tue Apr 22 15:15:00 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: Message-ID: <20030422154739.S14109-100000@www.missl.cs.umd.edu> Perhaps the LinuxBIOS project can commit to supporting *every* AMD 64-bit chipset? It seems like a good place to draw a line in the sand, as information is starting out freely available. Thoughts? On 22 Apr 2003, Eric W. Biederman wrote: > So far I have just skimmed the Documentation but it appears that > AMD has now publicly published everything needed to do Hammer > LinuxBIOS port. > > No great surprise but it is nice to actually see it ! > > Yeah! The chaos with the Dual Athlons has finally paid off :) > > Eric > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From nathanael at gnat.ca Tue Apr 22 15:46:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 22 15:46:01 2003 Subject: mtd & bios flashing Message-ID: <9B4596A4-74FF-11D7-B156-0003931B4D6A@gnat.ca> Hello again, you may be sick of my questions today but well here's another one..;) I read over the Sis630 howto. It does explain pretty well how to get this all flashed. I have two problems, it uses the flash_on / flash_and_burn programs which don't work on my board (sis 530 based). I don't know where to start to get that working pointers would be nice if this is the road I should take. I say should take, because it was mentioned that I could use the mtd utils to burn to the flash that the original bios is on. I don't understand how I can do this. I haven't seen anything in the mtd packages that would allow me to find the bios holding flash chip. I also noticed that some of the instructions assume that the DoC and the flash are in the same socket (as in you remove the flash chip and replace it with a DoC) my particular board has a PLCC type flash AND a DoC. I have the DoC working close to perfectly. I had grub installed and able to boot from it. I was thinking of installing a jffs2 file system on it. I also am just realizing that I wouldn't be using grub or any "bootloader" persay. Which is good because well, the grub doesn't support jffs2 I don't think. In anycase, here is what I wonder about. Do I need to go about finding out how to access the flash holding the regular bios so as to be able to flash it? Is there documentation on basic procedures to do this. If there is an easier way, please let me know of that instead. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From ebiederman at lnxi.com Tue Apr 22 21:28:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Tue Apr 22 21:28:00 2003 Subject: [ANNOUNCE] mkelfImage-2.4 Message-ID: mkelfImage-2.4 is a bug fix release. Available at: ftp://ftp.lnxi.com/pub/src/mkelfImage/mkelfImage-2.4.tar.gz * 2.4 22 April 2003 - Fix code that failed to write more than the first ELF note. Resulting in checksums and other information not being present in the generated ELF image. This bug has been present since mkelfImage-2.0 From dkotian3 at vsnl.net Tue Apr 22 23:07:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Tue Apr 22 23:07:01 2003 Subject: Debugging the linuxBIOS code Message-ID: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From sivakumar.subramani at wipro.com Tue Apr 22 23:19:00 2003 From: sivakumar.subramani at wipro.com (siva) Date: Tue Apr 22 23:19:00 2003 Subject: Debugging the linuxBIOS code References: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> Message-ID: <0e0501c3094b$ac082d00$2a0806c0@wiproxcoio3398> Hi, You should get LinuxBIOS banner message even if your northbridge initialization is not correct. I think there is some problem in your serial port setting. Check it out. -Siva ----- Original Message ----- From: To: Sent: Wednesday, April 23, 2003 9:44 AM Subject: Debugging the linuxBIOS code > Hi, > > Is there a recommended way of debugging LinuxBIOS code. > > I burnt my FLASHROM with LinuxBIOS, nothing came > up at all. Not a single message came at the serial Output. > I knew that my North bridge code has issues, but still > something should have come. I guess.. > > I had connected the serial out of my PC to another PC > and I am using an Windows HyperTerminal to view > the messages. > Should it not show something at all even if north bridge code > is broken. > > Could anyone give their views on this one.. > > Regards > Deepak > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios **************************Disclaimer************************************ Information contained in this E-MAIL being proprietary to Wipro Limited is 'privileged' and 'confidential' and intended for use only by the individual or entity to which it is addressed. You are notified that any use, copying or dissemination of the information contained in the E-MAIL in any manner whatsoever is strictly prohibited. *************************************************************************** From rminnich at lanl.gov Tue Apr 22 23:19:09 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 22 23:19:09 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> Message-ID: On Wed, 23 Apr 2003 dkotian3 at vsnl.net wrote: > Is there a recommended way of debugging LinuxBIOS code. a rom emulator is good. Or, if you have more money, and America Arium ICE. > I had connected the serial out of my PC to another PC > and I am using an Windows HyperTerminal to view > the messages. never use hyperterminal. It is a piece of shit. It will lose character output all the time. Use minicom on linux. ron From aip at cwlinux.com Tue Apr 22 23:19:13 2003 From: aip at cwlinux.com (Andrew Ip) Date: Tue Apr 22 23:19:13 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in>; from dkotian3@vsnl.net on Wed, Apr 23, 2003 at 09:14:04AM +0500 References: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> Message-ID: <20030423115226.C3013@mail.cwlinux.com> Deepak, > Is there a recommended way of debugging LinuxBIOS code. > I burnt my FLASHROM with LinuxBIOS, nothing came > up at all. Not a single message came at the serial Output. > I knew that my North bridge code has issues, but still > something should have come. I guess.. > I had connected the serial out of my PC to another PC > and I am using an Windows HyperTerminal to view > the messages. > Should it not show something at all even if north bridge code > is broken. > Could anyone give their views on this one.. POST card would be helpful if you don't have emulator. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From steve at nexpath.com Wed Apr 23 00:14:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 23 00:14:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: References: Message-ID: <3EA61CF3.6020101@nexpath.com> > never use hyperterminal. It is a piece of shit. It will lose character > output all the time. Use minicom on linux. > > ron > I agree, hyperterminal is really bad. I use Tera Term Pro when using Windows and it seems okay. /sg From adam at cfar.umd.edu Wed Apr 23 00:18:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 23 00:18:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030423115226.C3013@mail.cwlinux.com> Message-ID: <20030423005522.G14763-100000@www.missl.cs.umd.edu> On Wed, 23 Apr 2003, Andrew Ip wrote: > POST card would be helpful if you don't have emulator. Speaking of which. Does anyone know 3.3V PORT-80 PCI cards instead of your usual 5V ? (and preferably does not cost $100 either). -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From CHall at optos.com Wed Apr 23 04:31:00 2003 From: CHall at optos.com (Colin Hall) Date: Wed Apr 23 04:31:00 2003 Subject: Debugging the linuxBIOS code Message-ID: Adam, My board has a 3v3-only PCI bus, so I have the same requirement. The cards I have considered are listed below. My understanding is that the Micro-2000 POST-Probe [5] is the only one that can be powered from 3v3. It does not meet your price point, I'm afraid. I am about to place an order for the rather expensive Micro-2000 POST-Probe, so anyone who can recommend something cheaper for a 3v3-only bus would be doing me a great favour. Best Regards, Colin. Refs: [1] "ToolStar PCI Professional POST Card", ProTech Diagnostics, [2] "P.H.D. PCI", Ultra-X (UK) Limited, [3] "PCI / ISA bus test tools", PC Engines GmbH, [4] "PCI POST Power-On Self-Test Card", pcwiz, [5] "POST-Probe ", Micro2000, > -----Original Message----- > From: Adam Sulmicki [mailto:adam at cfar.umd.edu] > Sent: 23 April 2003 05:57 > To: Andrew Ip > Cc: dkotian3 at vsnl.net; linuxbios at clustermatic.org > Subject: Re: Debugging the linuxBIOS code > > > On Wed, 23 Apr 2003, Andrew Ip wrote: > > > POST card would be helpful if you don't have emulator. > > Speaking of which. Does anyone know 3.3V PORT-80 PCI cards > instead of your > usual 5V ? (and preferably does not cost $100 either). > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > ********************************************************************** This footnote confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com ********************************************************************** This e-mail transmission is intended for the named addressee only. Its contents are private and confidential and should not be read, copied or disclosed by any other person. If you have received this transmission in error, please notify the sender immediately by e-mail. Opinions, conclusions and other information in this message that do not relate to the official business of my Company shall be understood as neither given nor endorsed by it. ________________________________________________ Internet e-mails are not necessarily secure. Optos plc does not accept responsibility for changes made to this message after it was sent. Copyright Optos plc All rights reserved. No part of this communication may be reproduced without the prior written permission of Optos plc. It is expressly declared that this e-mail does not constitute nor form part of a contract or unilateral obligation. From sivakumar.subramani at wipro.com Wed Apr 23 05:04:01 2003 From: sivakumar.subramani at wipro.com (siva) Date: Wed Apr 23 05:04:01 2003 Subject: Where to get the latest LinuxBIOS source code. Message-ID: <0f3501c3097b$f08289a0$2a0806c0@wiproxcoio3398> Hi, I am not able to use the CVS server to download the LinuxBIOS source code. Is it possible to download the latest CDS code in TAR format. Thanks, siva **************************Disclaimer************************************ Information contained in this E-MAIL being proprietary to Wipro Limited is 'privileged' and 'confidential' and intended for use only by the individual or entity to which it is addressed. You are notified that any use, copying or dissemination of the information contained in the E-MAIL in any manner whatsoever is strictly prohibited. *************************************************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: From dkotian3 at vsnl.net Wed Apr 23 06:38:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Wed Apr 23 06:38:01 2003 Subject: Where to get the latest LinuxBIOS source code. Message-ID: <20030423114437.A090D4FEEF@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From justin at street-vision.com Wed Apr 23 06:54:00 2003 From: justin at street-vision.com (Justin Cormack) Date: Wed Apr 23 06:54:00 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: <20030422154739.S14109-100000@www.missl.cs.umd.edu> References: <20030422154739.S14109-100000@www.missl.cs.umd.edu> Message-ID: <1051097256.5446.154.camel@lotte> On Tue, 2003-04-22 at 20:53, Adam Agnew wrote: > > Perhaps the LinuxBIOS project can commit to supporting *every* AMD 64-bit > chipset? It seems like a good place to draw a line in the sand, as > information is starting out freely available. Thoughts? I doubt it. Nvidia are not usually very forthcoming. Several of the chipsets look like being quite odd hybrids of existing athlon chipsets. Justin From shubhangi.jadhav at patni.com Wed Apr 23 07:19:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Wed Apr 23 07:19:00 2003 Subject: Subsystem ID Message-ID: Hi, Can we map the device id from lspci output be mapped to the Subsystem ID - the SID register that needs to be set while initializing any chip on the motherboard. Thanks, Shubhangi From hcyun at etri.re.kr Wed Apr 23 07:51:00 2003 From: hcyun at etri.re.kr (hcyun at etri.re.kr) Date: Wed Apr 23 07:51:00 2003 Subject: Reducing kernel ide driver probing time. Message-ID: <8470181DABD5D511B3E700D0B7A8AC4A9CD72D@cms3> Hello I found that ide driver take too much time for probing unnecessary drives. (drivers/ide-probe.c) it probe from hda to hdl as you can see below and it takes about 3 ~ 4 seconds. However common main board have only 2 ide controller so probing over hde is unnecessary. My solution is redefine MAX_HWIF as 2 (originally 6) so I can reduce probing time down to 1 second. Is there any other way which can control this such as using kernel parameter? VP_IDE: VIA vt8235 (rev 00) IDE UDMA133 controller on pci00:11.1 ide0: BM-DMA at 0x1ce0-0x1ce7, BIOS settings: hda:pio, hdb:pio ide1: BM-DMA at 0x1ce8-0x1cef, BIOS settings: hdc:pio, hdd:pio probing for hda: present=1, media=32, probetype=ATA hda: MAXAN MFD-44-32, ATA DISK drive probing for hdb: present=0, media=32, probetype=ATA probing for hdb: present=0, media=32, probetype=ATAPI probing for hdc: present=0, media=32, probetype=ATA probing for hdc: present=0, media=32, probetype=ATAPI probing for hdd: present=0, media=32, probetype=ATA probing for hdd: present=0, media=32, probetype=ATAPI probing for hde: present=0, media=32, probetype=ATA probing for hde: present=0, media=32, probetype=ATAPI probing for hdf: present=0, media=32, probetype=ATA probing for hdf: present=0, media=32, probetype=ATAPI probing for hdg: present=0, media=32, probetype=ATA probing for hdg: present=0, media=32, probetype=ATAPI probing for hdh: present=0, media=32, probetype=ATA probing for hdh: present=0, media=32, probetype=ATAPI probing for hdi: present=0, media=32, probetype=ATA probing for hdi: present=0, media=32, probetype=ATAPI probing for hdj: present=0, media=32, probetype=ATA probing for hdj: present=0, media=32, probetype=ATAPI probing for hdk: present=0, media=32, probetype=ATA probing for hdk: present=0, media=32, probetype=ATAPI probing for hdl: present=0, media=32, probetype=ATA probing for hdl: present=0, media=32, probetype=ATAPI ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 hda: attached ide-disk driver. Regard, Heechul. -------------- next part -------------- An HTML attachment was scrubbed... URL: From adam at cfar.umd.edu Wed Apr 23 08:31:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 23 08:31:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: Message-ID: <20030423090422.Q16182-100000@www.missl.cs.umd.edu> On Wed, 23 Apr 2003, Colin Hall wrote: > My understanding is that the Micro-2000 POST-Probe [5] is the only one > that can be powered from 3v3. > [5] "POST-Probe ", Micro2000, call me stupid, but where does it say it is an 3.3V? in fact looking at pinout of PCI interface in their manual it seems to me like it is 5V card. an 3.3V card will have Pins A1 and B1 on the same side that "noth" is. See http://www.interfacemasters.com/images/P_MINIPCI_TO_PCI_01.jpg yet on page 271 of http://micro2000.com/downloads/PPMan.pdf it is other way around. > It does not meet your price point, I'm afraid. It is mostly disgust with price googuing sicne I'm told you can get reular Port-80 card for $3. (at china). -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rminnich at lanl.gov Wed Apr 23 09:06:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 09:06:00 2003 Subject: Debugging the linuxBIOS code In-Reply-To: Message-ID: Maybe the New FAQ Guy can add that previous message about POST cards :-) ron From rminnich at lanl.gov Wed Apr 23 09:07:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 09:07:01 2003 Subject: Where to get the latest LinuxBIOS source code. In-Reply-To: <0f3501c3097b$f08289a0$2a0806c0@wiproxcoio3398> Message-ID: On Wed, 23 Apr 2003, siva wrote: > I am not able to use the CVS server to download the LinuxBIOS source > code. Is it possible to download the latest CDS code in TAR format. k I will try to set up a cron job. What's the problem with CVS, just curious. ron From sivakumar.subramani at wipro.com Wed Apr 23 09:13:00 2003 From: sivakumar.subramani at wipro.com (siva) Date: Wed Apr 23 09:13:00 2003 Subject: Where to get the latest LinuxBIOS source code. References: Message-ID: <116301c3099e$a9d00190$2a0806c0@wiproxcoio3398> Due to some firewall configuration in our office I am not able to access the CVS server. I have download the CVS tar from the "daily snap shot " link in the Linuxbios homepage. IS it the latest code? Thanks, -SIVA ----- Original Message ----- From: "ron minnich" To: "siva" Cc: Sent: Wednesday, April 23, 2003 7:09 PM Subject: Re: Where to get the latest LinuxBIOS source code. > On Wed, 23 Apr 2003, siva wrote: > > > I am not able to use the CVS server to download the LinuxBIOS source > > code. Is it possible to download the latest CDS code in TAR format. > k > I will try to set up a cron job. > > What's the problem with CVS, just curious. > > ron > **************************Disclaimer************************************ Information contained in this E-MAIL being proprietary to Wipro Limited is 'privileged' and 'confidential' and intended for use only by the individual or entity to which it is addressed. You are notified that any use, copying or dissemination of the information contained in the E-MAIL in any manner whatsoever is strictly prohibited. *************************************************************************** From prl-linuxbios at sychron.com Wed Apr 23 11:14:01 2003 From: prl-linuxbios at sychron.com (prl-linuxbios at sychron.com) Date: Wed Apr 23 11:14:01 2003 Subject: Where to get the latest LinuxBIOS source code. In-Reply-To: Your message of "Wed, 23 Apr 2003 19:15:54 +0530." <116301c3099e$a9d00190$2a0806c0@wiproxcoio3398> Message-ID: > Due to some firewall configuration in our office I am not able to access the > CVS server. I have download the CVS tar from the "daily snap shot " link in > the Linuxbios homepage. IS it the latest code? Are you able to get out on SSH, port 22? If so, set up CVS via SSH. Sourceforge make it quite easy to set up. My Etherboot CVS environment looks like this... CVSROOT=:ext:prl at cvs.etherboot.sourceforge.net:/cvsroot/etherboot CVS_RSH=/usr/bin/ssh From CHall at optos.com Wed Apr 23 11:25:01 2003 From: CHall at optos.com (Colin Hall) Date: Wed Apr 23 11:25:01 2003 Subject: Debugging the linuxBIOS code Message-ID: Adam, > > My understanding is that the Micro-2000 POST-Probe [5] is > the only one > > that can be powered from 3v3. > > > [5] "POST-Probe ", Micro2000, > > > call me stupid, but where does it say it is an 3.3V? It doesn't, and this is an important point that I am trying to clarify with their engineers who claim that it will work from 3v3. As I said, if anyone can recommend something better I would love to hear about it. Regards, Colin. ********************************************************************** This footnote confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com ********************************************************************** This e-mail transmission is intended for the named addressee only. Its contents are private and confidential and should not be read, copied or disclosed by any other person. If you have received this transmission in error, please notify the sender immediately by e-mail. Opinions, conclusions and other information in this message that do not relate to the official business of my Company shall be understood as neither given nor endorsed by it. ________________________________________________ Internet e-mails are not necessarily secure. Optos plc does not accept responsibility for changes made to this message after it was sent. Copyright Optos plc All rights reserved. No part of this communication may be reproduced without the prior written permission of Optos plc. It is expressly declared that this e-mail does not constitute nor form part of a contract or unilateral obligation. From rsmith at bitworks.com Wed Apr 23 11:46:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 23 11:46:00 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <3EA61CF3.6020101@nexpath.com> References: <3EA61CF3.6020101@nexpath.com> Message-ID: <3EA6BD0C.4060408@bitworks.com> Steve Gehlbach wrote: > > I agree, hyperterminal is really bad. I use Tera Term Pro when using > Windows and it seems okay. > Tera Term Pro what I use as well. Speaking of.. has anyone found a good free windows terminal program that will display the values in hex as well as ascii? Realterm is the only thing I've found so far. It's useable but not quite there yet. A split screen ascii/hex setup would be ideal. Outputing values in binary form is really handy for the ram init code where you can't use any of the nice bin2hex debugging macros due to lack of registers. -- Richard A. Smith rsmith at bitworks.com From ebiederman at lnxi.com Wed Apr 23 11:48:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 11:48:00 2003 Subject: Reducing kernel ide driver probing time. In-Reply-To: <8470181DABD5D511B3E700D0B7A8AC4A9CD72D@cms3> References: <8470181DABD5D511B3E700D0B7A8AC4A9CD72D@cms3> Message-ID: hcyun at etri.re.kr writes: > Hello > > I found that ide driver take too much time for probing unnecessary drives. > (drivers/ide-probe.c) > it probe from hda to hdl as you can see below and it takes about 3 ~ 4 > seconds. > However common main board have only 2 ide controller so probing over hde is > unnecessary. > My solution is redefine MAX_HWIF as 2 (originally 6) so I can reduce > probing time down to 1 second. > Is there any other way which can control this such as using kernel > parameter? iden=noprobe. Eric From ebiederman at lnxi.com Wed Apr 23 11:52:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 11:52:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: References: Message-ID: ron minnich writes: > On Wed, 23 Apr 2003 dkotian3 at vsnl.net wrote: > > > Is there a recommended way of debugging LinuxBIOS code. > > a rom emulator is good. Or, if you have more money, and America Arium ICE. On Athlons and Hammers there is a jtag port that with the right adapter causes the cpu itself to act like an ICE. I think the P4 has something similar but I don't remember. Except for getting that initial character of output, and discerning between a hardware and a software failure. I don't much see the need for an ICE. Eric From steve at nexpath.com Wed Apr 23 11:53:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Wed Apr 23 11:53:01 2003 Subject: [ANNOUNCE] mkelfImage-2.4 In-Reply-To: References: Message-ID: <3EA6C065.1010606@nexpath.com> Eric W. Biederman wrote: > mkelfImage-2.4 is a bug fix release Down loaded and compiled it, but I get: Internal error convert_magic 00000070 != a5a5a5a5 for bzImage or vmlinux. Debian gcc 2.95.4 and gas 2.12.90.0.1, linux 2.4.20. -Steve From ebiederman at lnxi.com Wed Apr 23 11:54:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 11:54:00 2003 Subject: Status freebios2 In-Reply-To: <20030423125322.60294.qmail@web13208.mail.yahoo.com> References: <20030423125322.60294.qmail@web13208.mail.yahoo.com> Message-ID: Please don't just email me directly on general questions like this. Please at least CC the list. zhu shi song writes: > where can i download freebios2 thanks zhu In CVS. check out freebios2 instead of freebios. Eric From ebiederman at lnxi.com Wed Apr 23 11:56:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 11:56:01 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: <20030422154739.S14109-100000@www.missl.cs.umd.edu> References: <20030422154739.S14109-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > Perhaps the LinuxBIOS project can commit to supporting *every* AMD 64-bit > chipset? It seems like a good place to draw a line in the sand, as > information is starting out freely available. Thoughts? We can only do that if someone has either the economic incentive to do it, or they know they will have the free time. We don't seem to have that yet. Eric From ebiederman at lnxi.com Wed Apr 23 12:06:59 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 12:06:59 2003 Subject: [ANNOUNCE] mkelfImage-2.4 In-Reply-To: <3EA6C065.1010606@nexpath.com> References: <3EA6C065.1010606@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > mkelfImage-2.4 is a bug fix release > > Down loaded and compiled it, but I get: > > Internal error convert_magic 00000070 != a5a5a5a5 > > for bzImage or vmlinux. Debian gcc 2.95.4 and gas 2.12.90.0.1, linux > 2.4.20. Yep. That is a symptom of the binutils bug. It would be nice to get a configure test for that. But at least I catch it at runtime. A newer version should fix the problem. I am currently running: as --version GNU assembler 2.13.90.0.10 20021010 Debian GNU/Linux Basically some versions of gas append junk to the end of assembly files. It does not show up as a problem except when your assembly code falls of the end of the object file on purpose, or you care a lot about size. Eric From antonio.catani at seceti.it Wed Apr 23 12:38:01 2003 From: antonio.catani at seceti.it (Antonio Catani) Date: Wed Apr 23 12:38:01 2003 Subject: award bios source code Message-ID: <000001c309ba$a2ab53d0$7c03020a@catani> Hi list, i dont know if is usefull, but i have downloaded award bios source code it is 2.59 mb zipped, if you want i can upload where you want -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Wed Apr 23 12:50:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 12:50:01 2003 Subject: award bios source code In-Reply-To: <000001c309ba$a2ab53d0$7c03020a@catani> Message-ID: no thank you, we have no interest in that source code. It is IP of the company that wrote it, and we are very careful about not transgressing IP. ron From joshua at joshuawise.com Wed Apr 23 12:52:00 2003 From: joshua at joshuawise.com (Joshua Wise) Date: Wed Apr 23 12:52:00 2003 Subject: award bios source code In-Reply-To: <000001c309ba$a2ab53d0$7c03020a@catani> References: <000001c309ba$a2ab53d0$7c03020a@catani> Message-ID: <200304231325.16384.joshua@joshuawise.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Wednesday 23 April 2003 1:06 pm, Antonio Catani wrote: > Hi list, i dont know if is usefull, but i have downloaded award bios > source code it is 2.59 mb zipped, if you want i can upload where you > want PLEASE do not do so. If anyone looks at this code, they will be 'tainted' - they cannot release any further code they write for LinuxBIOS. Thx, - -j - -- Joshua Wise | www.joshuawise.com GPG Key | 0xEA80E0B3 Quote | I akilled *@* by mistake -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQE+psx4Pn9tWOqA4LMRAmjyAJ9kszTTAWq2b/F2lyOpgjaX7nL+igCgg7PX sgUWWu/Eo9p0Ik5IoP0GDdM= =kzJa -----END PGP SIGNATURE----- From dkotian3 at vsnl.net Wed Apr 23 13:18:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 23 13:18:01 2003 Subject: Debugging the linuxBIOS code References: <20030423051552.90E854FF07@bom6.vsnl.net.in> <20030423140439.B4356@mail.cwlinux.com> Message-ID: <012201c309c0$d5a77c40$253e41db@vsnl.net> Thanks for the information, Andrew. Correct, this would be good in case machine hang. But still it should be very useful. May be in future , a smart "POST code" could have some small internal buffering with some kind of manual switch to move up and down the display to track the code. This may or may not sound sense , but just a thought. Regards Deepak ----- Original Message ----- From: "Andrew Ip" To: Cc: ; ; Sent: Wednesday, April 23, 2003 11:34 AM Subject: Re: Debugging the linuxBIOS code > Deepak, > > > Thanks. I like this list for it's good responses. > > Could you please elaborate on "POST card", what is it exactly ? > > You are correct, I do not have a simulator. > It is a PCI card which displays the value of port address 80. In > LinuxBIOS source, it is not hard to find, eg. > > intel_chip_post_macro(0x11) > > It will dump 0x11 to port 80, and POST card will display 11 on its > LED. > > While debugging, you can record the value and search thru the source > to see where LinuxBIOS hangs. However, you can't do backtracing with > this device. > > -Andrew > > -- > Andrew Ip > Email: aip at cwlinux.com > Tel: (852) 2542 2046 > Fax: (852) 2542 2036 > Mobile: (852) 9201 9866 > > Cwlinux Limited > Unit 202B 2/F Lai Cheong Factory Building, > 479-479A Castle Peak Road, > Lai Chi Kok, Kowloon, > Hong Kong. > > Tel: (852)2542 2046 > Fax: (852)2542 2036 > > For public pgp key, please obtain it from http://www.keyserver.net/en. > From rminnich at lanl.gov Wed Apr 23 13:24:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 13:24:00 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <012201c309c0$d5a77c40$253e41db@vsnl.net> Message-ID: On Wed, 23 Apr 2003, Deepak Kotian wrote: > May be in future , a smart > "POST code" could have some small internal buffering > with some kind of manual switch to move up and > down the display to track the code. > This may or may not sound sense , but just a thought. I think it would cost too much :-) ron From dkotian3 at vsnl.net Wed Apr 23 13:25:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 23 13:25:00 2003 Subject: Hammer Documentation is now Available at AMD's website References: <20030422154739.S14109-100000@www.missl.cs.umd.edu> Message-ID: <015501c309c1$db4b7740$253e41db@vsnl.net> Sorry. Could one please give the exact link/url where AMD has now publicly published the LinuxBIOS Port documentation. Did not find it easily. Regards Deepak ----- Original Message ----- From: "Adam Agnew" To: "Eric W. Biederman" Cc: "LinuxBIOS" Sent: Wednesday, April 23, 2003 1:23 AM Subject: Re: Hammer Documentation is now Available at AMD's website > > Perhaps the LinuxBIOS project can commit to supporting *every* AMD 64-bit > chipset? It seems like a good place to draw a line in the sand, as > information is starting out freely available. Thoughts? > > On 22 Apr 2003, Eric W. Biederman wrote: > > > So far I have just skimmed the Documentation but it appears that > > AMD has now publicly published everything needed to do Hammer > > LinuxBIOS port. > > > > No great surprise but it is nice to actually see it ! > > > > Yeah! The chaos with the Dual Athlons has finally paid off :) > > > > Eric > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From adam at cfar.umd.edu Wed Apr 23 13:43:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 23 13:43:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: Message-ID: <20030423141517.I18577-100000@www.missl.cs.umd.edu> > > > My understanding is that the Micro-2000 POST-Probe [5] is > > the only one > > > that can be powered from 3v3. > > > > > [5] "POST-Probe ", Micro2000, > > > It doesn't, and this is an important point that I am trying to clarify with > their engineers who claim that it will work from 3v3. Allright. As far as I can tell it is 5V card. The red PCI book Chapter 15 : MECHANICAL SPECIFICATION CONNECTOR PHYSICAL DESCRIPTION PG 471 [..] The same physical connector is used for the 32-bit signaling environment voltages. In one orientation the keys accept 5 volt boards. Rotated 180 degress, the connector accepts 3.3 volt signaling boards. The pin numbering of the connector changes (!) for the different signaling environemtns to maintain the same relative position of signals on the connector. For what it is worth, if anyone needs MiniPCI Port-80 boards: http://www.pctestpro.com/post/pcipost.htm (all the way at bottom) http://www.costronic.com/Ev014mp.htm http://siliconkit.dnsalias.com/cart/tpcim.html -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From dkotian3 at vsnl.net Wed Apr 23 14:11:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 23 14:11:01 2003 Subject: VIA C3 cpu config code in LinuxBIOS Message-ID: <02d701c309c8$4df4bda0$253e41db@vsnl.net> Hi, The VIA C3 processor code in LinuxBIOS has just invd statement in premtrr.inc in freebios/src/config/cpu/c3 directory. Is that it, just wanted to confirm on this .. Can someone please elaborate on this. Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From hansolofalcon at worldnet.att.net Wed Apr 23 14:27:00 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Wed Apr 23 14:27:00 2003 Subject: HTML mail on list Message-ID: <002d01c309ca$a72d1820$0100a8c0@who5> Hello again from Gregg C Levine Ron, folks, can we cease and desist using HTM mail on the list, except when necessary? Our friend David Woodhouse, who runs the Linux-MTD list, recently, had problems with such mail on the list, and had some comments on it I'd quote it but I'm better off posting his location here, http://www.infradead.org/~dwmw2/email.html that one describes everything. His list, also caught a large amount of junk, and, ah, he posted two messages to the list regarding such activities, in the form of a "rant". He called the message, "List etiquette.", even replied to himself which we all try not to do, to add to it, to add further complaints, which I'm not going echo here. David, I remember seeing a few of your posts here, do you mind my using the reference? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) From rminnich at lanl.gov Wed Apr 23 14:29:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 14:29:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030423141517.I18577-100000@www.missl.cs.umd.edu> Message-ID: On Wed, 23 Apr 2003, Adam Sulmicki wrote: > Allright. As far as I can tell it is 5V card. and there are now motherboards that won't power on with a 5V post card in ANY PCI slot. ron From rminnich at lanl.gov Wed Apr 23 14:32:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 23 14:32:00 2003 Subject: HTML mail on list In-Reply-To: <002d01c309ca$a72d1820$0100a8c0@who5> Message-ID: On Wed, 23 Apr 2003, Gregg C Levine wrote: > Ron, folks, can we cease and desist using HTM mail on the list, except > when necessary? I haven't noticed any lately, but I can try to filter it. ron From jerj at coplanar.net Wed Apr 23 15:36:01 2003 From: jerj at coplanar.net (Jeremy Jackson) Date: Wed Apr 23 15:36:01 2003 Subject: Debugging the linuxBIOS code References: <20030423005522.G14763-100000@www.missl.cs.umd.edu> Message-ID: <019a01c309d4$bfed42e0$7c07a8c0@kennet.coplanar.net> The one I have is from JDR Microdevices. It is probably 5V, but I'm mentioning it because I had a very hard time finding one this cheap, and I thought others (without the 3.3V requirement) may find it useful. http://www.jdr.com/interact/item.asp?itemno=gr-pcode-p Cheers, Jeremy ----- Original Message ----- From: "Adam Sulmicki" To: "Andrew Ip" Cc: ; Sent: Wednesday, April 23, 2003 12:56 AM Subject: Re: Debugging the linuxBIOS code > On Wed, 23 Apr 2003, Andrew Ip wrote: > > > POST card would be helpful if you don't have emulator. > > Speaking of which. Does anyone know 3.3V PORT-80 PCI cards instead of your > usual 5V ? (and preferably does not cost $100 either). From stepan at suse.de Wed Apr 23 16:59:00 2003 From: stepan at suse.de (Stefan Reinauer) Date: Wed Apr 23 16:59:00 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: <015501c309c1$db4b7740$253e41db@vsnl.net>; from dkotian3@vsnl.net on Wed, Apr 23, 2003 at 11:27:49PM +0530 References: <20030422154739.S14109-100000@www.missl.cs.umd.edu> <015501c309c1$db4b7740$253e41db@vsnl.net> Message-ID: <20030423233158.A1848@suse.de> * Deepak Kotian [030423 19:57]: > Could one please give the exact link/url where > AMD has now publicly published the LinuxBIOS > Port documentation. Did not find it easily. AMD Opteron Tech Docs: http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_9003,00.html AMD 8000(tm) Series Chipset Tech Docs: http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_9004,00.html AMD64 Tech Docs: http://www.amd.com/us-en/Processors/TechnicalResources/0,,30_182_739_7044,00.html Especially interesting for LinuxBIOS is definitely this one: http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/26094.pdf Best regards, Stefan Reinauer -- The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offense. -- E. W. Dijkstra From stepan at suse.de Wed Apr 23 18:04:00 2003 From: stepan at suse.de (Stefan Reinauer) Date: Wed Apr 23 18:04:00 2003 Subject: gcc 3.3 and coreutils Message-ID: <20030424003727.A2662@suse.de> Hi LinuxBIOS hackers, while working on a port to the freshly released AMD64 Opteron systems I fixed a couple of warnings that probably occur due to stricter gcc 3.x syntax checks. This includes type signedness checking and stricter string concatenation. Also I noticed that coreutils 5.0 are stricter with their POSIX compliance in a way that breaks LinuxBIOS scripts. tail -5 has to be tail -n 5, head accordingly. I have not explicitly tested older compilers, but the changes are harmless and unlikely to break anything. Can this be included? Comments? Stefan -- The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offense. -- E. W. Dijkstra -------------- next part -------------- diff -urN freebios/src/arch/i386/boot/linuxbios_table.c freebios.hammer/src/arch/i386/boot/linuxbios_table.c --- freebios/src/arch/i386/boot/linuxbios_table.c 2003-01-28 21:03:28.000000000 +0100 +++ freebios.hammer/src/arch/i386/boot/linuxbios_table.c 2003-04-11 17:36:04.000000000 +0200 @@ -115,11 +115,11 @@ { LB_TAG_LINKER, linuxbios_linker, }, { LB_TAG_ASSEMBLER, linuxbios_assembler, }, }; - int i; + unsigned int i; for(i = 0; i < sizeof(strings)/sizeof(strings[0]); i++) { struct lb_string *rec; size_t len; - rec = lb_new_record(header); + rec = (struct lb_string *)lb_new_record(header); len = strlen(strings[i].string); rec->tag = strings[i].tag; rec->size = (sizeof(*rec) + len + 1 + 3) & ~3; diff -urN freebios/src/arch/i386/lib/c_start.S freebios.hammer/src/arch/i386/lib/c_start.S --- freebios/src/arch/i386/lib/c_start.S 2002-10-28 08:41:54.000000000 +0100 +++ freebios.hammer/src/arch/i386/lib/c_start.S 2003-04-11 18:13:13.000000000 +0200 @@ -10,7 +10,7 @@ cli lgdt %cs:gdtaddr ljmp $0x10, $1f -1: movl $0x18, %ax +1: movl $0x18, %eax movl %eax, %ds movl %eax, %es movl %eax, %ss diff -urN freebios/src/arch/i386/smp/ioapic.c freebios.hammer/src/arch/i386/smp/ioapic.c --- freebios/src/arch/i386/smp/ioapic.c 2001-11-03 03:11:48.000000000 +0100 +++ freebios.hammer/src/arch/i386/smp/ioapic.c 2003-04-11 18:13:42.000000000 +0200 @@ -54,7 +54,7 @@ void setup_ioapic(void) { - int i; + unsigned int i; unsigned long value_low, value_high; unsigned long nvram = 0xfec00000; volatile unsigned long *l; diff -urN freebios/src/config/Config freebios.hammer/src/config/Config --- freebios/src/config/Config 2003-04-17 04:12:58.000000000 +0200 +++ freebios.hammer/src/config/Config 2003-04-17 04:20:15.000000000 +0200 @@ -20,8 +20,8 @@ option LINUXBIOS_COMPILE_BY = $(shell whoami) option LINUXBIOS_COMPILE_HOST = $(shell hostname) option LINUXBIOS_COMPILE_DOMAIN = $(shell dnsdomainname) -option LINUXBIOS_COMPILER = $(shell $(CC) $(CFLAGS) -v 2>&1 | tail -1) -option LINUXBIOS_LINKER = $(shell $(CC) -Wl,-v 2>&1 | grep version | tail -1) +option LINUXBIOS_COMPILER = $(shell $(CC) $(CFLAGS) -v 2>&1 | tail -n 1) +option LINUXBIOS_LINKER = $(shell $(CC) -Wl,-v 2>&1 | grep version | tail -n 1) option LINUXBIOS_ASSEMBLER = $(shell touch dummy.s ; $(CC) -c -Wa,-v dummy.s 2>&1; rm -f dummy.s dummy.o ) makerule ldscript.ld : ldoptions $(LDSUBSCRIPTS-1) ; echo "INCLUDE ldoptions" > $@ ; for file in $(LDSUBSCRIPTS-1) ; do echo "INCLUDE $$file" >> $@ ; done diff -urN freebios/src/cpu/p5/delay_tsc.c freebios.hammer/src/cpu/p5/delay_tsc.c --- freebios/src/cpu/p5/delay_tsc.c 2002-10-26 02:49:25.000000000 +0200 +++ freebios.hammer/src/cpu/p5/delay_tsc.c 2003-04-11 18:19:17.000000000 +0200 @@ -99,10 +99,10 @@ static unsigned long long calibrate_tsc(void) { unsigned long long start, end, delta; - unsigned long allones = (unsigned long) -1, result; + // unsigned long allones = (unsigned long) -1; unsigned long startlow, starthigh; unsigned long endlow, endhigh; - unsigned long count; + unsigned long count, result; rdtsc(startlow,starthigh); // no udivdi3, dammit. @@ -129,7 +129,7 @@ endhigh, endlow, starthigh, startlow); printk_spew("32-bit delta %d\n", (unsigned long) delta); - printk_spew(__FUNCTION__ " 32-bit result is %d\n", result); + printk_spew("%s 32-bit result is %d\n", __FUNCTION__, result); return delta; } diff -urN freebios/src/cpu/p6/mtrr.c freebios.hammer/src/cpu/p6/mtrr.c --- freebios/src/cpu/p6/mtrr.c 2003-01-28 21:04:18.000000000 +0100 +++ freebios.hammer/src/cpu/p6/mtrr.c 2003-04-11 17:53:35.000000000 +0200 @@ -129,8 +129,6 @@ /* setting variable mtrr, comes from linux kernel source */ void set_var_mtrr(unsigned int reg, unsigned long base, unsigned long size, unsigned char type) { - unsigned int tmp; - if (reg >= 8) return; diff -urN freebios/src/lib/clog2.c freebios.hammer/src/lib/clog2.c --- freebios/src/lib/clog2.c 2002-07-19 23:06:42.000000000 +0200 +++ freebios.hammer/src/lib/clog2.c 2003-04-11 18:03:06.000000000 +0200 @@ -7,8 +7,7 @@ unsigned long pow = sizeof(x) * 8 - 1; if (! x) { - printk_emerg(__FUNCTION__ - " called with invalid parameter of 0\n"); + printk_emerg("%s called with invalid parameter of 0\n",__FUNCTION__); return -1; } for(; i > x; i >>= 1, pow--) diff -urN freebios/src/lib/elfboot.c freebios.hammer/src/lib/elfboot.c --- freebios/src/lib/elfboot.c 2003-01-28 21:04:44.000000000 +0100 +++ freebios.hammer/src/lib/elfboot.c 2003-04-11 18:03:58.000000000 +0200 @@ -109,7 +109,7 @@ unsigned long lb_size; unsigned long mem_entries; unsigned long buffer; - int i; + unsigned int i; lb_size = (unsigned long)(&_eram_seg - &_ram_seg); /* Double linuxBIOS size so I have somewhere to place a copy to return to */ lb_size = lb_size + lb_size; @@ -207,7 +207,7 @@ * the segment that was passed in is completely contained * in RAM. */ - int i; + unsigned int i; unsigned long end = start + len; unsigned long mem_entries = (mem->size - sizeof(*mem))/sizeof(mem->map[0]); @@ -424,7 +424,7 @@ struct segment *head, struct stream *stream, unsigned char *header, unsigned long header_size) { - unsigned long offset; + long offset; struct segment *ptr; offset = 0; @@ -598,8 +598,8 @@ { Elf_ehdr *ehdr; static unsigned char header[ELF_HEAD_SIZE]; - int header_offset; - int i, result; + int header_offset, result; + unsigned int i; result = 0; printk_info("\n"); diff -urN freebios/src/lib/malloc.c freebios.hammer/src/lib/malloc.c --- freebios/src/lib/malloc.c 2002-02-05 01:06:20.000000000 +0100 +++ freebios.hammer/src/lib/malloc.c 2003-04-11 17:59:52.000000000 +0200 @@ -28,7 +28,7 @@ { void *p; - MALLOCDBG((__FUNCTION__ " Enter, size %d, free_mem_ptr %p\n", size, free_mem_ptr)); + MALLOCDBG(("%s Enter, size %d, free_mem_ptr %p\n", __FUNCTION__, size, free_mem_ptr)); if (size < 0) error("Error! malloc: Size < 0"); if (free_mem_ptr <= 0) diff -urN freebios/src/lib/memcpy.c freebios.hammer/src/lib/memcpy.c --- freebios/src/lib/memcpy.c 2001-03-13 05:22:19.000000000 +0100 +++ freebios.hammer/src/lib/memcpy.c 2003-04-11 17:58:58.000000000 +0200 @@ -1,7 +1,7 @@ #include void *memcpy(void *__dest, __const void *__src, size_t __n) { - int i; + unsigned int i; char *d = (char *) __dest, *s = (char *) __src; for (i = 0; i < __n; i++) diff -urN freebios/src/lib/memset.c freebios.hammer/src/lib/memset.c --- freebios/src/lib/memset.c 2001-03-13 05:22:19.000000000 +0100 +++ freebios.hammer/src/lib/memset.c 2003-04-11 17:58:43.000000000 +0200 @@ -2,7 +2,7 @@ void *memset(void *s, int c, size_t n) { - int i; + unsigned int i; char *ss = (char *) s; for (i = 0; i < n; i++) diff -urN freebios/src/lib/newpci.c freebios.hammer/src/lib/newpci.c --- freebios/src/lib/newpci.c 2003-04-17 04:13:12.000000000 +0200 +++ freebios.hammer/src/lib/newpci.c 2003-04-17 22:42:50.000000000 +0200 @@ -439,7 +439,7 @@ struct resource *result = 0; int seen_last = 0; for(curdev = bus->children; curdev; curdev = curdev->sibling) { - int i; + unsigned int i; for(i = 0; i < curdev->resources; i++) { struct resource *resource = &curdev->resource[i]; /* If it isn't the right kind of resource ignore it */ diff -urN freebios/src/pc80/mc146818rtc.c freebios.hammer/src/pc80/mc146818rtc.c --- freebios/src/pc80/mc146818rtc.c 2003-01-28 21:07:09.000000000 +0100 +++ freebios.hammer/src/pc80/mc146818rtc.c 2003-04-11 18:21:11.000000000 +0200 @@ -136,7 +136,7 @@ PC_CKS_RANGE_END,PC_CKS_LOC); if (invalid || cmos_invalid || checksum_invalid) { - int i; + unsigned int i; printk_warning("RTC:%s%s%s zeroing cmos\n", invalid?" Clear requested":"", cmos_invalid?" Power Problem":"", diff -urN freebios/util/nrv2b/nrv2b.c freebios.hammer/util/nrv2b/nrv2b.c --- freebios/util/nrv2b/nrv2b.c 2002-10-26 01:06:54.000000000 +0200 +++ freebios.hammer/util/nrv2b/nrv2b.c 2003-03-19 16:37:08.000000000 +0100 @@ -783,8 +783,7 @@ break; if (in != m) if (memcmp(in,ip,c->m_len+1) == 0) - printf("%p %p %p -%5d\n",in,ip,m,c->m_len); + printf("%p %p %p\n%5d\n",in,ip,m,c->m_len); in++; } @@ -1445,11 +1444,9 @@ rewind(infile = f); } else if (argc != 4) { - Fprintf((stderr, "'lzhuf e file1 file2' encodes file1 into -file2.\n" + Fprintf((stderr, "'lzhuf e file1 file2' encodes file1 into file2.\n" - "'lzhuf d file2 file1' decodes file2 into -file1.\n")); + "'lzhuf d file2 file1' decodes file2 into file1.\n")); return EXIT_FAILURE; } diff -urN freebios/util/options/build_opt_tbl.c freebios.hammer/util/options/build_opt_tbl.c --- freebios/util/options/build_opt_tbl.c 2003-01-28 21:08:09.000000000 +0100 +++ freebios.hammer/util/options/build_opt_tbl.c 2003-04-11 17:52:04.000000000 +0200 @@ -26,7 +26,7 @@ void test_for_entry_overlaps(int entry_start,int entry_end) { int ptr; - int buffer_bit_size; + unsigned int buffer_bit_size; int offset; int byte; int byte_length; @@ -74,7 +74,7 @@ } } else { /* test if bits overlap byte boundaries */ - if(ce->length>(8-offset)) { + if((int)ce->length>(8-offset)) { printf("Error - Entry %s length overlaps a byte boundry\n", ce->name); exit(1); } @@ -142,7 +142,7 @@ */ int main(int argc, char **argv) { - int i; + unsigned int i; char *config=0; char *option=0; FILE *fp; @@ -163,7 +163,7 @@ int len; unsigned char buf[16]; - for(i=1;i References: Message-ID: <3EA72B34.30808@bitworks.com> I've just discovered a compact flash that isn't identified properly by the polling IDE code. Its an 8Meg CF and Linux identifies it as a Hitachi CV 7.1.1, ATA DISK drive with CHS=246/2/32 The polling IDE code sees that there is a drive there issues the info command and dosen't generate an error but the model info is blank and the CHS info shows all zeros. Same code seems to work fine with the SanDisk and Lexar CFs I have. So it looks like the IDE_CMD_GET_INFO command is returning back all zeros. If that's the case then where is Linux getting its info from? -- Richard A. Smith rsmith at bitworks.com From agnew at cs.umd.edu Wed Apr 23 21:08:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Wed Apr 23 21:08:01 2003 Subject: Polling IDE and CF issue. In-Reply-To: <3EA72B34.30808@bitworks.com> Message-ID: <20030423214618.I20517-100000@www.missl.cs.umd.edu> Sorry I have to ask, but.. which polled ide code are you talking about? LinuxBIOS native, Etherboot 5.0.6 or Etherboot 5.1.7? Thanks - Adam Agnew On Wed, 23 Apr 2003, Richard Smith wrote: > I've just discovered a compact flash that isn't identified properly by > the polling IDE code. > > Its an 8Meg CF and Linux identifies it as a Hitachi CV 7.1.1, ATA DISK > drive with CHS=246/2/32 > > The polling IDE code sees that there is a drive there issues the info > command and dosen't generate an error but the model info is blank and > the CHS info shows all zeros. Same code seems to work fine with the > SanDisk and Lexar CFs I have. > > So it looks like the IDE_CMD_GET_INFO command is returning back all zeros. > > If that's the case then where is Linux getting its info from? > > -- > Richard A. Smith > rsmith at bitworks.com > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From ebiederman at lnxi.com Wed Apr 23 22:17:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 22:17:01 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: <200304231727.h3NHRFqV031140@tenor.codegen.com> References: <200304231727.h3NHRFqV031140@tenor.codegen.com> Message-ID: "Thomas J. Merritt" writes: > |<><><><><> Original message from Eric W. Biederman <><><><><> > |So far I have just skimmed the Documentation but it appears that > |AMD has now publicly published everything needed to do Hammer > |LinuxBIOS port. > | > |No great surprise but it is nice to actually see it ! > | > |Yeah! The chaos with the Dual Athlons has finally paid off :) > > It does appear that they have released all of the documentation > necessary. The learning curve is pretty steep though. There are > a lot of registers to program to make a box sing. We have been > able to bring SmartFirmware up on our AMD Solo box. It has taken > about 5 months to get to the point that we can now boot NetBSD. > We're currently unable to boot Linux, but making progress on that front. Hmm. I wonder if that is because of the open firmware chaos. So far I have not seen anything out of the ordinary in the way of a learning curve, and I just booted a linux kernel. > On the LinuxBIOS front we're planning to leverage our x86-64 experience > and provide commercial support for LinuxBIOS on AMD64 processors. > With AMD's public release of the processor and chipset documentation we > should be releived of most of our NDA obligations, and this will now > become possible. If anyone has hardware and is contemplating bring up > LinuxBIOS please let me know, since I don't want to duplicate effort if > we can avoid it. You have heard the talk about the freebios2 tree? Anyway Stefan Reinauer at SuSe and I are busily bringing it up. So far so good. We are still in the stage where you hard code things, so you know you understand what registers need to be programmed, but everything has been straight forward so far. Eric From aip at cwlinux.com Wed Apr 23 22:50:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Wed Apr 23 22:50:01 2003 Subject: Polling IDE and CF issue. In-Reply-To: <3EA72B34.30808@bitworks.com>; from Richard Smith on Wed, Apr 23, 2003 at 07:09:24PM -0500 References: <3EA72B34.30808@bitworks.com> Message-ID: <20030424112349.A21650@mail.cwlinux.com> Richard, > I've just discovered a compact flash that isn't identified properly by > the polling IDE code. > Its an 8Meg CF and Linux identifies it as a Hitachi CV 7.1.1, ATA DISK > drive with CHS=246/2/32 > The polling IDE code sees that there is a drive there issues the info > command and dosen't generate an error but the model info is blank and > the CHS info shows all zeros. Same code seems to work fine with the > SanDisk and Lexar CFs I have. > So it looks like the IDE_CMD_GET_INFO command is returning back all zeros. > If that's the case then where is Linux getting its info from? I have noticed that some CF's require more delay than others. For instance, I had to increase WAIT_COUNT from 0xFFFF to 0xFFFFFF in Adam's ide patch in order to get certain card to work. Hope this help. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From ebiederman at lnxi.com Wed Apr 23 22:55:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 22:55:01 2003 Subject: Polling IDE and CF issue. In-Reply-To: <20030423214618.I20517-100000@www.missl.cs.umd.edu> References: <20030423214618.I20517-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > Sorry I have to ask, but.. which polled ide code are you talking about? > LinuxBIOS native, Etherboot 5.0.6 or Etherboot 5.1.7? Given that the mentioned command does not occur in etherboot 5.1.7 it sounds like the builtin IDE driver. Richard please try the etherboot one. Ron was looking at backporting it. There are no magic delay funnies and the code has been tested with quite a few drives. Eric From ebiederman at lnxi.com Wed Apr 23 22:59:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 22:59:00 2003 Subject: Subsystem ID In-Reply-To: References: Message-ID: "Shubhangi Jadhav" writes: > Hi, > > Can we map the device id from lspci output be mapped to the Subsystem > ID - > > the SID register that needs to be set while initializing any chip on the > motherboard. This question at least with context makes no sense to me. Eric From ebiederman at lnxi.com Wed Apr 23 23:08:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 23 23:08:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> References: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> Message-ID: dkotian3 at vsnl.net writes: > Hi, > > Is there a recommended way of debugging LinuxBIOS code. > > I burnt my FLASHROM with LinuxBIOS, nothing came > up at all. Not a single message came at the serial Output. > I knew that my North bridge code has issues, but still > something should have come. I guess.. > > I had connected the serial out of my PC to another PC > and I am using an Windows HyperTerminal to view > the messages. > Should it not show something at all even if north bridge code > is broken. > > Could anyone give their views on this one.. If there is not a easy solution review the code generated and see if it is initializing your serial port correctly. It is the insane way but on known good hardware it tends to work well. Oh, and check your baud rate settings.. Eric From bari at onelabs.com Thu Apr 24 01:06:01 2003 From: bari at onelabs.com (Bari Ari) Date: Thu Apr 24 01:06:01 2003 Subject: Debugging the linuxBIOS code References: <20030423141517.I18577-100000@www.missl.cs.umd.edu> Message-ID: <3EA77885.4070208@onelabs.com> Adam Sulmicki wrote: >For what it is worth, if anyone needs MiniPCI Port-80 boards: > > http://www.pctestpro.com/post/pcipost.htm (all the way at bottom) > > http://www.costronic.com/Ev014mp.htm > > http://siliconkit.dnsalias.com/cart/tpcim.html > > > Another option is to use a miniPCI to PCI adapter which you can plug your regular 3.3V PCI POST card into. http://www.interfacemasters.com/products/pci_tools/mini_pci_to_pci/index.html --Bari From adam at cfar.umd.edu Thu Apr 24 01:18:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Thu Apr 24 01:18:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <3EA77885.4070208@onelabs.com> Message-ID: <20030424015619.G21280-100000@www.missl.cs.umd.edu> > >For what it is worth, if anyone needs MiniPCI Port-80 boards: > > http://www.pctestpro.com/post/pcipost.htm (all the way at bottom) > > http://www.costronic.com/Ev014mp.htm > > http://siliconkit.dnsalias.com/cart/tpcim.html > Another option is to use a miniPCI to PCI adapter which you can plug > your regular 3.3V PCI POST card into. > > http://www.interfacemasters.com/products/pci_tools/mini_pci_to_pci/index.html yeah, with the $65'000 question being where do I find such 3.3V POST card. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From ebiederman at lnxi.com Thu Apr 24 01:57:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 24 01:57:01 2003 Subject: Hammer Documentation is now Available at AMD's website In-Reply-To: <200304240416.h3O4GcqV037513@tenor.codegen.com> References: <200304240416.h3O4GcqV037513@tenor.codegen.com> Message-ID: "Thomas J. Merritt" writes: > |<><><><><> Original message from Eric W. Biederman <><><><><> > |"Thomas J. Merritt" writes: > | > |> |<><><><><> Original message from Eric W. Biederman <><><><><> > |> |So far I have just skimmed the Documentation but it appears that > |> |AMD has now publicly published everything needed to do Hammer > |> |LinuxBIOS port. > |> | > |> |No great surprise but it is nice to actually see it ! > |> | > |> |Yeah! The chaos with the Dual Athlons has finally paid off :) > |> > |> It does appear that they have released all of the documentation > |> necessary. The learning curve is pretty steep though. There are > |> a lot of registers to program to make a box sing. We have been > |> able to bring SmartFirmware up on our AMD Solo box. It has taken > |> about 5 months to get to the point that we can now boot NetBSD. > |> We're currently unable to boot Linux, but making progress on that front. > | > |Hmm. I wonder if that is because of the open firmware chaos. > > Open Firmware is pretty much out of the picture when the kernel takes > over. The primary problem is the wired assumptions in the legacy PC > chaos. I guess that is what I meant the chaos of having Open Firmware or anything else besides exactly what the kernel was expecting. > |So far I have not seen anything out of the ordinary in the way of > |a learning curve, and I just booted a linux kernel. > > Keep in mind that our background is primarily PowerPC, Sparc, MIPS and ARM > processors and we've avoided x86 for the most part. Learning the > Hammer side of things has been relatively easy, it has been more > painful learning the legacy PC stuff that hangs out in the 8111 > I/O Hub. Not a problem. The comment was meant more for than you to clarify that the Hammer was exceptionally complex, for an x86 cpu. > |> On the LinuxBIOS front we're planning to leverage our x86-64 experience > |> and provide commercial support for LinuxBIOS on AMD64 processors. > |> With AMD's public release of the processor and chipset documentation we > |> should be releived of most of our NDA obligations, and this will now > |> become possible. If anyone has hardware and is contemplating bring up > |> LinuxBIOS please let me know, since I don't want to duplicate effort if > |> we can avoid it. > | > |You have heard the talk about the freebios2 tree? > | > |Anyway Stefan Reinauer at SuSe and I are busily bringing it up. > |So far so good. We are still in the stage where you hard code things, > |so you know you understand what registers need to be programmed, > |but everything has been straight forward so far. > > As straight forward as a PC can I would say. Given that you have > experience with prior AMD chipsets, the Hammer should look pretty > familiar. We've reviewed about 5000 pages of documentation in the > last several months and have most of it digested now. Yep. There can certainly be a lot to review and digest if you are new to x86. The only good thing is that basic assumption stay the same for a long time. Eric From yenneo at oreka.com Thu Apr 24 01:59:00 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Thu Apr 24 01:59:00 2003 Subject: help : flash an epia Message-ID: <200304240632.IAA93577@mailhub5.isdnet.net> I want to load linuxbios on an epia board. When i try to execute the flash_rom utility it compile the different file and give me this mistake: flash_rom.c:35: pci/pci.h: No such file or directory make: *** [all] Error 1 I?ve put rhe romimage into the directory /util/flash_and_burn and executed the makefile I also tried the command "setpci -s 0:11.0 40.b=3D54" but it gives an error of syntax... So how should i do? Thanks im advance. -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From aip at cwlinux.com Thu Apr 24 02:31:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Thu Apr 24 02:31:01 2003 Subject: help : flash an epia In-Reply-To: <200304240632.IAA93577@mailhub5.isdnet.net>; from Philippe CABANNES on Thu, Apr 24, 2003 at 08:24:27AM +0100 References: <200304240632.IAA93577@mailhub5.isdnet.net> Message-ID: <20030424150421.A25586@mail.cwlinux.com> Philippe, > I want to load linuxbios on an epia board. > When i try to execute the flash_rom utility it compile the different > file and give me this mistake: > flash_rom.c:35: pci/pci.h: No such file or directory > make: *** [all] Error 1 You need pciutils-dev package. > I?ve put rhe romimage into the directory /util/flash_and_burn and > executed the makefile > I also tried the command "setpci -s 0:11.0 40.b=3D54" but it gives an > error of syntax... > So how should i do? In the README of via/epia, it says $ setpci -s 0:11.0 40.b=54 Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From bari at onelabs.com Thu Apr 24 02:35:01 2003 From: bari at onelabs.com (Bari Ari) Date: Thu Apr 24 02:35:01 2003 Subject: Debugging the linuxBIOS code References: <20030424015619.G21280-100000@www.missl.cs.umd.edu> Message-ID: <3EA78D85.4010908@onelabs.com> Adam Sulmicki wrote: > >yeah, with the $65'000 question being where do I find such 3.3V POST card. > > > We build lots of our tools. Check with NSC. They have a PCIWatcher Debug Card that has a LCD display and has buffers to pass the data to a RS-232 serial port. It captures not only port 80/84 but all PCI I/O, memory, cinfig, INTacks, etc. --Bari From yapeehuey at hotmail.com Thu Apr 24 08:08:01 2003 From: yapeehuey at hotmail.com (Yap Ee Huey) Date: Thu Apr 24 08:08:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image Message-ID: Hi hackers : I am a little bit confused here after building linuxbios, etherboot and making a tagged kernel image with mknbi. Hopefully I can get some help here : I have made ide_disk.elf, (22k) I tag bzImage with mknbi-linux, (888k) I have built linuxbios.rom (64k) I have 256k bios, so the kernel should be put somewhere on my harddisk. If i am not wrong my ide_disk.elf should be placed at the first sector of my bios rom. My question is: where should I put my Kernel Image and linuxbios.rom so that ide_disk.elf will know where to get and load them ? Do I need to modify the -DDEFAULT_BOOTFILE option which defaulted to tftp:///tftpboot/kernel ? Thanks a lot ! best regards, eehuey _________________________________________________________________ Are you in love? Find a date on MSN Personals http://match.msn.com.my/ From shubhangi.jadhav at patni.com Thu Apr 24 08:10:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Thu Apr 24 08:10:00 2003 Subject: linuxbios problem Message-ID: Hi, I'm trying to get my motherboard working with Linuxbios. I'm using ADLO with Linuxbios. The debug messages are as below. The problem is after jumping to linuxbiosmain (it prints "Welcome to start32......") it loops back , restarting the boot process all over again. what's wrong ? why does it go in a loop? ***************log********************************* LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 Ram2 Ram3 Ram Enable 1 Ram Enable 2 Ram Enable 3 Ram Enable 4 Ram Enable 5 Ram4 Ram5 Ram6 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 booting... Finding PCI configuration type. PCI: Using configuration type 1 handle_superio start, nsuperio 1 handle_superio Pass 0, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 0, Superio w83627hf handle_superio port 0x0, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 0, done #0 handle_superio done Scanning PCI bus...PCI: pci_scan_bus for bus 0 PCI: 00:00.0 [8086/7124] PCI: 00:1e.0 [8086/2418] PCI: 00:1f.0 [8086/2410] PCI: 00:1f.1 [8086/2411] PCI: 00:1f.2 [8086/2412] PCI: 00:1f.3 [8086/2413] PCI: 00:1f.5 [8086/2415] PCI: 00:1f.6 [8086/2416] PCI: pci_scan_bus for bus 1 PCI: 01:04.0 [10ec/8139] PCI: 01:05.0 [10ec/8139] PCI: pci_scan_bus returning with max=01 PCI: pci_scan_bus returning with max=01 done Allocating PCI resources... ASSIGN RESOURCES, bus 0 PCI: 00:1e.0 1c <- [0x00001000 - 0x00001fff] bus 1 io PCI: 00:1e.0 24 <- [0xfec00000 - 0xfebfffff] bus 1 prefmem PCI: 00:1e.0 20 <- [0xfeb00000 - 0xfebfffff] bus 1 mem ASSIGN RESOURCES, bus 1 PCI: 01:04.0 10 <- [0x00001000 - 0x000010ff] io PCI: 01:04.0 14 <- [0xfeb00000 - 0xfeb000ff] mem PCI: 01:05.0 10 <- [0x00001400 - 0x000014ff] io PCI: 01:05.0 14 <- [0xfeb01000 - 0xfeb010ff] mem ASSIGNED RESOURCES, bus 1 PCI: 00:1f.1 20 <- [0x000028e0 - 0x000028ef] io PCI: 00:1f.2 20 <- [0x000028c0 - 0x000028df] io PCI: 00:1f.3 20 <- [0x000028f0 - 0x000028ff] io PCI: 00:1f.5 10 <- [0x00002000 - 0x000020ff] io PCI: 00:1f.5 14 <- [0x00002880 - 0x000028bf] io PCI: 00:1f.6 10 <- [0x00002400 - 0x000024ff] io PCI: 00:1f.6 14 <- [0x00002800 - 0x0000287f] io ASSIGNED RESOURCES, bus 0 done. Enabling PCI resourcess...PCI: 00:00.0 cmd <- 06 PCI: 00:1e.0 cmd <- 07 PCI: 00:1f.0 cmd <- 0f PCI: 00:1f.1 cmd <- 01 PCI: 00:1f.2 cmd <- 01 PCI: 00:1f.3 cmd <- 01 PCI: 00:1f.5 cmd <- 01 PCI: 00:1f.6 cmd <- 01 PCI: 01:04.0 cmd <- 03 PCI: 01:05.0 cmd <- 03 done. Initializing PCI devices... PCI devices initialized DRP0 = 0x9 DIMM0 - size = 64M DIMM1 - size = 0M DRP1 = 0x0 DIMM2 - size = 0M totalram: 64M Initializing CPU #0 Updating microcode microcode_info: sig = 0x00000678 pf=0x00000001 rev = 0x00000000 Enabling cache... Setting fixed MTRRs(0-88) type: UC Setting fixed MTRRs(0-16) type: WB DONE fixed MTRRs Setting variable MTRR 0, base: 0MB, range: 64MB, type WB DONE variable MTRRs Clear out the extra MTRR's call intel_enable_fixed_mtrr() call intel_enable_var_mtrr() Leave setup_mtrrs done. Max cpuid index : 1 Vendor ID : CentaurHauls Processor Type : 0x00 Processor Family : 0x06 Processor Model : 0x07 Processor Mask : 0x00 Processor Stepping : 0x08 Feature flags : 0x00803035 MTRR check Fixed MTRRs : Enabled Variable MTRRs: Enabled Configuring L2 cache...Not 'GenuineIntel' Processor Enable Cache done. Disabling local apic...done. CPU #0 Initialized ioapic southbridge enabled 180 RTC Init Invalid CMOS LB checksum set power on after power fail Please turn on nvram handle_superio start, nsuperio 1 handle_superio Pass 1, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 1, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e Call init Enabling com device: 03 iobase = 0x02f8 irq=3 handle_superio Pass 1, done #0 handle_superio done handle_superio start, nsuperio 1 handle_superio Pass 2, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 2, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 2, done #0 handle_superio done Wrote linuxbios table at: 00000500 - 00000664 checksum 7535 Jumping to linuxbiosmain()... Welcome to start32, the open sourced starter. This space will eventually hold more diagnostic information. January 2000, James Hendricks, Dale Webster, and Ron Minnich. Version 0.1 LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 Ram2 Ram3 Ram Enable 1 Ram Enable 2 Ram Enable 3 Ram Enable 4 Ram Enable 5 Ram4 Ram5 Ram6 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. ***************log********************************* Regards, Shubhangi From shubhangi.jadhav at patni.com Thu Apr 24 08:29:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Thu Apr 24 08:29:00 2003 Subject: Delay in copying Linuxbios to ram Message-ID: Hi, While trying to boot using linuxbios , I get a message " Copying LinuxBIOS to ram " and then the system pauses for 3-4 secs, before printing the next message - "Jumping to Linuxbios" Could someone tell me what causes this delay, does the process of uncompressing and copying linuxbios to RAM take this long? Any suggestions on how to reduce this delay? Here's the log of the debug messages. ***************log********************************* LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 Ram2 Ram3 Ram Enable 1 Ram Enable 2 Ram Enable 3 Ram Enable 4 Ram Enable 5 Ram4 Ram5 Ram6 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 booting... Finding PCI configuration type. PCI: Using configuration type 1 handle_superio start, nsuperio 1 handle_superio Pass 0, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 0, Superio w83627hf handle_superio port 0x0, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 0, done #0 handle_superio done Scanning PCI bus...PCI: pci_scan_bus for bus 0 PCI: 00:00.0 [8086/7124] PCI: 00:1e.0 [8086/2418] PCI: 00:1f.0 [8086/2410] PCI: 00:1f.1 [8086/2411] PCI: 00:1f.2 [8086/2412] PCI: 00:1f.3 [8086/2413] PCI: 00:1f.5 [8086/2415] PCI: 00:1f.6 [8086/2416] PCI: pci_scan_bus for bus 1 PCI: 01:04.0 [10ec/8139] PCI: 01:05.0 [10ec/8139] PCI: pci_scan_bus returning with max=01 PCI: pci_scan_bus returning with max=01 done Allocating PCI resources... ASSIGN RESOURCES, bus 0 PCI: 00:1e.0 1c <- [0x00001000 - 0x00001fff] bus 1 io PCI: 00:1e.0 24 <- [0xfec00000 - 0xfebfffff] bus 1 prefmem PCI: 00:1e.0 20 <- [0xfeb00000 - 0xfebfffff] bus 1 mem ASSIGN RESOURCES, bus 1 PCI: 01:04.0 10 <- [0x00001000 - 0x000010ff] io PCI: 01:04.0 14 <- [0xfeb00000 - 0xfeb000ff] mem PCI: 01:05.0 10 <- [0x00001400 - 0x000014ff] io PCI: 01:05.0 14 <- [0xfeb01000 - 0xfeb010ff] mem ASSIGNED RESOURCES, bus 1 PCI: 00:1f.1 20 <- [0x000028e0 - 0x000028ef] io PCI: 00:1f.2 20 <- [0x000028c0 - 0x000028df] io PCI: 00:1f.3 20 <- [0x000028f0 - 0x000028ff] io PCI: 00:1f.5 10 <- [0x00002000 - 0x000020ff] io PCI: 00:1f.5 14 <- [0x00002880 - 0x000028bf] io PCI: 00:1f.6 10 <- [0x00002400 - 0x000024ff] io PCI: 00:1f.6 14 <- [0x00002800 - 0x0000287f] io ASSIGNED RESOURCES, bus 0 done. Enabling PCI resourcess...PCI: 00:00.0 cmd <- 06 PCI: 00:1e.0 cmd <- 07 PCI: 00:1f.0 cmd <- 0f PCI: 00:1f.1 cmd <- 01 PCI: 00:1f.2 cmd <- 01 PCI: 00:1f.3 cmd <- 01 PCI: 00:1f.5 cmd <- 01 PCI: 00:1f.6 cmd <- 01 PCI: 01:04.0 cmd <- 03 PCI: 01:05.0 cmd <- 03 done. Initializing PCI devices... PCI devices initialized DRP0 = 0x9 DIMM0 - size = 64M DIMM1 - size = 0M DRP1 = 0x0 DIMM2 - size = 0M totalram: 64M Initializing CPU #0 Updating microcode microcode_info: sig = 0x00000678 pf=0x00000001 rev = 0x00000000 Enabling cache... Setting fixed MTRRs(0-88) type: UC Setting fixed MTRRs(0-16) type: WB DONE fixed MTRRs Setting variable MTRR 0, base: 0MB, range: 64MB, type WB DONE variable MTRRs Clear out the extra MTRR's call intel_enable_fixed_mtrr() call intel_enable_var_mtrr() Leave setup_mtrrs done. Max cpuid index : 1 Vendor ID : CentaurHauls Processor Type : 0x00 Processor Family : 0x06 Processor Model : 0x07 Processor Mask : 0x00 Processor Stepping : 0x08 Feature flags : 0x00803035 MTRR check Fixed MTRRs : Enabled Variable MTRRs: Enabled Configuring L2 cache...Not 'GenuineIntel' Processor Enable Cache done. Disabling local apic...done. CPU #0 Initialized ioapic southbridge enabled 180 RTC Init Invalid CMOS LB checksum set power on after power fail Please turn on nvram handle_superio start, nsuperio 1 handle_superio Pass 1, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 1, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e Call init Enabling com device: 03 iobase = 0x02f8 irq=3 handle_superio Pass 1, done #0 handle_superio done handle_superio start, nsuperio 1 handle_superio Pass 2, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 2, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 2, done #0 handle_superio done Wrote linuxbios table at: 00000500 - 00000664 checksum 7535 Jumping to linuxbiosmain()... Welcome to start32, the open sourced starter. This space will eventually hold more diagnostic information. January 2000, James Hendricks, Dale Webster, and Ron Minnich. Version 0.1 Regards, Shubhangi From shubhangi.jadhav at patni.com Thu Apr 24 08:45:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Thu Apr 24 08:45:00 2003 Subject: setting USE_ELF_BOOT Message-ID: Hi, When do we need to set USE_ELF_BOOT option ? Is elfboot loader required to load ADLO ? or is it required only in case of etherboot? Regards, Shubhangi From rminnich at lanl.gov Thu Apr 24 10:10:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 24 10:10:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: Message-ID: one quick question: are you using etherboot to load the kernel image? ro From rminnich at lanl.gov Thu Apr 24 10:15:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 24 10:15:01 2003 Subject: help : flash an epia In-Reply-To: <20030424150421.A25586@mail.cwlinux.com> Message-ID: On Thu, 24 Apr 2003, Andrew Ip wrote: > In the README of via/epia, it says > $ setpci -s 0:11.0 40.b=54 we need to get this bit combination into flash_rom. ron From rminnich at lanl.gov Thu Apr 24 10:18:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 24 10:18:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: Message-ID: On 23 Apr 2003, Eric W. Biederman wrote: > If there is not a easy solution review the code generated > and see if it is initializing your serial port correctly. it is worth the exercise to hexdump the romimage and look at the end of the image, where the jump is, and where it jumps to, to make sure that there is some sanity in the generated code. ron From steve at nexpath.com Thu Apr 24 10:37:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 24 10:37:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: References: Message-ID: <3EA80024.1080401@nexpath.com> Shubhangi Jadhav wrote: > While trying to boot using linuxbios , I get a message " Copying LinuxBIOS > to ram " and then the system pauses for 3-4 secs, before printing the next > message - "Jumping to Linuxbios" > Could someone tell me what causes this delay, does the process of > uncompressing and copying linuxbios to RAM take this long? > Any suggestions on how to reduce this delay? Sounds like the decompression time. Try disabling compression (option CONFIG_COMPRESS=0) and see if it speeds up. What is biosbase set to (default is 0xf0000)? It might be caching is not set for the region of your flash, since it is executing from flash at this point. -Steve From rsmith at bitworks.com Thu Apr 24 11:01:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Thu Apr 24 11:01:01 2003 Subject: Debugging the linuxBIOS code Message-ID: <3EA803F0.1060609@bitworks.com> Eric W. Biederman wrote: > > Oh, and check your baud rate settings.. > If you use realterm (realterm.sf.net) then it will show you that if you rx line has any activity on it and if the UART threw an error. That way you can tell if you have no output at all or just no viewable output. -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Thu Apr 24 11:05:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Thu Apr 24 11:05:00 2003 Subject: Fwd: Re: Polling IDE and CF issue Message-ID: <3EA804A0.1010308@bitworks.com> eric W. Biederman wrote: > Adam Agnew writes: >>Sorry I have to ask, but.. which polled ide code are you talking about? >>LinuxBIOS native, Etherboot 5.0.6 or Etherboot 5.1.7? > > Given that the mentioned command does not occur in etherboot > 5.1.7 it sounds like the builtin IDE driver. Richard please > try the etherboot one. Ron was looking at backporting it. > Correct. its the LinuxBIOS native one. > There are no magic delay funnies and the code has been tested > with quite a few drives. Ok so that would be go get Etherboot 5.1.7 compile it so it includes IDE right? Does the make file create an Elf image or do I have to use mkelfimage? Which mainbord has a good config file example for this? -- Richard A. Smith rsmith at bitworks.com -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Thu Apr 24 11:51:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Thu Apr 24 11:51:01 2003 Subject: Etherboot compile error Message-ID: <3EA80F7C.5020002@bitworks.com> The etherboot page only had 5.1.8 available for download. When I try to compile I get a mutiple definition of 'get_memsizes' error. Who's got the 5.1.7 tarball handy? -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Thu Apr 24 12:19:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Thu Apr 24 12:19:00 2003 Subject: Nevermind In-Reply-To: <3EA80F7C.5020002@bitworks.com> References: <3EA80F7C.5020002@bitworks.com> Message-ID: <3EA81624.3090108@bitworks.com> Richard Smith wrote: > The etherboot page only had 5.1.8 available for download. > > When I try to compile I get a mutiple definition of 'get_memsizes' error. > > Who's got the 5.1.7 tarball handy? > I fixed it.. I didn't have -DPCBIOS commented out. -- Richard A. Smith rsmith at bitworks.com From dkotian3 at vsnl.net Thu Apr 24 13:44:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Thu Apr 24 13:44:00 2003 Subject: Delay in copying Linuxbios to ram References: <3EA80024.1080401@nexpath.com> Message-ID: <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> 3-4 seconds looks a long time, should uncompression of few 100 KBbytes take that long. Also, could someone please elaborate the signifigance of biosbase set with this reference. Regards Deepak ----- Original Message ----- From: "Steve Gehlbach" To: "Shubhangi Jadhav" Cc: "LinuxBIOS" Sent: Thursday, April 24, 2003 8:47 PM Subject: Re: Delay in copying Linuxbios to ram > Shubhangi Jadhav wrote: > > While trying to boot using linuxbios , I get a message " Copying LinuxBIOS > > to ram " and then the system pauses for 3-4 secs, before printing the next > > message - "Jumping to Linuxbios" > > Could someone tell me what causes this delay, does the process of > > uncompressing and copying linuxbios to RAM take this long? > > Any suggestions on how to reduce this delay? > > Sounds like the decompression time. Try disabling compression (option > CONFIG_COMPRESS=0) and see if it speeds up. What is biosbase set to > (default is 0xf0000)? It might be caching is not set for the region of > your flash, since it is executing from flash at this point. > > -Steve > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From ebiederman at lnxi.com Thu Apr 24 14:23:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 24 14:23:00 2003 Subject: Polling IDE and CF issue. In-Reply-To: <3EA802E3.8000905@bitworks.com> References: <20030423214618.I20517-100000@www.missl.cs.umd.edu> <3EA802E3.8000905@bitworks.com> Message-ID: Richard Smith writes: > Eric W. Biederman wrote: > > > Adam Agnew writes: > > >>Sorry I have to ask, but.. which polled ide code are you talking about? > >>LinuxBIOS native, Etherboot 5.0.6 or Etherboot 5.1.7? > > Given that the mentioned command does not occur in etherboot > > 5.1.7 it sounds like the builtin IDE driver. Richard please > > try the etherboot one. Ron was looking at backporting it. > > > Correct. its the LinuxBIOS native one. > > > There are no magic delay funnies and the code has been tested > > with quite a few drives. > > Ok so that would be go get Etherboot 5.1.7 compile it so it includes IDE right? > Does the make file create an Elf image or do I have to use mkelfimage? mkelfImage is not required. make ide_disk.zelf or make nic1--nic2--nic3--ide_disk.zelf. It's weird but it allows etherboot to build multiple drivers. 5.1.8 now that it is released is probably better. I need to see how well make etherboot.zelf and make etherboot-pci.zelf work. The first one includes all of the drivers and the second one includes all of the pci drivers (which should include ide_disk). > Which mainbord has a good config file example for this? Basically you just need to use the ELF bootloader. Eric From ebiederman at lnxi.com Thu Apr 24 14:24:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 24 14:24:00 2003 Subject: [Richard Smith ] Re: Debugging the linuxBIOS code In-Reply-To: References: <20030423041404.7E7CB4FF29@bom6.vsnl.net.in> Message-ID: An embedded message was scrubbed... From: Richard Smith Subject: Re: Debugging the linuxBIOS code Date: Thu, 24 Apr 2003 10:32:16 -0500 Size: 1973 URL: From rminnich at lanl.gov Thu Apr 24 14:32:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Thu Apr 24 14:32:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> Message-ID: On Thu, 24 Apr 2003, Deepak Kotian wrote: > 3-4 seconds looks a long time, should uncompression of few 100 KBbytes > take that long. sure, if you are running un-cached. I had it take 20 seconds on an L440GX+ before we got caching right. ron From steve at nexpath.com Thu Apr 24 15:19:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Thu Apr 24 15:19:01 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> Message-ID: <3EA8421F.3040805@nexpath.com> Deepak Kotian wrote: > 3-4 seconds looks a long time, should uncompression of few 100 KBbytes > take that long. > Also, could someone please elaborate the signifigance of biosbase set with > this reference. > biosbase defaults to 0xf0000 and I believe this region is automatically cached by linuxbios (maybe not?). If biosbase is set to 0xffff0000, then in your config you need to set: option XIP_ROM_SIZE= 0x01000000 option XIP_ROM_BASE= 0xff000000 This will cache the upper 16M of the 4G address space. Basically, the chipsets generally map both to the same physical memory, the flash, ie, they aliasing 0xf0000 to 0xffff0000. You can cover 0xf0000 with ram, by setting proper bits in the bridge, but not 0xffff0000. But on startup by default most if not all chipsets alias the top 64k under 1M to the top 64k under 4G. It all has to do with legacy, real-mode stuff. linuxbios is in 32-bit mode, though, after a few initial instructions. -Steve From yapeehuey at hotmail.com Thu Apr 24 20:59:00 2003 From: yapeehuey at hotmail.com (Yap Ee Huey) Date: Thu Apr 24 20:59:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image Message-ID: Yes, I am going to use etherboot to load kernel image, but I don't know how to link those files. Bendany told me : "etherboot will init the harddisk, and get the linux kernel from the first partition of your harddisk. so you just put the kernel into your harddisk's first partition and the kernel name must be "kernel". filesystem can be ext2. " But I have multiple partitions on my harddisk, Linux is not on /dev/hda1 , is there any other solutions ? Thanks. >From: ron minnich >To: Yap Ee Huey >CC: linuxbios at clustermatic.org, >Subject: Re: Linuxbios + Etherboot + tagged Kernel Image >Date: Thu, 24 Apr 2003 08:42:08 -0600 (MDT) > > >one quick question: are you using etherboot to load the kernel image? > >ro > _________________________________________________________________ Using a handphone prepaid card? Reload your credit online! http://www.msn.com.my/reloadredir/default.asp From aip at cwlinux.com Thu Apr 24 21:10:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Thu Apr 24 21:10:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: ; from Yap Ee Huey on Fri, Apr 25, 2003 at 01:32:30AM +0000 References: Message-ID: <20030425094350.D7000@mail.cwlinux.com> > "etherboot will init the harddisk, and get the linux kernel > from the first partition of your harddisk. > so you just put the kernel into your harddisk's first partition > and the kernel name must be "kernel". filesystem can be ext2. " > But I have multiple partitions on my harddisk, Linux is not on > /dev/hda1 , is there any other solutions ? I think the patch support ext2, ext3 and vfat at the moment. If you are using something other than these. You still can extract file system code from grub to support other filessystem, eg reiserfs, xfs, jfs. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From ebiederman at lnxi.com Thu Apr 24 22:36:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Thu Apr 24 22:36:01 2003 Subject: [ANNOUNCE] mkelfImage-2.5 Message-ID: At: ftp://ftp.lnxi.com/pub/src/mkelfImage/mkelfImage-2.5.tar.gz * 2.5 24 April 2003 - Fix the checksum code so that it checksums the bss sections as well. How I missed the checksum code was broken in 2.4 I don't know. Eric From agnew at cs.umd.edu Fri Apr 25 00:09:01 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Fri Apr 25 00:09:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <20030425094350.D7000@mail.cwlinux.com> Message-ID: <20030425004629.V25844-100000@www.missl.cs.umd.edu> On Fri, 25 Apr 2003, Andrew Ip wrote: > > "etherboot will init the harddisk, and get the linux kernel > > from the first partition of your harddisk. > > so you just put the kernel into your harddisk's first partition > > and the kernel name must be "kernel". filesystem can be ext2. " > > But I have multiple partitions on my harddisk, Linux is not on > > /dev/hda1 , is there any other solutions ? > I think the patch support ext2, ext3 and vfat at the moment. > If you are using something other than these. You still can > extract file system code from grub to support other filessystem, > eg reiserfs, xfs, jfs. Wrong etherboot. For file system support you want my patch against etherboot-5.0.6 found at http://www.missl.cs.umd.edu/~agnew/ (the rules change though, now you want to "make bin32/sis900.ebi") From ts1 at cma.co.jp Fri Apr 25 03:00:00 2003 From: ts1 at cma.co.jp (SONE Takeshi) Date: Fri Apr 25 03:00:00 2003 Subject: help : flash an epia In-Reply-To: References: <20030424150421.A25586@mail.cwlinux.com> Message-ID: <20030425073307.GA944@cma.co.jp> On Thu, Apr 24, 2003 at 08:47:19AM -0600, ron minnich wrote: > On Thu, 24 Apr 2003, Andrew Ip wrote: > > > In the README of via/epia, it says > > $ setpci -s 0:11.0 40.b=54 > > we need to get this bit combination into flash_rom. It's already in there, you applied my patch. -- Takeshi From yenneo at oreka.com Fri Apr 25 03:00:40 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Fri Apr 25 03:00:40 2003 Subject: help: flash an epia Message-ID: <200304250734.JAA43339@mailhub10.isdnet.net> I'm sorry but i still have problems. the setpci command looks ok. thanks ;-) When i execute the makefile i don't have error anymore but i can't run flash_rom. I've read this : >Basically you need flash_rom to flash romimage. It can be built from >cvs >tree(freebios/util/flash_and_burn). Here is the step to flash, >1. enable flash writable >$ setpci -s 0:11.0 40.b=3D54 >2. flash romimage >$ flash_rom romimage but when i execute "flash_rom romimage" it says me "command not found" -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From shubhangi.jadhav at patni.com Fri Apr 25 03:07:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Fri Apr 25 03:07:01 2003 Subject: linuxbios problem In-Reply-To: Message-ID: Hi, When executing Linuxbios the machine resets after printing the message "Welcome to start32 ..." And then Linuxbios goes in a loop, restarting the boot process. what's wrong here and how to fix it? I am using Linuxbios with ADLO. ****************************************************** Welcome to start32, the open sourced starter. This space will eventually hold more diagnostic information. January 2000, James Hendricks, Dale Webster, and Ron Minnich. Version 0.1 LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 ****************************************************** Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Shubhangi Jadhav Sent: Thursday, April 24, 2003 6:16 PM To: LinuxBIOS Subject: linuxbios problem Hi, I'm trying to get my motherboard working with Linuxbios. I'm using ADLO with Linuxbios. The debug messages are as below. The problem is after jumping to linuxbiosmain (it prints "Welcome to start32......") it loops back , restarting the boot process all over again. what's wrong ? why does it go in a loop? ***************log********************************* LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 Ram2 Ram3 Ram Enable 1 Ram Enable 2 Ram Enable 3 Ram Enable 4 Ram Enable 5 Ram4 Ram5 Ram6 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 booting... Finding PCI configuration type. PCI: Using configuration type 1 handle_superio start, nsuperio 1 handle_superio Pass 0, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 0, Superio w83627hf handle_superio port 0x0, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 0, done #0 handle_superio done Scanning PCI bus...PCI: pci_scan_bus for bus 0 PCI: 00:00.0 [8086/7124] PCI: 00:1e.0 [8086/2418] PCI: 00:1f.0 [8086/2410] PCI: 00:1f.1 [8086/2411] PCI: 00:1f.2 [8086/2412] PCI: 00:1f.3 [8086/2413] PCI: 00:1f.5 [8086/2415] PCI: 00:1f.6 [8086/2416] PCI: pci_scan_bus for bus 1 PCI: 01:04.0 [10ec/8139] PCI: 01:05.0 [10ec/8139] PCI: pci_scan_bus returning with max=01 PCI: pci_scan_bus returning with max=01 done Allocating PCI resources... ASSIGN RESOURCES, bus 0 PCI: 00:1e.0 1c <- [0x00001000 - 0x00001fff] bus 1 io PCI: 00:1e.0 24 <- [0xfec00000 - 0xfebfffff] bus 1 prefmem PCI: 00:1e.0 20 <- [0xfeb00000 - 0xfebfffff] bus 1 mem ASSIGN RESOURCES, bus 1 PCI: 01:04.0 10 <- [0x00001000 - 0x000010ff] io PCI: 01:04.0 14 <- [0xfeb00000 - 0xfeb000ff] mem PCI: 01:05.0 10 <- [0x00001400 - 0x000014ff] io PCI: 01:05.0 14 <- [0xfeb01000 - 0xfeb010ff] mem ASSIGNED RESOURCES, bus 1 PCI: 00:1f.1 20 <- [0x000028e0 - 0x000028ef] io PCI: 00:1f.2 20 <- [0x000028c0 - 0x000028df] io PCI: 00:1f.3 20 <- [0x000028f0 - 0x000028ff] io PCI: 00:1f.5 10 <- [0x00002000 - 0x000020ff] io PCI: 00:1f.5 14 <- [0x00002880 - 0x000028bf] io PCI: 00:1f.6 10 <- [0x00002400 - 0x000024ff] io PCI: 00:1f.6 14 <- [0x00002800 - 0x0000287f] io ASSIGNED RESOURCES, bus 0 done. Enabling PCI resourcess...PCI: 00:00.0 cmd <- 06 PCI: 00:1e.0 cmd <- 07 PCI: 00:1f.0 cmd <- 0f PCI: 00:1f.1 cmd <- 01 PCI: 00:1f.2 cmd <- 01 PCI: 00:1f.3 cmd <- 01 PCI: 00:1f.5 cmd <- 01 PCI: 00:1f.6 cmd <- 01 PCI: 01:04.0 cmd <- 03 PCI: 01:05.0 cmd <- 03 done. Initializing PCI devices... PCI devices initialized DRP0 = 0x9 DIMM0 - size = 64M DIMM1 - size = 0M DRP1 = 0x0 DIMM2 - size = 0M totalram: 64M Initializing CPU #0 Updating microcode microcode_info: sig = 0x00000678 pf=0x00000001 rev = 0x00000000 Enabling cache... Setting fixed MTRRs(0-88) type: UC Setting fixed MTRRs(0-16) type: WB DONE fixed MTRRs Setting variable MTRR 0, base: 0MB, range: 64MB, type WB DONE variable MTRRs Clear out the extra MTRR's call intel_enable_fixed_mtrr() call intel_enable_var_mtrr() Leave setup_mtrrs done. Max cpuid index : 1 Vendor ID : CentaurHauls Processor Type : 0x00 Processor Family : 0x06 Processor Model : 0x07 Processor Mask : 0x00 Processor Stepping : 0x08 Feature flags : 0x00803035 MTRR check Fixed MTRRs : Enabled Variable MTRRs: Enabled Configuring L2 cache...Not 'GenuineIntel' Processor Enable Cache done. Disabling local apic...done. CPU #0 Initialized ioapic southbridge enabled 180 RTC Init Invalid CMOS LB checksum set power on after power fail Please turn on nvram handle_superio start, nsuperio 1 handle_superio Pass 1, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 1, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e Call init Enabling com device: 03 iobase = 0x02f8 irq=3 handle_superio Pass 1, done #0 handle_superio done handle_superio start, nsuperio 1 handle_superio Pass 2, check #0, s 0000cde0 s->super 0000e138 handle_superio: Pass 2, Superio w83627hf handle_superio port 0x2e, defaultport 0x2e handle_superio Using port 0x2e handle_superio Pass 2, done #0 handle_superio done Wrote linuxbios table at: 00000500 - 00000664 checksum 7535 Jumping to linuxbiosmain()... Welcome to start32, the open sourced starter. This space will eventually hold more diagnostic information. January 2000, James Hendricks, Dale Webster, and Ron Minnich. Version 0.1 LinuxBIOS-1.0.0 Thu Apr 24 06:42:06 IST 2003 starting... Ram1 Ram2 Ram3 Ram Enable 1 Ram Enable 2 Ram Enable 3 Ram Enable 4 Ram Enable 5 Ram4 Ram5 Ram6 Copying LinuxBIOS to ram. Jumping to LinuxBIOS. ***************log********************************* Regards, Shubhangi _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From shubhangi.jadhav at patni.com Fri Apr 25 03:35:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Fri Apr 25 03:35:00 2003 Subject: Using ADLO with Linuxbios Message-ID: Hi, I have 512k bios, so the kernel should be put somewhere on the harddisk. Hence I'm using ADLO with Linuxbios. However I'm not able to get linuxbios working. I followed these steps : Build the ADLO payload as per the instructions in the readme for ADLO. Appended the following entry to the config file : USE_ELF_BOOT=0 payload /home/shubhangi/freebios/util/ADLO/payload Build the romimage using this config file. Is this the correct way to use ADLO? Thanks for any help. Regards, Shubhangi From ts1 at cma.co.jp Fri Apr 25 04:38:00 2003 From: ts1 at cma.co.jp (SONE Takeshi) Date: Fri Apr 25 04:38:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <3EA8421F.3040805@nexpath.com> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> Message-ID: <20030425091117.GA5673@cma.co.jp> On Thu, Apr 24, 2003 at 12:59:27PM -0700, Steve Gehlbach wrote: > Deepak Kotian wrote: > >3-4 seconds looks a long time, should uncompression of few 100 KBbytes > >take that long. > >Also, could someone please elaborate the signifigance of biosbase set with > >this reference. > > > > biosbase defaults to 0xf0000 and I believe this region is automatically > cached by linuxbios (maybe not?). If biosbase is set to 0xffff0000, > then in your config you need to set: > > option XIP_ROM_SIZE= 0x01000000 > option XIP_ROM_BASE= 0xff000000 I have the same delay problem on VIA EPIA board. It didn't run with biosbase set to 0xffff0000 (hangs after first banner, and I don't have a POST card to track it down). In cpu/p6/earlymtrr.inc, 0xf0000-0xfffff is left uncached. (Only 0-640KB is cached, 640KB-1M is uncached) I will try adding code there to set fixed MTRR for 0xf0000-0xfffff WP caching. It should be harmless for other boards that do not need this, what do you think? -- Takeshi From bendany at mistdl.com Fri Apr 25 07:58:01 2003 From: bendany at mistdl.com (bendany) Date: Fri Apr 25 07:58:01 2003 Subject: Using ADLO with Linuxbios References: Message-ID: <002901c30ba4$4faac6c0$8e00a8c0@benbook> ----- Original Message ----- From: "Shubhangi Jadhav" To: "LinuxBIOS" Sent: Friday, April 25, 2003 1:10 AM Subject: Using ADLO with Linuxbios > Hi, > > I have 512k bios, so the kernel should be put somewhere on the harddisk. > Hence I'm using ADLO with Linuxbios. However I'm not able to get linuxbios > working. > I followed these steps : > > Build the ADLO payload as per the instructions in the readme for ADLO. > Appended the following entry to the config file : > USE_ELF_BOOT=0 why set it to 0? ADLO+BOCHS+VIDEOBIOS will be a ELF program. so you can put the ADLO to the anywhere, such as rom/harddisk/tftp server/ first, you can test the LinuxBIOS , make sure that it can run on your mb correctly. Also make sure that the linuxbios can load the ELF kernel correctly. and next you can replace the linux ELF kernel with the ADLO, patch the rombios.c with the serial patch. and try the next test. BTW: what is your mb's type? sis630,or else? > payload /home/shubhangi/freebios/util/ADLO/payload > Build the romimage using this config file. > > Is this the correct way to use ADLO? Thanks for any help. > > Regards, > Shubhangi > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From shubhangi.jadhav at patni.com Fri Apr 25 08:19:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Fri Apr 25 08:19:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <002901c30ba4$4faac6c0$8e00a8c0@benbook> Message-ID: Hi, If I burn ADLO alongwith linuxbios on rom do we need to specify BOOT_IDE=1 ? Thanks, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of bendany Sent: Saturday, April 26, 2003 9:01 AM To: Shubhangi Jadhav Cc: Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios ----- Original Message ----- From: "Shubhangi Jadhav" To: "LinuxBIOS" Sent: Friday, April 25, 2003 1:10 AM Subject: Using ADLO with Linuxbios > Hi, > > I have 512k bios, so the kernel should be put somewhere on the harddisk. > Hence I'm using ADLO with Linuxbios. However I'm not able to get linuxbios > working. > I followed these steps : > > Build the ADLO payload as per the instructions in the readme for ADLO. > Appended the following entry to the config file : > USE_ELF_BOOT=0 why set it to 0? ADLO+BOCHS+VIDEOBIOS will be a ELF program. so you can put the ADLO to the anywhere, such as rom/harddisk/tftp server/ first, you can test the LinuxBIOS , make sure that it can run on your mb correctly. Also make sure that the linuxbios can load the ELF kernel correctly. and next you can replace the linux ELF kernel with the ADLO, patch the rombios.c with the serial patch. and try the next test. BTW: what is your mb's type? sis630,or else? > payload /home/shubhangi/freebios/util/ADLO/payload > Build the romimage using this config file. > > Is this the correct way to use ADLO? Thanks for any help. > > Regards, > Shubhangi > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Fri Apr 25 09:25:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 25 09:25:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: Message-ID: On Fri, 25 Apr 2003, Yap Ee Huey wrote: > Yes, I am going to use etherboot to load kernel image, > but I don't know how to link those files. Bendany told me : Here is what I did: I used fdisk to make an empty partition at the front of the drive, at hda1. When I had a kernel ready: cat vmlinux.gz > /dev/hda1 That puts it where etherboot can get it. ron From agnew at cs.umd.edu Fri Apr 25 11:29:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Fri Apr 25 11:29:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: <20030425120632.B27749-100000@www.missl.cs.umd.edu> On Fri, 25 Apr 2003, Shubhangi Jadhav wrote: > Hi, > > If I burn ADLO alongwith linuxbios on rom do we need to specify BOOT_IDE=1 Nope. BOOT_IDE is merely to use the ide support thats directly in LinuxBIOS, in which case it would skip right over ADLO. From rsmith at bitworks.com Fri Apr 25 11:33:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Fri Apr 25 11:33:00 2003 Subject: Polling IDE and CF issue. In-Reply-To: References: <20030423214618.I20517-100000@www.missl.cs.umd.edu> <3EA802E3.8000905@bitworks.com> Message-ID: <3EA95CD3.4030306@bitworks.com> Eric W. Biederman wrote: > make ide_disk.zelf > or Assumint you are in the etherboot-x.x.x/src directory That should be: make bin/ide_disk.zelf > make nic1--nic2--nic3--ide_disk.zelf. Couldn't make this work. tried "bin/3c509.zelf--ide_disk.zelf" and "bin/3c509.zelf--bin/ide_disk.zelf" both yeilded "No rule to make target" > 5.1.8 now that it is released is probably better. Thats what I used. > I need to see how well make etherboot.zelf and make etherboot-pci.zelf Both bin/etherboot.zelf and bin/etherboot-pci.zelf worked fine. -- Richard A. Smith rsmith at bitworks.com From steve at nexpath.com Fri Apr 25 13:17:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 25 13:17:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <20030425091117.GA5673@cma.co.jp> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> Message-ID: <3EA9770D.4040507@nexpath.com> > I have the same delay problem on VIA EPIA board. > It didn't run with biosbase set to 0xffff0000 (hangs after first banner, > and I don't have a POST card to track it down). Not surprising, some implementations depend on executing at 0xf0000. If the serial port is working, you can put in prints etc and track it down without a POST card. > > In cpu/p6/earlymtrr.inc, 0xf0000-0xfffff is left uncached. > (Only 0-640KB is cached, 640KB-1M is uncached) > I will try adding code there to set fixed MTRR for 0xf0000-0xfffff > WP caching. > It should be harmless for other boards that do not need this, > what do you think? > > Not sure. Need to see if the RAM is turned on for this region later after the C-code starts, if so, then the cache memory type may have to be changed from WP to normal. -Steve From dkotian3 at vsnl.net Fri Apr 25 14:37:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Fri Apr 25 14:37:00 2003 Subject: Using ADLO with Linuxbios References: <20030425120632.B27749-100000@www.missl.cs.umd.edu> Message-ID: <038a01c30b5e$32602920$f02041db@vsnl.net> So, if I want to boot from Compact Flash IDE which has Linux, would LinuxBIOS directly boot it. Do I really need ADLO for this purpose ? And also can one do away with etherboot as well ? If ADLO is not needed, then would the kernel image be read by the LINUXBIOS directly from the IDE if we specific the path of the vmlinuz image. I read a message from Ron, that one can just do cat vmlinux..gz > /dev/hda1 and etherboot would be OK. Should it done something like this when BOOT_IDE is set. Could you please elaborate. Regards Deepak ----- Original Message ----- From: "Adam Agnew" To: "Shubhangi Jadhav" Cc: "bendany" ; Sent: Friday, April 25, 2003 9:38 PM Subject: RE: Using ADLO with Linuxbios > > On Fri, 25 Apr 2003, Shubhangi Jadhav wrote: > > > Hi, > > > > If I burn ADLO alongwith linuxbios on rom do we need to specify BOOT_IDE=1 > > Nope. BOOT_IDE is merely to use the ide support thats directly in > LinuxBIOS, in which case it would skip right over ADLO. > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Fri Apr 25 15:10:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 25 15:10:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <038a01c30b5e$32602920$f02041db@vsnl.net> Message-ID: On Sat, 26 Apr 2003, Deepak Kotian wrote: > So, if I want to boot from Compact Flash IDE which has Linux, would > LinuxBIOS directly boot it. > Do I really need ADLO for this purpose ? And also can one do away with > etherboot as > well ? > If ADLO is not needed, then would the kernel image be read by the LINUXBIOS > directly > from the IDE if we specific the path of the vmlinuz image. > > I read a message from Ron, that one can just > do > cat vmlinux..gz > /dev/hda1 and etherboot would be OK. cat elfImage > /dev/hda1 make sure it is a linuxbios-format elfImage ron From dkotian3 at vsnl.net Fri Apr 25 15:15:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Fri Apr 25 15:15:01 2003 Subject: Using ADLO with Linuxbios References: Message-ID: <001001c30b63$94f11040$353e41db@vsnl.net> Sorry, did not quite get it. What is this elfimage, I have my own kernel image which I would like to boot from LinuxBIOS. Can I convert my kernel image to LinuxBIOS format, what would be procedure.. Could you Please elaborate.. ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: "Adam Agnew" ; "Shubhangi Jadhav" ; "bendany" ; Sent: Saturday, April 26, 2003 1:12 AM Subject: Re: Using ADLO with Linuxbios > On Sat, 26 Apr 2003, Deepak Kotian wrote: > > > So, if I want to boot from Compact Flash IDE which has Linux, would > > LinuxBIOS directly boot it. > > Do I really need ADLO for this purpose ? And also can one do away with > > etherboot as > > well ? > > If ADLO is not needed, then would the kernel image be read by the LINUXBIOS > > directly > > from the IDE if we specific the path of the vmlinuz image. > > > > I read a message from Ron, that one can just > > do > > cat vmlinux..gz > /dev/hda1 and etherboot would be OK. > > cat elfImage > /dev/hda1 > > make sure it is a linuxbios-format elfImage > > ron > > From rsmith at bitworks.com Fri Apr 25 15:46:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Fri Apr 25 15:46:01 2003 Subject: etherboot dosen't find my elfkernel Message-ID: <3EA997F5.2070402@bitworks.com> I've got etherboot loading now. I took a bzImage and turned it into a elfimage with mkelfimage bzImage elfimage using mkelfimage 2.5 then stuck that on my CF in the first partition ie 'cat elfimage > /dev/hde1' where hde is the pcmcia mount of my CF Stick it in my target and boot. I get: < linux bios stuff snipped> Etherboot 5.1.8 (GPL) Tagged ELF for [IDE] Relocating _text from: [00023d20,00032e30) to [07ff0ef0,08000000) Probing pci disk... [IDE]disk-1 7872k cap: 0200 Searching for image... ................................ Probing pci disk... [IDE]Probing isa disk... Probing pci disk... [IDE]disk0 7872k cap: 0200 Searching for image... ................................ Probing pci disk... [IDE]Probing isa disk... Probing pci disk... [IDE]disk1 7872k cap: 0200 Searching for image... ................................ Probing pci disk... [IDE]Probing isa disk... And it just goes on and on checking different disks Did I miss something? Is Tagged ELF different from the elf image that I created with mkelfimage? -- Richard A. Smith rsmith at bitworks.com From agnew at cs.umd.edu Fri Apr 25 16:06:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Fri Apr 25 16:06:00 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: <3EA997F5.2070402@bitworks.com> Message-ID: <20030425164501.B28568-100000@www.missl.cs.umd.edu> PCMCIA mount? Etherboot wont look at pcmcia devices. If its going to find a compact flash card, it will be because there's a compact flash on IDE adapter. Then it will see it as a hard disk. On Fri, 25 Apr 2003, Richard Smith wrote: > I've got etherboot loading now. > > I took a bzImage and turned it into a elfimage with > > mkelfimage bzImage elfimage > > using mkelfimage 2.5 > > then stuck that on my CF in the first partition ie 'cat elfimage > > /dev/hde1' where hde is the pcmcia mount of my CF > > Stick it in my target and boot. > > I get: > > < linux bios stuff snipped> > > Etherboot 5.1.8 (GPL) Tagged ELF for [IDE] > Relocating _text from: [00023d20,00032e30) to [07ff0ef0,08000000) > Probing pci disk... > [IDE]disk-1 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > Probing pci disk... > [IDE]disk0 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > Probing pci disk... > [IDE]disk1 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > > And it just goes on and on checking different disks > > Did I miss something? Is Tagged ELF different from the elf image that I > created with mkelfimage? > > -- > Richard A. Smith > rsmith at bitworks.com > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From steve at nexpath.com Fri Apr 25 16:14:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 25 16:14:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <001001c30b63$94f11040$353e41db@vsnl.net> References: <001001c30b63$94f11040$353e41db@vsnl.net> Message-ID: <3EA99FD5.4000104@nexpath.com> Deepak Kotian wrote: > Sorry, did not quite get it. > What is this elfimage, I have my own kernel image which I would like > to boot from LinuxBIOS. > Can I convert my kernel image to LinuxBIOS format, what > would be procedure.. > > Could you Please elaborate.. In the current linuxbios, you can also boot directly without an elf image file. Look at the configuration freebios/util/config/pcchips787.config; this boots from CF on IDE device 2 = /dev/hdc1 (cat linux.bin.gz > /dev/hdc1) and mounts /dev/hda2 as root. linux.bin.gz is made in the build process by the linuxbios makefile, from vmlinux in your kernel source directory. It is my understanding this direct mechanism will be removed, though, in linuxbios2, so that only elf images can be used in the future. -Steve From rminnich at lanl.gov Fri Apr 25 16:43:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Fri Apr 25 16:43:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EA99FD5.4000104@nexpath.com> Message-ID: On Fri, 25 Apr 2003, Steve Gehlbach wrote: > It is my understanding this direct mechanism will be removed, though, in > linuxbios2, so that only elf images can be used in the future. yes but booting elfimages direct from linuxbios will still be possible. ron From ebiederman at lnxi.com Fri Apr 25 20:03:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:03:00 2003 Subject: Polling IDE and CF issue. In-Reply-To: <3EA95CD3.4030306@bitworks.com> References: <20030423214618.I20517-100000@www.missl.cs.umd.edu> <3EA802E3.8000905@bitworks.com> <3EA95CD3.4030306@bitworks.com> Message-ID: Richard Smith writes: > Eric W. Biederman wrote: > > > make ide_disk.zelf or > > Assumint you are in the etherboot-x.x.x/src directory That should be: > > make bin/ide_disk.zelf > > > make nic1--nic2--nic3--ide_disk.zelf. > > Couldn't make this work. tried "bin/3c509.zelf--ide_disk.zelf" and > "bin/3c509.zelf--bin/ide_disk.zelf" > > both yeilded "No rule to make target" The supported target is: bin/3x509--ide_disk.zelf. > > 5.1.8 now that it is released is probably better. > > Thats what I used. > > > I need to see how well make etherboot.zelf and make etherboot-pci.zelf > > Both bin/etherboot.zelf and bin/etherboot-pci.zelf worked fine. Good. Eric From ebiederman at lnxi.com Fri Apr 25 20:12:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:12:00 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: <3EA997F5.2070402@bitworks.com> References: <3EA997F5.2070402@bitworks.com> Message-ID: Richard Smith writes: > I've got etherboot loading now. > > I took a bzImage and turned it into a elfimage with > > mkelfimage bzImage elfimage > > using mkelfimage 2.5 Yeah. At least someone does not have broken binutils :) > then stuck that on my CF in the first partition ie 'cat elfimage > /dev/hde1' > where hde is the pcmcia mount of my CF > > Stick it in my target and boot. > > I get: > > < linux bios stuff snipped> > > Etherboot 5.1.8 (GPL) Tagged ELF for [IDE] > Relocating _text from: [00023d20,00032e30) to [07ff0ef0,08000000) > Probing pci disk... > [IDE]disk-1 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > Probing pci disk... > [IDE]disk0 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > Probing pci disk... > [IDE]disk1 7872k cap: 0200 > Searching for image... > ................................ > Probing pci disk... > [IDE]Probing isa disk... > > > And it just goes on and on checking different disks > > Did I miss something? Is Tagged ELF different from the elf image that I created > with mkelfimage? Nope. the file format is right. Etherboot just works the way the LinuxBIOS code originally did. In particular it scans the first 8K of disk for an ELF header. This means you don't have to be at the absolute start of the disk for your ELF image you do need to be close. Think of what is happening as a super master boot record instead of a stupid bootloader. As there is a very strong reluctance to upgrade firmware I try to make the as simple as possible while still being powerful enough to be useful. Hopefully at some point there a bootloader I can read off of the disk. With the appropriate tools it is possible to force your first partition early in the disk. By default the first partition is often aligned on a cylinder boundary, which is probably what you are running into. The fact the IDE driver is reading your disk and not having problems is good. Eric From ebiederman at lnxi.com Fri Apr 25 20:16:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:16:01 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: <20030425164501.B28568-100000@www.missl.cs.umd.edu> References: <20030425164501.B28568-100000@www.missl.cs.umd.edu> Message-ID: Adam Agnew writes: > PCMCIA mount? > Etherboot wont look at pcmcia devices. If its going to find a compact > flash card, it will be because there's a compact flash on IDE adapter. > Then it will see it as a hard disk. You may have a point. Richard how many IDE disks do you have? I guess I am wondering if it looks like the correct disk is being checked. I still favor the partition is to late on the disk theory but, not checking the CF works as well. For a pure test case it is possible to just dd the ELF image onto the start of the disk. If you don't need partitions. That is the way Ron is currently using the code. Eric From ebiederman at lnxi.com Fri Apr 25 20:18:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:18:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <3EA9770D.4040507@nexpath.com> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> Message-ID: Steve Gehlbach writes: > > I have the same delay problem on VIA EPIA board. > > It didn't run with biosbase set to 0xffff0000 (hangs after first banner, > > and I don't have a POST card to track it down). > > Not surprising, some implementations depend on executing at 0xf0000. If the > serial port is working, you can put in prints etc and track it down without a > POST card. > > > In cpu/p6/earlymtrr.inc, 0xf0000-0xfffff is left uncached. > > (Only 0-640KB is cached, 640KB-1M is uncached) > > I will try adding code there to set fixed MTRR for 0xf0000-0xfffff > > WP caching. > > It should be harmless for other boards that do not need this, > > what do you think? > > > > Not sure. Need to see if the RAM is turned on for this region later after the > C-code starts, if so, then the cache memory type may have to be changed from WP > to normal. The generic code should handle this. I believe all of the mtrrs are rebuilt. Eric From ebiederman at lnxi.com Fri Apr 25 20:19:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:19:00 2003 Subject: setting USE_ELF_BOOT In-Reply-To: References: Message-ID: "Shubhangi Jadhav" writes: > Hi, > > When do we need to set USE_ELF_BOOT option ? Whenever you are not using the deprecated builtin linux loader. > Is elfboot loader required to load ADLO ? or is it required only in case > of etherboot? The ELF loader is required. Eric From ebiederman at lnxi.com Fri Apr 25 20:27:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:27:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: References: Message-ID: ron minnich writes: > On Fri, 25 Apr 2003, Steve Gehlbach wrote: > > > It is my understanding this direct mechanism will be removed, though, in > > linuxbios2, so that only elf images can be used in the future. Already done. The basic rule so we have a line against code bloat is that: LinuxBIOS will boot an ELF image non-interactively off of 1 device. It can be any device you want, but it is one ELF image. > yes but booting elfimages direct from linuxbios will still be possible. Yep. So far I only have the simple rom driver in the tree. But moving some other ones over as well should not be a problem. In building up the tree I am going back and forth between making the Opteron port work (with everything hard coded I can boot linux now), and simplifying and generalizing the generic code. An interesting facet of how Hypertransport works even device numbers are no longer fixed. So our current interrupt code must be reworked. When the bus and device number change copying static tables of static information just does not work. Eric From ebiederman at lnxi.com Fri Apr 25 20:29:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Fri Apr 25 20:29:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: References: Message-ID: ron minnich writes: > On Fri, 25 Apr 2003, Yap Ee Huey wrote: > > > Yes, I am going to use etherboot to load kernel image, > > but I don't know how to link those files. Bendany told me : > > Here is what I did: > > I used fdisk to make an empty partition at the front of the drive, at > hda1. When I had a kernel ready: > > cat vmlinux.gz > /dev/hda1 > > That puts it where etherboot can get it. Note. Etherboot is not limited to only one hard disk so it may make sense to put it on a secondary disk. Eric From steve at nexpath.com Fri Apr 25 22:57:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Fri Apr 25 22:57:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: References: Message-ID: <3EAA0166.9090107@nexpath.com> >>cat vmlinux.gz > /dev/hda1 >> >>That puts it where etherboot can get it. > > > Note. Etherboot is not limited to only one hard disk so it may make > sense to put it on a secondary disk. > > Eric I don't see how this works, with an 8kbytes limit how etherboot could find it in hda1. Typical disks are 63 sectors, so the initial skip to partition 1 is 63 x 512 ~= 32K, or much larger than 8K. -Steve From shubhangi.jadhav at patni.com Sat Apr 26 09:20:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Sat Apr 26 09:20:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: So now from what understand, if I want to use Linuxbios to directly boot my elfimage how do I go about build linuxbios using the option BOOT_IDE in the config file build an elfimage from my kernel image setting OUTPUT_FORMAT="elf32-i386" in mkelfimage cat elfimage > /dev/hda1 burn linuxbios on the flash rom Is that all I need to get going ? Thanks, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Eric W. Biederman Sent: Saturday, April 26, 2003 6:35 AM To: ron minnich Cc: Steve Gehlbach; Deepak Kotian; Adam Agnew; Shubhangi Jadhav; bendany; Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios ron minnich writes: > On Fri, 25 Apr 2003, Steve Gehlbach wrote: > > > It is my understanding this direct mechanism will be removed, though, in > > linuxbios2, so that only elf images can be used in the future. Already done. The basic rule so we have a line against code bloat is that: LinuxBIOS will boot an ELF image non-interactively off of 1 device. It can be any device you want, but it is one ELF image. > yes but booting elfimages direct from linuxbios will still be possible. Yep. So far I only have the simple rom driver in the tree. But moving some other ones over as well should not be a problem. In building up the tree I am going back and forth between making the Opteron port work (with everything hard coded I can boot linux now), and simplifying and generalizing the generic code. An interesting facet of how Hypertransport works even device numbers are no longer fixed. So our current interrupt code must be reworked. When the bus and device number change copying static tables of static information just does not work. Eric _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From shubhangi.jadhav at patni.com Sat Apr 26 09:41:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Sat Apr 26 09:41:01 2003 Subject: setting USE_ELF_BOOT In-Reply-To: Message-ID: I build the ADLO payload (note I'm not using etherboot) and linuxbios (using USE_ELF_BOOT option). created the romimage from these and burnt the image on flash rom. But when I try to boot from it I get the following error ********************************************************* Welcome to elfboot, the open sourced starter. January 2002, Eric Biederman. Version 1.2 203:init_bytes() - zkernel_start:0xfffc0000 zkernel_mask:0x0000007f Searching for 16 byte tags 64:rom_read_bytes() - overflowed source buffer. max_block = 3 init_bytes found 0 tags Cannot Load ELF Image ********************************************************* What could be wrong here? Thanks, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Eric W. Biederman Sent: Saturday, April 26, 2003 6:27 AM To: Shubhangi Jadhav Cc: LinuxBIOS Subject: Re: setting USE_ELF_BOOT "Shubhangi Jadhav" writes: > Hi, > > When do we need to set USE_ELF_BOOT option ? Whenever you are not using the deprecated builtin linux loader. > Is elfboot loader required to load ADLO ? or is it required only in case > of etherboot? The ELF loader is required. Eric _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From ebiederman at lnxi.com Sat Apr 26 15:04:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Sat Apr 26 15:04:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <3EAA0166.9090107@nexpath.com> References: <3EAA0166.9090107@nexpath.com> Message-ID: Steve Gehlbach writes: > >>cat vmlinux.gz > /dev/hda1 > >> > >>That puts it where etherboot can get it. > > Note. Etherboot is not limited to only one hard disk so it may make > > sense to put it on a secondary disk. > > Eric > > I don't see how this works, with an 8kbytes limit how etherboot could find it in > > hda1. Typical disks are 63 sectors, so the initial skip to partition 1 is 63 x > 512 ~= 32K, or much larger than 8K. So again my goal was not to be able to boot out of hda1 but to provide a MBR like boot mechanism that would work with older systems. It happens that you can tell some of the partition programs to create a partition that is not aligned on cylinder boundaries. So they can create a partition hda1 that you can just throw the ELF image into. The way etherboot loads the image is it interprets all addresses in the ELF image as disk addresses with offsets from the start of the ELF image. There is support for ELF64 so you can if you wish scatter the image across an entire disk. The only restriction is there is an assumption that the entire ELF header, and ELF program header are read into memory with the first 8K read. I have a proof of concept utility that will look up the addresses a ELF images is stored on disk (like lilo) and generate a new 64bit ELF header that will allow it to be loaded from anywhere on the disk. I have not published it because the address it puts the ELF header at is currently hard coded, and there are several other stupid assumptions. If you want the code holler and I will send it to you. Eric From ebiederman at lnxi.com Sat Apr 26 15:07:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Sat Apr 26 15:07:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: References: Message-ID: "Shubhangi Jadhav" writes: > So now from what understand, if I want to use Linuxbios to directly boot my > elfimage how do I go about > build linuxbios using the option BOOT_IDE in the config file > build an elfimage from my kernel image setting > OUTPUT_FORMAT="elf32-i386" > > in mkelfimage > cat elfimage > /dev/hda1 > burn linuxbios on the flash rom > Is that all I need to get going ? Yes I think so. The BOOT_IDE driver has been modified in strange ways to find things in the first partition and I have not had a chance to look over the code to see what is going on there yet. Eric From root at hamburg.de Sat Apr 26 17:35:01 2003 From: root at hamburg.de (Felix Kloeckner) Date: Sat Apr 26 17:35:01 2003 Subject: post data on sis950 lpt Message-ID: <20030427000807.A16839@synapse.pentanet> hi, some time ago i made linuxbios successfully running on the elitegroup p6stp-fl. because i am using the onboard parallel port (sis950) i found that the port will be correctly switched on, but it's left in "post data port" mode. i modified "src/superio/sis/950/superio.c" so post data port mode will be switched off. may someone patch that file or does anybody need the parallel port in that mode? (patch is attached) felix -------------- next part -------------- A non-text attachment was scrubbed... Name: sis950lpt.patch Type: application/octet-stream Size: 709 bytes Desc: not available URL: From rminnich at lanl.gov Sat Apr 26 22:03:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sat Apr 26 22:03:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <3EAA0166.9090107@nexpath.com> Message-ID: On Fri, 25 Apr 2003, Steve Gehlbach wrote: > >>cat vmlinux.gz > /dev/hda1 > I don't see how this works, with an 8kbytes limit how etherboot could > find it in hda1. Typical disks are 63 sectors, so the initial skip to > partition 1 is 63 x 512 ~= 32K, or much larger than 8K. my mistake. What I have done is this for now: dd if=elfImage of=/dev/hda bs=4096 skip=1 preserves partition information. A kludge, but it works. ron From steve at nexpath.com Sat Apr 26 22:05:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Sat Apr 26 22:05:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: References: Message-ID: <3EAB46AE.6090005@nexpath.com> Eric W. Biederman wrote: >The BOOT_IDE driver has been modified in strange > ways to find things in the first partition and I have not had a chance > to look over the code to see what is going on there yet. > Not strange to me, by default it simply skips 63x512 bytes (0x7e00), which is normally for most disks today where partition 1 starts. Of course, it should read and decode the partition table, but this is much harder. -Steve From rminnich at lanl.gov Sat Apr 26 22:23:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sat Apr 26 22:23:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: On Sat, 26 Apr 2003, Shubhangi Jadhav wrote: > cat elfimage > /dev/hda1 I made a mistake. Sorry. This is what I had to do. dd if=elfImage of=/dev/hda bs=4096 seek=1 This leaves partition tables, mbr, etc. intact. ron From rminnich at lanl.gov Sat Apr 26 22:29:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Sat Apr 26 22:29:00 2003 Subject: post data on sis950 lpt In-Reply-To: <20030427000807.A16839@synapse.pentanet> Message-ID: I will try to apply your patch next week. I think the linuxbios freeze will be may 1, I'm making sure Eric is happy that he synced up LNXI.com completely. So committers, get your patches in. ron From steve at nexpath.com Sat Apr 26 22:49:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Sat Apr 26 22:49:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: References: Message-ID: <3EAB5125.4030106@nexpath.com> ron minnich wrote: > my mistake. What I have done is this for now: > dd if=elfImage of=/dev/hda bs=4096 skip=1 > > preserves partition information. A kludge, but it works. > IMHO it is a better solution than non-standard partitioning. Actually I think the region past the first sector (512 bytes) is unused, if I am reading http://www.ata-atapi.com/hiwtab.htm correctly. One should also point out that this overlaps into partition 1, if that is not obvious, so partition 1 should be large enough and unused. -Steve From aip at cwlinux.com Sun Apr 27 03:17:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Sun Apr 27 03:17:01 2003 Subject: [COMMIT] epia-m support Message-ID: <20030427155106.A31982@mail.cwlinux.com> I'm pleased to announce EPIA-M is now supported. I would like to thank Nougen for sponsoring this. At this moment, the checkin supports ddr ram init, etherboot, linux and proper irq routing. However, VGA is not done. There is bug where serial baudrate always start in 57600 from cold boot even 115200 is set, but reset from award bios works fine. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From adam at cfar.umd.edu Sun Apr 27 03:33:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sun Apr 27 03:33:01 2003 Subject: [COMMIT] epia-m support In-Reply-To: <20030427155106.A31982@mail.cwlinux.com> Message-ID: <20030427040810.R35121-100000@www.missl.cs.umd.edu> Sounds like great news. Now if I only had an EPIA-M board :-P FYI: people have gotten VGA and ADLO (fully working; with VGA!) to work successfully under EPIA-M. It seems there may be some NDA's issues to it may be few months before they can contribute back. Anyway, bottom line is that it shouldn't be that hard. On Sun, 27 Apr 2003, Andrew Ip wrote: > > I'm pleased to announce EPIA-M is now supported. I would like > to thank Nougen for sponsoring this. At this moment, the checkin > supports ddr ram init, etherboot, linux and proper irq routing. However, > VGA is not done. There is bug where serial baudrate always start in > 57600 from cold boot even 115200 is set, but reset from award bios works > fine. > > -Andrew > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From aip at cwlinux.com Sun Apr 27 03:50:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Sun Apr 27 03:50:01 2003 Subject: [COMMIT] epia-m support In-Reply-To: <20030427040810.R35121-100000@www.missl.cs.umd.edu>; from Adam Sulmicki on Sun, Apr 27, 2003 at 04:13:09AM -0400 References: <20030427155106.A31982@mail.cwlinux.com> <20030427040810.R35121-100000@www.missl.cs.umd.edu> Message-ID: <20030427162422.A32313@mail.cwlinux.com> Adam, > Sounds like great news. Now if I only had an EPIA-M board :-P > FYI: people have gotten VGA and ADLO (fully working; with VGA!) to work > successfully under EPIA-M. It seems there may be some NDA's issues to it > may be few months before they can contribute back. Is it? I would like to try it first. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From hcyun at etri.re.kr Sun Apr 27 06:21:01 2003 From: hcyun at etri.re.kr (hcyun at etri.re.kr) Date: Sun Apr 27 06:21:01 2003 Subject: [COMMIT] epia-m support Message-ID: <8470181DABD5D511B3E700D0B7A8AC4A9CD730@cms3> Well, I also made a linuxbios port for epia-m and ADLO and bochs bios patch which can run original vgabios of epia-m (Adam helped me a lot). So I can see vga text console and also I can run XFree86 or what so ever. And also my collegue made a framebuffer driver. However, because of NDA, I'm afraid I can't share it with you for now. (I don't know anything about specific details of NDA, but my boss told me to do so until our project is done). However, as for ADLO, I think If you have the CLE266 programming guide, you can do it by implementing bios APIs of chapter 7 onto bochs bios code and slight modification of loader.s of ADLO. BTW, Andrew I have a question to you how you can release code. You may need via documents which require NDA. Is VIA approved or sponsoring you? Regard, Heechul > -----Original Message----- > From: Andrew Ip [mailto:aip at cwlinux.com] > Sent: Sunday, April 27, 2003 5:24 PM > To: Adam Sulmicki > Cc: linuxbios at clustermatic.org > Subject: Re: [COMMIT] epia-m support > > > Adam, > > > Sounds like great news. Now if I only had an EPIA-M board :-P > > FYI: people have gotten VGA and ADLO (fully working; with > VGA!) to work > > successfully under EPIA-M. It seems there may be some NDA's > issues to it > > may be few months before they can contribute back. > Is it? I would like to try it first. > > -Andrew > > -- > Andrew Ip > Email: aip at cwlinux.com > Tel: (852) 2542 2046 > Fax: (852) 2542 2036 > Mobile: (852) 9201 9866 > > Cwlinux Limited > Unit 202B 2/F Lai Cheong Factory Building, > 479-479A Castle Peak Road, > Lai Chi Kok, Kowloon, > Hong Kong. > > Tel: (852)2542 2046 > Fax: (852)2542 2036 > > For public pgp key, please obtain it from http://www.keyserver.net/en. > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -------------- next part -------------- An HTML attachment was scrubbed... URL: From dkotian3 at vsnl.net Sun Apr 27 08:43:00 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Sun Apr 27 08:43:00 2003 Subject: Using ADLO with Linuxbios Message-ID: <20030427135047.5854D4FE86@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From root at hamburg.de Sun Apr 27 09:00:01 2003 From: root at hamburg.de (Felix Kloeckner) Date: Sun Apr 27 09:00:01 2003 Subject: low compression of kernel Message-ID: <20030427153246.A2054@synapse.pentanet> hello, just a simple question: what's the reason for using only a compression level of 3 instead of default or even maximum? 9 works fine for me and give me some additional space for the kernel "makerule linux.bin.gz: linux.bin ;gzip -f -3 linux.bin" in "src/arch/i386/config/make.base", line 39 felix From adam at cfar.umd.edu Sun Apr 27 10:20:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sun Apr 27 10:20:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <20030427135047.5854D4FE86@bom6.vsnl.net.in> Message-ID: <20030427110027.X35121-100000@www.missl.cs.umd.edu> when you needs bios services, ex when booting windows. On Sun, 27 Apr 2003 dkotian3 at vsnl.net wrote: > question: > When exactly one goes for ADLO, because the direct > elf image boot is also possible from LinuxBIOS. > Could someone please elaborate on this. > > Regards > Deepak > > > > > rminnich at lanl.gov wrote > On Sat, 26 Apr 2003, Shubhangi Jadhav wrote: > > > cat elfimage > /dev/hda1 > > I made a mistake. Sorry. This is what I had to do. > > dd if=elfImage of=/dev/hda bs=4096 seek=1 > > This leaves partition tables, mbr, etc. intact. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From dkotian3 at vsnl.net Sun Apr 27 10:35:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Sun Apr 27 10:35:01 2003 Subject: Using ADLO with Linuxbios Message-ID: <20030427154216.156D04FE81@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From dkotian3 at vsnl.net Sun Apr 27 10:41:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Sun Apr 27 10:41:01 2003 Subject: Using ADLO with Linuxbios Message-ID: <20030427154542.071AD4FE8F@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From adam at cfar.umd.edu Sun Apr 27 10:46:13 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Sun Apr 27 10:46:13 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <20030427154216.156D04FE81@bom6.vsnl.net.in> Message-ID: <20030427111849.U35121-100000@www.missl.cs.umd.edu> http://www.missl.cs.umd.edu/Projects/sebos/winint/index2.html On Sun, 27 Apr 2003 dkotian3 at vsnl.net wrote: > Thanks.. > What are "bios services", could you please give > an example and which is not needed for Linux.. > > Regards > Deepak > > > > adam at cfar.umd.edu wrote > > when you needs bios services, ex when booting windows. > > On Sun, 27 Apr 2003 dkotian3 at vsnl.net wrote: > > > question: > > When exactly one goes for ADLO, because the direct > > elf image boot is also possible from LinuxBIOS. > > Could someone please elaborate on this. > > > > Regards > > Deepak > > > > > > > > > > rminnich at lanl.gov wrote > > On Sat, 26 Apr 2003, Shubhangi Jadhav wrote: > > > > > cat elfimage > /dev/hda1 > > > > I made a mistake. Sorry. This is what I had to do. > > > > dd if=elfImage of=/dev/hda bs=4096 seek=1 > > > > This leaves partition tables, mbr, etc. intact. > > > > ron > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rminnich at lanl.gov Sun Apr 27 11:21:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Sun Apr 27 11:21:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <3EAB5125.4030106@nexpath.com> Message-ID: On Sat, 26 Apr 2003, Steve Gehlbach wrote: > IMHO it is a better solution than non-standard partitioning. Actually I > think the region past the first sector (512 bytes) is unused, if I am > reading http://www.ata-atapi.com/hiwtab.htm correctly. you are but Plan 9 uses the second 512 bytes and just to be safe I decided to put it at 4096. > One should also point out that this overlaps into partition 1, if that > is not obvious, so partition 1 should be large enough and unused. yes. ron From ebiederman at lnxi.com Sun Apr 27 12:08:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Sun Apr 27 12:08:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EAB46AE.6090005@nexpath.com> References: <3EAB46AE.6090005@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > >The BOOT_IDE driver has been modified in strange > > ways to find things in the first partition and I have not had a chance > > to look over the code to see what is going on there yet. > > > > Not strange to me, by default it simply skips 63x512 bytes (0x7e00), which is > normally for most disks today where partition 1 starts. Of course, it should > read and decode the partition table, but this is much harder. No. I very deliberately do not read partition tables. And not because it is difficult, but more because there is not a consensus (especially across architectures) on what partition tables should look like. The code used to work just like the etherboot code and read the first 8K but that has obviously changed while I was not looking. Current x86 partition tables are limited to 2TB disks. Not an immediate problem I admit but with 120+GB drives today it is a limit we are quickly approaching. Eric From ebiederman at lnxi.com Sun Apr 27 12:16:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Sun Apr 27 12:16:01 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <3EAB5125.4030106@nexpath.com> References: <3EAB5125.4030106@nexpath.com> Message-ID: Steve Gehlbach writes: > ron minnich wrote: > > > my mistake. What I have done is this for now: > > dd if=elfImage of=/dev/hda bs=4096 skip=1 > > preserves partition information. A kludge, but it works. > > > > IMHO it is a better solution than non-standard partitioning. Actually I think > the region past the first sector (512 bytes) is unused, if I am reading > http://www.ata-atapi.com/hiwtab.htm correctly. Right the region past the first sector (512 bytes) is unused. Which is why I just move up the start of first partition into the unused space. I don't see how using the standard partition tables to force a partition to a particular location on the disk is non-standard partitioning. I agree it drops compatibility with ancient versions of DOS but otherwise it is just reserving space. > One should also point out that this overlaps into partition 1, if that is not > obvious, so partition 1 should be large enough and unused. So why should partition 1 not contain the correct information. If you really want to dig into this, my proof of concept utility that will allow much more interesting layouts than starting an ELF image within 8K of the start of the disk. ftp://ftp.lnxi.com/pub/src/linuxbios/installelf.c With that you can just put start of the ELF image in the unused space (normally 63 sectors I think) and put the ELF image on a filesystem. Eric From stuge-linuxbios at cdy.org Sun Apr 27 23:47:01 2003 From: stuge-linuxbios at cdy.org (Peter Stuge) Date: Sun Apr 27 23:47:01 2003 Subject: Debugging the linuxBIOS code In-Reply-To: References: Message-ID: <20030428041050.GG26298@foo.birdnet.se> On Wed, Apr 23, 2003 at 10:05:06AM +0100, Colin Hall wrote: > This e-mail transmission is intended for the named addressee only. > Its contents are private and confidential and should not be read, > copied or disclosed by any other person. Please do not send confidential email to public mailing lists. //Peter From shubhangi.jadhav at patni.com Mon Apr 28 00:35:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Mon Apr 28 00:35:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: I give the following options in my config file option BOOT_IDE=1 option IDE_BOOT_DRIVE=2 option ONE_TRACK=32 commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 option USE_ELF_BOOT=1 Is this correct if I want Linuxbios to directly boot the elfimage? However when I try to build linuxbios using these options I get an error- file linux.bin not found. Now why does it look for this file. Do I need to specify some payload in the config file. what payload is it? Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of ron minnich Sent: Sunday, April 27, 2003 8:26 AM To: Shubhangi Jadhav Cc: Eric W. Biederman; Steve Gehlbach; Deepak Kotian; Adam Agnew; bendany; Linuxbios at clustermatic.org Subject: RE: Using ADLO with Linuxbios On Sat, 26 Apr 2003, Shubhangi Jadhav wrote: > cat elfimage > /dev/hda1 I made a mistake. Sorry. This is what I had to do. dd if=elfImage of=/dev/hda bs=4096 seek=1 This leaves partition tables, mbr, etc. intact. ron _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From hansolofalcon at worldnet.att.net Mon Apr 28 00:49:00 2003 From: hansolofalcon at worldnet.att.net (Gregg C Levine) Date: Mon Apr 28 00:49:00 2003 Subject: Debugging the linuxBIOS code In-Reply-To: <20030428041050.GG26298@foo.birdnet.se> Message-ID: <000001c30d46$47b7b740$0100a8c0@who5> Hello again from Gregg C Levine Peter that is a good observation. A decidedly good one. Colin, regarding your disclaimer, is it possible to have it disabled for traffic on whatever mailing lists that you belong to? Or at least have a disclaimer of your own inserted, stating that it is company policy to use, and not yours? ------------------- Gregg C Levine hansolofalcon at worldnet.att.net ------------------------------------------------------------ "The Force will be with you...Always." Obi-Wan Kenobi "Use the Force, Luke."? Obi-Wan Kenobi (This company dedicates this E-Mail to General Obi-Wan Kenobi ) (This company dedicates this E-Mail to Master Yoda ) > -----Original Message----- > From: linuxbios-admin at clustermatic.org [mailto:linuxbios- > admin at clustermatic.org] On Behalf Of Peter Stuge > Sent: Monday, April 28, 2003 12:11 AM > To: Colin Hall > Cc: linuxbios at clustermatic.org > Subject: Re: Debugging the linuxBIOS code > > On Wed, Apr 23, 2003 at 10:05:06AM +0100, Colin Hall wrote: > > This e-mail transmission is intended for the named addressee only. > > Its contents are private and confidential and should not be read, > > copied or disclosed by any other person. > > Please do not send confidential email to public mailing lists. > > > //Peter > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From yenneo at oreka.com Mon Apr 28 02:34:01 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Mon Apr 28 02:34:01 2003 Subject: Epia -m : EEPROM not found Message-ID: <200304280708.JAA90294@mailhub1.isdnet.net> When i run the flash utility in flash_and_burn directory, it check for the rom chip but doesn?t find it. At the end it give this message : "EEPROM not found". I?m using a bios savior RD1 but i think it shouldn?t do problems. Can i use this utility with an epia-m or should i use an other one? -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From yenneo at oreka.com Mon Apr 28 03:13:01 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Mon Apr 28 03:13:01 2003 Subject: Epia-m : downloaded missing =?iso-8859-1?q?files=3F?= Message-ID: <200304280747.JAA59831@mailhub10.isdnet.net> I've just downloaded the latest code for epia-m but it looks like thoses two files are missings : freebios/src/northbridge/via/vt8623/config and freebios/src/southbridge/via/vt8235/config ... -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From rminnich at lanl.gov Mon Apr 28 09:11:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 09:11:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: On Mon, 28 Apr 2003, Shubhangi Jadhav wrote: > I give the following options in my config file > > option BOOT_IDE=1 > option IDE_BOOT_DRIVE=2 > option ONE_TRACK=32 > commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 > option USE_ELF_BOOT=1 > > Is this correct if I want Linuxbios to directly boot the elfimage? > However when I try to build linuxbios using these options I get an error- > file linux.bin not found. Now why does it look for this file. > Do I need to specify some payload in the config file. what payload is it? > you should look at the sample config files in the source tree. Try looking in util/config ron From rminnich at lanl.gov Mon Apr 28 09:13:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 09:13:00 2003 Subject: Epia -m : EEPROM not found In-Reply-To: <200304280708.JAA90294@mailhub1.isdnet.net> Message-ID: On Mon, 28 Apr 2003, Philippe CABANNES wrote: > Can i use this utility with an epia-m or should i use an other one? They key question: is flash write enabled Philippe send me the output of flash_rom ron From rminnich at lanl.gov Mon Apr 28 09:13:07 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 09:13:07 2003 Subject: Epia-m : downloaded missing =?iso-8859-1?q?files=3F?= In-Reply-To: <200304280747.JAA59831@mailhub10.isdnet.net> Message-ID: On Mon, 28 Apr 2003, Philippe CABANNES wrote: > I've just downloaded the latest code for epia-m but it looks like > thoses two files are missings : > > freebios/src/northbridge/via/vt8623/config > > and > > freebios/src/southbridge/via/vt8235/config > sorry but epia-m has gotten yanked temporarily to ensure that there are no NDA issues. ron From nathanael at gnat.ca Mon Apr 28 10:40:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Mon Apr 28 10:40:01 2003 Subject: Bios flashing continued Message-ID: <044AB1B8-798C-11D7-AA31-0003931B4D6A@gnat.ca> Hello again, I'm at the point trying to figure out how to get linuxbios flashed onto the flash chip. The board I have is a sis530. The flash_on program just tells me it can't access the sis 950 and ./flash_rom gives me a bunch of messages of what it is trying to find. I'm guessing because flash_on didn't work. I've built a kernel with all the chips set as modules and preloaded all the modules. Ron told me that the 550 could be flashed and this is probably not much different, which still may be the case but well... at the moment it doesn't. So I'm wondering, what is the method I would use to go about finding how to enable the flash so that I can burn it. The flash_on is trying to get the sis 950 which was the superio chip used on most of those board. My board is the ITE 8661F. pointers on where to proceed would be appreciated. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Mon Apr 28 10:51:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 10:51:01 2003 Subject: Bios flashing continued In-Reply-To: <044AB1B8-798C-11D7-AA31-0003931B4D6A@gnat.ca> Message-ID: if you get really stuck, try the MTD drivers from linux. I am hoping Ollie can help tell what's going on. ron From rsmith at bitworks.com Mon Apr 28 10:56:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 10:56:01 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: References: <20030425164501.B28568-100000@www.missl.cs.umd.edu> Message-ID: <3EAD48B9.7000500@bitworks.com> Eric W. Biederman wrote: >>PCMCIA mount? >>Etherboot wont look at pcmcia devices. If its going to find a compact >>flash card, it will be because there's a compact flash on IDE adapter. >>Then it will see it as a hard disk. > > You may have a point. Richard how many IDE disks do you have? Slight miscommunication there. The CF is pcmcia mounted on my _developement_ system. I was just trying to clarify why my disk mount was /dev/hde for the 'cat >' rather than hda. Looks like I just confused things further. :( On my _target_ device the CF is a connector on the PCB. Its hda and the only IDE device I have. > I still favor the partition is to late on the disk theory but, not checking > the CF works as well. > > For a pure test case it is possible to just dd the ELF image onto the start > of the disk. If you don't need partitions. That is the way Ron is currently > using the code. Thanks I'll give that a try. I also see just where fdisk put the 1st partition. -- Richard A. Smith rsmith at bitworks.com From steve at nexpath.com Mon Apr 28 11:00:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Mon Apr 28 11:00:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: References: Message-ID: <3EAD4BD0.2030302@nexpath.com> Shubhangi Jadhav wrote: > I give the following options in my config file > > option BOOT_IDE=1 > option IDE_BOOT_DRIVE=2 > option ONE_TRACK=32 > commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 > option USE_ELF_BOOT=1 > > Is this correct if I want Linuxbios to directly boot the elfimage? > However when I try to build linuxbios using these options I get an error- > file linux.bin not found. Now why does it look for this file. > Do I need to specify some payload in the config file. what payload is it? > I believe you will see this error if you do not set "linux /usr/src/linux" or where ever your linux kernel source tree is. Put this in, and I think this error will go away. But I should mention that I have never gotten this to boot with USE_ELF_BOOT=1, though I am told it should. It has only worked for me using vmlinux.gin.gz, ie, direct loading a non-elf image. -Steve From russell at abc9986.demon.co.uk Mon Apr 28 12:47:00 2003 From: russell at abc9986.demon.co.uk (Russell Gower) Date: Mon Apr 28 12:47:00 2003 Subject: Via EPIA Message-ID: <006001c30daa$9f405260$0f01a8c0@winxp> Could someone please point me in the right direction. I need to enable wake on Ring for the onboard serial port and/or AC power recovery on a via epia 5000 mainboard. Thanks Russell From dkotian3 at vsnl.net Mon Apr 28 13:58:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 28 13:58:01 2003 Subject: Using ADLO with Linuxbios References: <3EAD4BD0.2030302@nexpath.com> Message-ID: <001601c30db4$294b9740$343e41db@vsnl.net> Good information. Can someone please stick out the config file content over here for the following situation, Case 1: 1. When LinuxBIOS in flash ROM and image is on compact flash without hdc1. Case 2: 2. Any sample config file which uses ADLO payload with LinuxBIOS on FLASHOMS. Regards Deepak ----- Original Message ----- From: "Steve Gehlbach" To: "Shubhangi Jadhav" Cc: "ron minnich" ; "Eric W. Biederman" ; "Deepak Kotian" ; "Adam Agnew" ; "bendany" ; Sent: Monday, April 28, 2003 9:12 PM Subject: Re: Using ADLO with Linuxbios > Shubhangi Jadhav wrote: > > I give the following options in my config file > > > > option BOOT_IDE=1 > > option IDE_BOOT_DRIVE=2 > > option ONE_TRACK=32 > > commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 > > option USE_ELF_BOOT=1 > > > > Is this correct if I want Linuxbios to directly boot the elfimage? > > However when I try to build linuxbios using these options I get an error- > > file linux.bin not found. Now why does it look for this file. > > Do I need to specify some payload in the config file. what payload is it? > > > > I believe you will see this error if you do not set "linux > /usr/src/linux" or where ever your linux kernel source tree is. Put > this in, and I think this error will go away. > > But I should mention that I have never gotten this to boot with > USE_ELF_BOOT=1, though I am told it should. It has only worked for me > using vmlinux.gin.gz, ie, direct loading a non-elf image. > > -Steve > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios From dkotian3 at vsnl.net Mon Apr 28 14:04:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 28 14:04:00 2003 Subject: Using ADLO with Linuxbios References: <3EAD4BD0.2030302@nexpath.com> <001601c30db4$294b9740$343e41db@vsnl.net> Message-ID: <004301c30db5$4ea0ab60$343e41db@vsnl.net> Sorry, I meant > Case 1: > 1. When LinuxBIOS in flash ROM and image is on compact flash with /dev/hdc1 block device and no harddisk. > ----- Original Message ----- From: "Deepak Kotian" To: "Steve Gehlbach" ; "Shubhangi Jadhav" Cc: "ron minnich" ; "Eric W. Biederman" ; "Adam Agnew" ; "bendany" ; Sent: Monday, April 28, 2003 11:58 PM Subject: Re: Using ADLO with Linuxbios > Good information. > > > Can someone please stick out the config file content over here for the > following situation, > > Case 1: > 1. When LinuxBIOS in flash ROM and image is on compact flash without hdc1. > > Case 2: > 2. Any sample config file which uses ADLO payload with LinuxBIOS on > FLASHOMS. > > > > Regards > Deepak > > > > ----- Original Message ----- > From: "Steve Gehlbach" > To: "Shubhangi Jadhav" > Cc: "ron minnich" ; "Eric W. Biederman" > ; "Deepak Kotian" ; "Adam Agnew" > ; "bendany" ; > > Sent: Monday, April 28, 2003 9:12 PM > Subject: Re: Using ADLO with Linuxbios > > > > Shubhangi Jadhav wrote: > > > I give the following options in my config file > > > > > > option BOOT_IDE=1 > > > option IDE_BOOT_DRIVE=2 > > > option ONE_TRACK=32 > > > commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 > > > option USE_ELF_BOOT=1 > > > > > > Is this correct if I want Linuxbios to directly boot the elfimage? > > > However when I try to build linuxbios using these options I get an > error- > > > file linux.bin not found. Now why does it look for this file. > > > Do I need to specify some payload in the config file. what payload is > it? > > > > > > > I believe you will see this error if you do not set "linux > > /usr/src/linux" or where ever your linux kernel source tree is. Put > > this in, and I think this error will go away. > > > > But I should mention that I have never gotten this to boot with > > USE_ELF_BOOT=1, though I am told it should. It has only worked for me > > using vmlinux.gin.gz, ie, direct loading a non-elf image. > > > > -Steve > > > > _______________________________________________ > > Linuxbios mailing list > > Linuxbios at clustermatic.org > > http://www.clustermatic.org/mailman/listinfo/linuxbios > > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Mon Apr 28 14:11:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Mon Apr 28 14:11:00 2003 Subject: Using ADLO with Linuxbios References: Message-ID: <007b01c30db6$3dd0aa00$343e41db@vsnl.net> I tried this a similar command on a Compact flash IDE after partioning. Say. > dd if=elfImage of=/dev/hdc1 bs=4096 seek=1 It said Invalid argument "/dev/hdc1" Any inputs, must be simple.. Regards Deepak ----- Original Message ----- From: "ron minnich" To: "Shubhangi Jadhav" Cc: "Eric W. Biederman" ; "Steve Gehlbach" ; "Deepak Kotian" ; "Adam Agnew" ; "bendany" ; Sent: Sunday, April 27, 2003 8:25 AM Subject: RE: Using ADLO with Linuxbios > On Sat, 26 Apr 2003, Shubhangi Jadhav wrote: > > > cat elfimage > /dev/hda1 > > I made a mistake. Sorry. This is what I had to do. > > dd if=elfImage of=/dev/hda bs=4096 seek=1 > > This leaves partition tables, mbr, etc. intact. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From nathanael at gnat.ca Mon Apr 28 14:33:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Mon Apr 28 14:33:01 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: On Monday, April 28, 2003, at 09:23 AM, ron minnich wrote: > if you get really stuck, try the MTD drivers from linux. I am hoping > Ollie > can help tell what's going on. So I have been unable so far to make any progress. Using google to search for bios flashing mtd gets me nothing of value. I can't get the flash_and_burn stuff to do anything for me so I hope Ollie can help as well. If there are any resources I should look/read please pass them this way as well. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From ebiederman at lnxi.com Mon Apr 28 14:39:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 28 14:39:01 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <20030428183028.GA11030@tsn.or.jp> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> Message-ID: SONE Takeshi writes: > On Fri, Apr 25, 2003 at 06:55:37PM -0600, Eric W. Biederman wrote: > > > > In cpu/p6/earlymtrr.inc, 0xf0000-0xfffff is left uncached. > > > > (Only 0-640KB is cached, 640KB-1M is uncached) > > > > I will try adding code there to set fixed MTRR for 0xf0000-0xfffff > > > > WP caching. > > > > It should be harmless for other boards that do not need this, > > > > what do you think? > > > > > > > > > > Not sure. Need to see if the RAM is turned on for this region later after > the > > > > C-code starts, if so, then the cache memory type may have to be changed from > WP > > > > to normal. > > I tried it anyway and it worked great. > Now I can't see anything between 'Copying LinuxBIOS to ram' and > IDE loading message because it's too fast! > > > The generic code should handle this. I believe all of the mtrrs are rebuilt. > > I think it's rebuilt in mtrr.c. > > My modification is below. ``Write Protect'' in the mtrrs does not mean write protected. It is a strange messed up form of write-through. In particular it dumps the cache on writes it does not forbid writes. Unless you know of a situation where write protect is more appropriate please use write-through. It is less confusing, and since no one is writing to that area anyway it gives the exact same result, reads are cached. Eric From ebiederman at lnxi.com Mon Apr 28 14:46:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 28 14:46:00 2003 Subject: low compression of kernel In-Reply-To: <20030427153246.A2054@synapse.pentanet> References: <20030427153246.A2054@synapse.pentanet> Message-ID: Felix Kloeckner writes: > hello, > > just a simple question: > > what's the reason for using only a compression level of 3 instead of default or > even maximum? > > 9 works fine for me and give me some additional space for the kernel > > "makerule linux.bin.gz: linux.bin ;gzip -f -3 linux.bin" in > "src/arch/i386/config/make.base", line 39 Two reasons. 1) That code has been depricated for over a year. 2) There were bugs seen when decompressing a kernel when presented with a kernel with maximum decompression. Eric From steve at nexpath.com Mon Apr 28 15:11:01 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Mon Apr 28 15:11:01 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> Message-ID: <3EAD866C.3020508@nexpath.com> Eric W. Biederman wrote: > ``Write Protect'' in the mtrrs does not mean write protected. It is a strange > messed up form of write-through. In particular it dumps the cache on writes > it does not forbid writes. > > Unless you know of a situation where write protect is more appropriate please > use write-through. It is less confusing, and since no one is writing to that > area anyway it gives the exact same result, reads are cached. > > Eric Good point, but do you think it would matter for reflashing?. I don't know if Linux resets the MTRRs when booting, or assumes the BIOS has already done that. I got into the habit of using the WP setting because MS does it this way for the ROM areas for the Xbox (copying MS is probably not a good reason for doing anything!). For all to reference, Intel manual: ? Write protected (WP)?Reads come from cache lines when possible, and read misses cause cache fills. Writes are propagated to the system bus and cause corresponding cache lines on all processors on the bus to be invalidated. Speculative reads are allowed. This memory type is available in the Pentium 4, Intel Xeon, and P6 family processors by programming the MTRRs (seeTable 10-6). --------- -Steve From steve at nexpath.com Mon Apr 28 15:12:00 2003 From: steve at nexpath.com (Steve Gehlbach) Date: Mon Apr 28 15:12:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <007b01c30db6$3dd0aa00$343e41db@vsnl.net> References: <007b01c30db6$3dd0aa00$343e41db@vsnl.net> Message-ID: <3EAD86DD.3010305@nexpath.com> Deepak Kotian wrote: > I tried this a similar command on a Compact flash IDE after partioning. > Say. > >>dd if=elfImage of=/dev/hdc1 bs=4096 seek=1 >>--------------------------^^ What Ron used was /dev/hdc, ie, the drive from the beginning, not partition 1. The 4096 and seek=1 skips over the partition table and puts the image into the last part of the MBR/partition table area overlapping into partition 1. It's equivalent to seeking 8 sectors from the beginning when viewing the disk as linear sector addressing. I wouldn't do this sort of thing on any drive you care about, easy to make typos and wipe out the disk (or another disk if your fingers type /dev/hda when your brain says hdc). -Steve From ebiederman at lnxi.com Mon Apr 28 15:17:01 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 28 15:17:01 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <3EAD866C.3020508@nexpath.com> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> <3EAD866C.3020508@nexpath.com> Message-ID: Steve Gehlbach writes: > Eric W. Biederman wrote: > > > ``Write Protect'' in the mtrrs does not mean write protected. It is a strange > > > messed up form of write-through. In particular it dumps the cache on writes > > it does not forbid writes. > > Unless you know of a situation where write protect is more appropriate please > > use write-through. It is less confusing, and since no one is writing to that > > area anyway it gives the exact same result, reads are cached. > > Eric > > Good point, but do you think it would matter for reflashing?. Neither type will work reliably when reflashing. The read caching prevents polling of the status bits, to function properly. > I don't know if > Linux resets the MTRRs when booting, or assumes the BIOS has already done that. LinuxBIOS does that. And it is at least legal to manipulate that in an mtd map driver if necessary. > I got into the habit of using the WP setting because MS does it this way for the > ROM areas for the Xbox (copying MS is probably not a good reason for doing > anything!). WP is the least aggressive form of caching. And in principle I don't have any problems with it. But it's name is very non-intuitive. > For all to reference, Intel manual: > ? Write protected (WP)?Reads come from cache lines when possible, and read > > misses cause cache fills. Writes are propagated to the system bus and cause > corresponding cache lines on all processors on the bus to be > invalidated. Speculative reads are allowed. This memory type is available in the > > Pentium 4, Intel Xeon, and P6 family processors by > programming the MTRRs (seeTable 10-6). > --------- > > -Steve From rsmith at bitworks.com Mon Apr 28 15:20:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 15:20:01 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: <3EAD48B9.7000500@bitworks.com> References: <20030425164501.B28568-100000@www.missl.cs.umd.edu> <3EAD48B9.7000500@bitworks.com> Message-ID: <3EAD86A3.7000401@bitworks.com> Richard Smith wrote: >> For a pure test case it is possible to just dd the ELF image onto the >> start >> of the disk. If you don't need partitions. That is the way Ron is >> currently >> using the code. > > Thanks I'll give that a try. I also see just where fdisk put the 1st > partition. Ok... I'm booting now. Erik your hunch was correct. /dev/hda1 must have been outside the 8k section. I'm not really sure how to read fdisk's info and translate it into a physical offset on the disk but going directly to /dev/hda worked. Although Ron's dd command gave me an "Invalid argument /dev/hde" it appears to have written the data on to the disk as I can boot and the partitition table is still intact. Thanks for all the help. -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Mon Apr 28 15:23:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 15:23:00 2003 Subject: vgabios emulation status In-Reply-To: References: <20030427153246.A2054@synapse.pentanet> Message-ID: <3EAD8777.6070702@bitworks.com> Ok the last thing I need is to get LB to run the vga bios for the 69030 chips (there are 2). I haven't seen any activity on the vgabios facility of linuxbios in a while. Can someone give me a rundown on the current status and what needs to be done to make that happen. -- Richard A. Smith rsmith at bitworks.com From ebiederman at lnxi.com Mon Apr 28 15:31:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Mon Apr 28 15:31:00 2003 Subject: etherboot dosen't find my elfkernel In-Reply-To: <3EAD86A3.7000401@bitworks.com> References: <20030425164501.B28568-100000@www.missl.cs.umd.edu> <3EAD48B9.7000500@bitworks.com> <3EAD86A3.7000401@bitworks.com> Message-ID: Richard Smith writes: > Richard Smith wrote: > > >> For a pure test case it is possible to just dd the ELF image onto the start > >> of the disk. If you don't need partitions. That is the way Ron is currently > > >> using the code. > > Thanks I'll give that a try. I also see just where fdisk put the 1st > > partition. > > Ok... I'm booting now. Erik your hunch was correct. /dev/hda1 must have been > outside the 8k section. I'm not really sure how to read fdisk's info and > translate it into a physical offset on the disk but going directly to /dev/hda > worked. In the normal case just change the units to sectors. The 'u' command. A single sector is 512 bytes or 1/2 KiloByte. > Although Ron's dd command gave me an "Invalid argument /dev/hde" it appears to > have written the data on to the disk as I can boot and the partitition table is > still intact. Good. Eric From rsmith at bitworks.com Mon Apr 28 15:46:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 15:46:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EAD86DD.3010305@nexpath.com> References: <007b01c30db6$3dd0aa00$343e41db@vsnl.net> <3EAD86DD.3010305@nexpath.com> Message-ID: <3EAD8CF5.5010208@bitworks.com> Steve Gehlbach wrote: >>> dd if=elfImage of=/dev/hdc1 bs=4096 seek=1 > > >>--------------------------^^ > > What Ron used was /dev/hdc, ie, the drive from the beginning, not > partition 1. The 4096 and seek=1 skips over the partition table and > puts the image into the last part of the MBR/partition table area On my system if I add the seek option to dd it whines about /dev/hde being an invalid argument as well so I don't think its the partition number thats causing the issue. 'dd if=elfimage of=/dev/hde bs=4096 seek=1' still generates and Invalid argument but appears to work anyway as I get the proper ammount of records in/out and the CF boots. -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Mon Apr 28 15:48:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 15:48:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EAD86DD.3010305@nexpath.com> Message-ID: On Mon, 28 Apr 2003, Steve Gehlbach wrote: > I wouldn't do this sort of thing on any drive you care about, easy to > make typos and wipe out the disk (or another disk if your fingers type > /dev/hda when your brain says hdc). ah, yes, I've done that. :-) ron From rminnich at lanl.gov Mon Apr 28 15:51:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 15:51:01 2003 Subject: vgabios emulation status In-Reply-To: <3EAD8777.6070702@bitworks.com> Message-ID: On Mon, 28 Apr 2003, Richard Smith wrote: > Ok the last thing I need is to get LB to run the vga bios for the 69030 > chips (there are 2). > > I haven't seen any activity on the vgabios facility of linuxbios in a > while. Can someone give me a rundown on the current status and what > needs to be done to make that happen. it worked on a small scale. But I thought those 69030 parts were so simple you did not need a vga bios? ron From rminnich at lanl.gov Mon Apr 28 15:57:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 15:57:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EAD8CF5.5010208@bitworks.com> Message-ID: On Mon, 28 Apr 2003, Richard Smith wrote: > On my system if I add the seek option to dd it whines about /dev/hde > being an invalid argument as well so I don't think its the partition > number thats causing the issue. > > 'dd if=elfimage of=/dev/hde bs=4096 seek=1' I've never had that happen. Weird. ron From rsmith at bitworks.com Mon Apr 28 16:04:00 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 16:04:00 2003 Subject: vgabios emulation status In-Reply-To: References: Message-ID: <3EAD90FE.9070000@bitworks.com> ron minnich wrote: > it worked on a small scale. But I thought those 69030 parts were so simple > you did not need a vga bios? Oh man I wish but, no way, not in the least. There are bunches of registes that have to be setup. Especially the Flat Panel registers. The 030 is further complicated by the fact that it has 2 independent rendering pipelines. It can drive both a CRT and LCD at 2 different resolutions at the same time. The good thing is that the full docs on the chip are available from asiliants web page. Is there a config anywhere in the tree that shows and example setup? -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Mon Apr 28 16:19:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 16:19:01 2003 Subject: vgabios emulation status In-Reply-To: <3EAD90FE.9070000@bitworks.com> Message-ID: To get vgabios working I think you should first boot linux with serial console, then run the little user-mode vga emulator in the linuxbios tree and see if you can get vga up. This is the easiest way to work out the problems. ron From rsmith at bitworks.com Mon Apr 28 16:31:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 16:31:01 2003 Subject: vgabios emulation status In-Reply-To: References: Message-ID: <3EAD9785.3050407@bitworks.com> ron minnich wrote: > To get vgabios working I think you should first boot linux with serial > console, then run the little user-mode vga emulator in the linuxbios tree > and see if you can get vga up. This is the easiest way to work out the > problems. Ok.. I was hoping to give it a test and see if it "just worked" I don't yet have a useable filesystem setup. I sent you the only one I had. :) I'll see what I can cobble up together. I think I have a 128 MB CF around here somewhere. In the mean time can I just try-and-see? -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Mon Apr 28 19:24:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Mon Apr 28 19:24:01 2003 Subject: errors compiling the vga bios stuff In-Reply-To: <3EAD9785.3050407@bitworks.com> References: <3EAD9785.3050407@bitworks.com> Message-ID: <3EADC000.90708@bitworks.com> Richard Smith wrote: I'm trying to compile in the realmode vga bios stuff and I'm getting an undefined reference to 'pcibios' in function 'biosinit' So far the only place I've found the body of function pcibios is in the src/bioscall/pcibios.c file. That file is not included in my make file. I assume that CONFIG_PCIBIOS=1 would control the inclusion of this file but I can't find any rule in the Makefile that would include this file even though the options is set. Do I have to somehow manually include this file? Also if I set VIDEO_CONSOLE=1 I get a undefined ref to 'beep' in function 'video_tx_byte' but I haven't done much digging on that. I'm out of the office tomorrow so it will be Wed before I can respond to any questions. Thanks in advance -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Mon Apr 28 19:52:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 19:52:01 2003 Subject: vgabios emulation status In-Reply-To: <3EAD9785.3050407@bitworks.com> Message-ID: On Mon, 28 Apr 2003, Richard Smith wrote: > I'll see what I can cobble up together. I think I have a 128 MB CF > around here somewhere. In the mean time can I just try-and-see? sure. try and see, it might shock you by working. option CONFIG_REALMOE_IDT=1 option CONFIG_VGABIOS=1 ron From rminnich at lanl.gov Mon Apr 28 20:02:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Mon Apr 28 20:02:00 2003 Subject: errors compiling the vga bios stuff In-Reply-To: <3EADC000.90708@bitworks.com> Message-ID: dammit. Version creep. OK, I'll take a look. Last time I tried this it worked. For FreeBIOS 2 we're going to have nightly builds of all configurations to make sure this doesn't happen. ron From yapeehuey at hotmail.com Mon Apr 28 20:54:01 2003 From: yapeehuey at hotmail.com (Yap Ee Huey) Date: Mon Apr 28 20:54:01 2003 Subject: Bios flashing continued Message-ID: Hi, I have been looking for a program to flash my eeprom as well. The standard flash programs do not allow me to burn unidentified bios. I have found two very useful and general flash and burn programs : (1)UNIFLASH (DOS program) http://www.pppr.sk/rainbow/programs.html (2)/dev/bios (Linux program) http://www.openbios.info/development/devbios.html Hope it helps u ! Although they don't work on my board -> TUSL2-C (eeprom SST49LF002A) because of some undocumented write protect registers maybe. good luck, eehuey > >Message: 5 >Date: Mon, 28 Apr 2003 13:07:24 -0600 >Subject: Re: Bios flashing continued >From: Nathanael Noblet >To: linuxbios at clustermatic.org > > >On Monday, April 28, 2003, at 09:23 AM, ron minnich wrote: > > > if you get really stuck, try the MTD drivers from linux. I am hoping > > Ollie > > can help tell what's going on. > >So I have been unable so far to make any progress. Using google to >search for bios flashing mtd gets me nothing of value. I can't get the >flash_and_burn stuff to do anything for me so I hope Ollie can help as >well. If there are any resources I should look/read please pass them >this way as well. > >-- >Nathanael Noblet >Gnat Solutions >4604 Monterey Ave NW >Calgary, AB >T3B 5K4 > >T/F 403.288.5360 >C 403.809.5368 > >http://www.gnat.ca/ > > _________________________________________________________________ Download ringtones, logos and picture messages from MSN Malaysia http://www.msn.com.my/mobile/ringtones/default.asp From hcyun at etri.re.kr Tue Apr 29 00:55:00 2003 From: hcyun at etri.re.kr (hcyun at etri.re.kr) Date: Tue Apr 29 00:55:00 2003 Subject: vgabios emulation status Message-ID: <8470181DABD5D511B3E700D0B7A8AC4A9CD736@cms3> AFAIK, the best way of VGA emulation is using ADLO. Because ADLO run boch bios, VGABIOS can be excuted just like as it comes with other commercial bios environment. Heechul -------------- next part -------------- An HTML attachment was scrubbed... URL: From bendany at mistdl.com Tue Apr 29 03:41:01 2003 From: bendany at mistdl.com (bendany) Date: Tue Apr 29 03:41:01 2003 Subject: Another way of using ADLO Message-ID: <032001c30fba$3eacc6b0$2a00a8c0@ben> Hm, It seems that my way is buggy, But it works. ;-) I am trying to use the VGA with LinuxBIOS. since I am trying to use the Xwindows. if you want to start the Xserver, you must have the vgabios. I think when the Xserver startup, it will probe the videocard. but since the LinuxBIOS does not have the videobios. so the Xserver crash. you can using the Xserver with the kernel framebuffer driver. such as fbdev. but it lack the ability of hardware accel. such as play the movie fullscreen. so use the ADLO & bochs bios & vgabios will give you a second way to run the Xserver. since the LinuxBIOS switch the CPU into protected mode early, and the commercial vgabios runs in realmode. so the commercial vgabios cann't run after the LinuxBIOS. so the ADLO will do some preparation works. Load the bochs bios & vgabios. switch the CPU to the realmode. execute the bochs bios. the bochs will try to init the vgabios first.it scan the address 0xC0000 to 0xC7FFF in 2k increments. and then scan the General ROM from 0xC8000 to 0xDFFFF in 2k increments, so if we put the etherboot+AA patch after the address of vgabios. the etherboot will be excuted after bochs init the vgabios. the way why I do this is that the bochs bios is lack the ability of dealing with the real hardware. such as ide harddisk. In my experiment, when I use the IBM 30G harddisk. the speed of lilo & grub load the kernel normally, but when I use the flash IDE. the speed of lilo & grub , or syslinux, is too slow. if you want to run the etherboot with the bochs bios. you must do some work yourself. first, using ADLO to get your own vgabios. my mainboard chipset is sis630. the vgabios is 48kb. plus the etherboot+AA rom. and the bochs bios .altogether 144kb. and the ADLO only provide 3 types of elf loader. so you must create the 145kb loader yourself. and also modify the loader.s. and you will need to modify the etherboot5.0.6+AA patch too. hope help. Best regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: From agnew at cs.umd.edu Tue Apr 29 03:48:00 2003 From: agnew at cs.umd.edu (Adam Agnew) Date: Tue Apr 29 03:48:00 2003 Subject: Another way of using ADLO In-Reply-To: <032001c30fba$3eacc6b0$2a00a8c0@ben> Message-ID: <20030429042757.K44496-100000@www.missl.cs.umd.edu> Hi Bendany, Perhaps I missed something in your email.. But if you've loaded ADLO + Bochs BIOS then you have Int13 support. Why not run Grub from the flash disk then instead of etherboot+AA? Is it because of the speed? - Adam Agnew On Thu, 1 May 2003, bendany wrote: > Hm, It seems that my way is buggy, But it works. ;-) > > I am trying to use the VGA with LinuxBIOS. since I am trying to use the Xwindows. > if you want to start the Xserver, you must have the vgabios. I think when the Xserver startup, it will probe the videocard. but > since the LinuxBIOS does not have the videobios. so the Xserver crash. > you can using the Xserver with the kernel framebuffer driver. such as fbdev. but it lack the ability of hardware accel. such > as play the movie fullscreen. so use the ADLO & bochs bios & vgabios will give you a second way to run the Xserver. > since the LinuxBIOS switch the CPU into protected mode early, and the commercial vgabios runs in realmode. so the commercial > vgabios cann't run after the LinuxBIOS. so the ADLO will do some preparation works. Load the bochs bios & vgabios. switch the > CPU to the realmode. execute the bochs bios. > the bochs will try to init the vgabios first.it scan the address 0xC0000 to 0xC7FFF in 2k increments. and then scan the General > ROM from 0xC8000 to 0xDFFFF in 2k increments, so if we put the etherboot+AA patch after the address of vgabios. the etherboot > will be excuted after bochs init the vgabios. > the way why I do this is that the bochs bios is lack the ability of dealing with the real hardware. such as ide harddisk. In my experiment, > when I use the IBM 30G harddisk. the speed of lilo & grub load the kernel normally, but when I use the flash IDE. the speed of lilo & > grub , or syslinux, is too slow. > > if you want to run the etherboot with the bochs bios. you must do some work yourself. > first, using ADLO to get your own vgabios. my mainboard chipset is sis630. the vgabios is 48kb. plus the etherboot+AA rom. > and the bochs bios .altogether 144kb. and the ADLO only provide 3 types of elf loader. so you must create the 145kb loader yourself. > and also modify the loader.s. > and you will need to modify the etherboot5.0.6+AA patch too. > > hope help. > Best regards. > From yapeehuey at hotmail.com Tue Apr 29 03:58:01 2003 From: yapeehuey at hotmail.com (Yap Ee Huey) Date: Tue Apr 29 03:58:01 2003 Subject: Linuxbios + ADLO Message-ID: Hi hackers, Thanks for all the answers regarding linuxbios + etherboot +tagged kernel image . But I don't want to modify my partitions, so I have changed my approach. I am using ADLO with Linuxbios now, I have: ADLO file -> payload 65k Linuxbios ROM -> linuxbios.rom 64k I specify in my config file : option USE_ELF_BOOT=1 payload payload And then I build a romimage (256k) with payload padded to (192k) dd conv=sync bs=196608 if=payload of=payload.block and linuxbios.rom at the end 64k . cat payload.block linuxbios.rom > romimage Does this make sense ? The result is negative. I can't boot. I have tried to put linuxbios.rom at first 64k and then following with ADLO payload and padded to 256k. Negative result also. What is the correct placement in rom to load linuxbios.rom at the first place and then tell it to go to ADLO payload ? Thanks everyone, eehuey _________________________________________________________________ Using a handphone prepaid card? Reload your credit online! http://www.msn.com.my/reloadredir/default.asp From bendany at mistdl.com Tue Apr 29 04:04:01 2003 From: bendany at mistdl.com (bendany) Date: Tue Apr 29 04:04:01 2003 Subject: Fw: Another way of using ADLO Message-ID: <035301c30fbd$8d29d370$2a00a8c0@ben> ---- Original Message ----- From: "Adam Agnew" To: "bendany" Cc: Sent: Tuesday, April 29, 2003 4:29 PM Subject: Re: Another way of using ADLO > > Hi Bendany, > > Perhaps I missed something in your email.. But if you've loaded ADLO + > Bochs BIOS then you have Int13 support. Why not run Grub from the flash > disk then instead of etherboot+AA? Is it because of the speed? yes. I have try several time. using the ADLO with Int13 support. when I using the LILO & Grub , I works with the IBM harddisk. but when I use the Flash IDE disk. always give me LI 01 01 01 ... with LILO. or with Grub, too slow. but with the etherboot+AA . it sounds great!. :-) > > - Adam Agnew > > > On Thu, 1 May 2003, bendany wrote: > > > From adam at cfar.umd.edu Tue Apr 29 06:58:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Tue Apr 29 06:58:00 2003 Subject: Fw: Another way of using ADLO In-Reply-To: <035301c30fbd$8d29d370$2a00a8c0@ben> Message-ID: <20030429073039.C45447-100000@www.missl.cs.umd.edu> > > Perhaps I missed something in your email.. But if you've loaded ADLO + > > Bochs BIOS then you have Int13 support. Why not run Grub from the flash > > disk then instead of etherboot+AA? Is it because of the speed? > yes. I have try several time. using the ADLO with Int13 support. when I > using the LILO & Grub , I works with > the IBM harddisk. but when I use the Flash IDE disk. always give me LI 01 > 01 01 ... with LILO. or with Grub, too slow. > but with the etherboot+AA . it sounds great!. :-) what exactly are you booting using AA patch? it seems to me that using his patch you simply by pass either lilo or grub. as for lilo, it seems to me that your problem is that you install lilo using different system than you boot from so disk geometry changes and thus the "LI" error. as for grub. well it is weird. are you getting any errors? it seems to me like bochs bios would need fixing. btw: I assume you are using 16 bit PC BIOS modules from etherboot rather than 32 bit EBI modules from said package, right? Interesting. Opens lots of new possibilities. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From bendany at mistdl.com Tue Apr 29 07:14:00 2003 From: bendany at mistdl.com (bendany) Date: Tue Apr 29 07:14:00 2003 Subject: Fw: Another way of using ADLO References: <20030429073039.C45447-100000@www.missl.cs.umd.edu> Message-ID: <03a601c30fd7$fb4675b0$2a00a8c0@ben> ----- Original Message ----- From: "Adam Sulmicki" To: "bendany" Cc: "Linux BIOS Mailing List" Sent: Tuesday, April 29, 2003 7:39 PM Subject: Re: Fw: Another way of using ADLO > > > > Perhaps I missed something in your email.. But if you've loaded ADLO + > > > Bochs BIOS then you have Int13 support. Why not run Grub from the flash > > > disk then instead of etherboot+AA? Is it because of the speed? > > > yes. I have try several time. using the ADLO with Int13 support. when I > > using the LILO & Grub , I works with > > the IBM harddisk. but when I use the Flash IDE disk. always give me LI 01 > > 01 01 ... with LILO. or with Grub, too slow. > > but with the etherboot+AA . it sounds great!. :-) > > what exactly are you booting using AA patch? it seems to me that using his > patch you simply by pass either lilo or grub. I modify the makefile. that make etherboot can poll the IDE ELF kernel. > > as for lilo, it seems to me that your problem is that you install lilo > using different system than you boot from so disk geometry changes and > thus the "LI" error. yeah . one time I have let LILO work. but the speed still slow. > > as for grub. well it is weird. are you getting any errors? it seems to me > like bochs bios would need fixing. I agree with you. bochs bios of ATA need fixing. now it seem it does not suitable to the real hardware. > > btw: I assume you are using 16 bit PC BIOS modules from etherboot rather > than 32 bit EBI modules from said package, right? Interesting. Opens lots > of new possibilities. > right. I using the 16bit PC BIOS modules from etherboot. version 5.0.6. with the AA patch. > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > > From shubhangi.jadhav at patni.com Tue Apr 29 07:45:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 07:45:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: Hi, dd if=elfimage of=/dev/hdc bs=4096 seek=1 gives me an error "/dev/hdc invalid argument". But "dd if=elfimage of=/dev/hdc" works as in it shows me the in/out records. However this is overwriting the partition table. when I do a fdisk -l /dev/hdc it says "disk /dev/hdc doesn't contain a valid partition table" Any suggestions on what could be done?? Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of ron minnich Sent: Tuesday, April 29, 2003 2:00 AM To: Richard Smith Cc: Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios On Mon, 28 Apr 2003, Richard Smith wrote: > On my system if I add the seek option to dd it whines about /dev/hde > being an invalid argument as well so I don't think its the partition > number thats causing the issue. > > 'dd if=elfimage of=/dev/hde bs=4096 seek=1' I've never had that happen. Weird. ron _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Tue Apr 29 09:17:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:17:01 2003 Subject: Another way of using ADLO In-Reply-To: <032001c30fba$3eacc6b0$2a00a8c0@ben> Message-ID: On Thu, 1 May 2003, bendany wrote: > I am trying to use the VGA with LinuxBIOS. since I am trying to use the > Xwindows. if you want to start the Xserver, you must have the vgabios. I > think when the Xserver startup, it will probe the videocard. but since > the LinuxBIOS does not have the videobios. so the Xserver crash. no, the problem is that the Xserver emulation of an 8086 has bugs, and so it crashes. There are some real problems with the X11 VGA support. > you can using the Xserver with the kernel framebuffer driver. such as > fbdev. but it lack the ability of hardware accel. such as play the movie > fullscreen. so use the ADLO & bochs bios & vgabios will give you a > second way to run the Xserver. since the LinuxBIOS switch the CPU into > protected mode early, and the commercial vgabios runs in realmode. so > the commercial vgabios cann't run after the LinuxBIOS. so the ADLO will > do some preparation works. Load the bochs bios & vgabios. switch the CPU > to the realmode. execute the bochs bios. the bochs will try to init the > vgabios first.it scan the address 0xC0000 to 0xC7FFF in 2k increments. > and then scan the General ROM from 0xC8000 to 0xDFFFF in 2k increments, > so if we put the etherboot+AA patch after the address of vgabios. the > etherboot will be excuted after bochs init the vgabios. the way why I do > this is that the bochs bios is lack the ability of dealing with the real > hardware. such as ide harddisk. In my experiment, when I use the IBM 30G > harddisk. the speed of lilo & grub load the kernel normally, but when I > use the flash IDE. the speed of lilo & grub , or syslinux, is too slow. Actually, this is very helpful but not completely right. I am glad you got your vga to work however. ron From rminnich at lanl.gov Tue Apr 29 09:19:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:19:01 2003 Subject: Linuxbios + ADLO In-Reply-To: Message-ID: On Tue, 29 Apr 2003, Yap Ee Huey wrote: > payload payload > > And then I build a romimage (256k) with payload padded to (192k) > dd conv=sync bs=196608 if=payload of=payload.block > and linuxbios.rom at the end 64k . > cat payload.block linuxbios.rom > romimage this is correct. Consider enabling SERIAL_POST, since you will see POST information on the serial port. I have fixed SERIAL_POST, but was unable to commit yesterday, I will try again today. ron From rminnich at lanl.gov Tue Apr 29 09:21:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:21:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: On Tue, 29 Apr 2003, Shubhangi Jadhav wrote: > dd if=elfimage of=/dev/hdc bs=4096 seek=1 > gives me an error "/dev/hdc invalid argument". run it under strace and find the failing system call. Then let me know. ron From shubhangi.jadhav at patni.com Tue Apr 29 09:23:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 09:23:00 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: Hi, I got dd working for my CF, it was the buffer misalignment problem with ext2. Now with ext3 it works. But I still have problem with Linuxbios,it detects the CF now but still gives me the error: Cannot Load ELF Image Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Shubhangi Jadhav Sent: Tuesday, April 29, 2003 5:52 PM To: ron minnich; Richard Smith Cc: Linuxbios at clustermatic.org Subject: RE: Using ADLO with Linuxbios Hi, dd if=elfimage of=/dev/hdc bs=4096 seek=1 gives me an error "/dev/hdc invalid argument". But "dd if=elfimage of=/dev/hdc" works as in it shows me the in/out records. However this is overwriting the partition table. when I do a fdisk -l /dev/hdc it says "disk /dev/hdc doesn't contain a valid partition table" Any suggestions on what could be done?? Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of ron minnich Sent: Tuesday, April 29, 2003 2:00 AM To: Richard Smith Cc: Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios On Mon, 28 Apr 2003, Richard Smith wrote: > On my system if I add the seek option to dd it whines about /dev/hde > being an invalid argument as well so I don't think its the partition > number thats causing the issue. > > 'dd if=elfimage of=/dev/hde bs=4096 seek=1' I've never had that happen. Weird. ron _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From shubhangi.jadhav at patni.com Tue Apr 29 09:39:01 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 09:39:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: <3EAD4BD0.2030302@nexpath.com> Message-ID: >I believe you will see this error if you do not set "linux >/usr/src/linux" or where ever your linux kernel source tree is. Put >this in, and I think this error will go away. Tried it and it got the following error - make: *** No rule to make target `/usr/src/linux2.4.18-3/vmlinux', needed by `linux.bin'. Stop. Another question, if I'm directly booting an elf image from linuxbios would I need to set linux in the Config file? -Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Steve Gehlbach Sent: Monday, April 28, 2003 9:12 PM To: Shubhangi Jadhav Cc: ron minnich; Eric W. Biederman; Deepak Kotian; Adam Agnew; bendany; Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios Shubhangi Jadhav wrote: > I give the following options in my config file > > option BOOT_IDE=1 > option IDE_BOOT_DRIVE=2 > option ONE_TRACK=32 > commandline root=/dev/hda2 console=ttyS0,115200 console=tty0 > option USE_ELF_BOOT=1 > > Is this correct if I want Linuxbios to directly boot the elfimage? > However when I try to build linuxbios using these options I get an error- > file linux.bin not found. Now why does it look for this file. > Do I need to specify some payload in the config file. what payload is it? > I believe you will see this error if you do not set "linux /usr/src/linux" or where ever your linux kernel source tree is. Put this in, and I think this error will go away. But I should mention that I have never gotten this to boot with USE_ELF_BOOT=1, though I am told it should. It has only worked for me using vmlinux.gin.gz, ie, direct loading a non-elf image. -Steve _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From nathanael at gnat.ca Tue Apr 29 09:41:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 29 09:41:01 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: <10219FD6-7A4D-11D7-9304-0003931B4D6A@gnat.ca> On Tuesday, April 29, 2003, at 07:42 AM, ron minnich wrote: > On Mon, 28 Apr 2003, Nathanael Noblet wrote: > >> I ran these commands: >> ./dumpdevmem > bios >> hexdump bios | more > > This is with bios savior right? I had the same problem. I don't know > what > part it is. Bios savior? I don't think so. I have a SBC board, I haven't purchased anything special for it. I take it Bios Savior is some kind of device to make sure you don't loose the bios... > Do you have a plain old flash part you can put in there? SST 28SF040 is > ideal. I can put an empty new flash part WinBond W29C020CP9DB ( I think it often goes by W29c020) -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From nathanael at gnat.ca Tue Apr 29 09:45:01 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 29 09:45:01 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: <933A3E57-7A4D-11D7-9304-0003931B4D6A@gnat.ca> > > (1)UNIFLASH (DOS program) > http://www.pppr.sk/rainbow/programs.html Interesting... I don't have floppy access to the board so unfortunately this won't work either. > (2)/dev/bios (Linux program) > http://www.openbios.info/development/devbios.html This program segfaults when I try to load the module for the board, though it works fine on some of my other machines. -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Tue Apr 29 09:45:23 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:45:23 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: On Tue, 29 Apr 2003, Shubhangi Jadhav wrote: > >I believe you will see this error if you do not set "linux > >/usr/src/linux" or where ever your linux kernel source tree is. Put > >this in, and I think this error will go away. > > Tried it and it got the following error - > > make: *** No rule to make target `/usr/src/linux2.4.18-3/vmlinux', needed by > `linux.bin'. Stop. payload /usr/src/linux/arch/i386/boot/bzImage the 'linux' command is deprecated > Another question, if I'm directly booting an elf image from linuxbios would > I need to set linux in the Config file? payload, and yes you need to do that. ron From rminnich at lanl.gov Tue Apr 29 09:47:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:47:01 2003 Subject: Bios flashing continued In-Reply-To: <10219FD6-7A4D-11D7-9304-0003931B4D6A@gnat.ca> Message-ID: On Tue, 29 Apr 2003, Nathanael Noblet wrote: > I can put an empty new flash part WinBond W29C020CP9DB ( I think it > often goes by W29c020) that's not support either, but you can give it a try and just fix up flash_rom to make it go. ron From shubhangi.jadhav at patni.com Tue Apr 29 09:47:07 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 09:47:07 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: > But I still have problem with Linuxbios,it detects the CF now but still >gives me the error: Cannot Load ELF Image I get the following messages : **********************log************************************************** I am now initializing the ide system UQNAUT MIFERABLLcl1t 001 init_drive sectors_per_track=[63], num_heads=[16], num_cylinders=[16383] IDE0 16/16383/63 cap: 0f00 init_drive sectors_per_track=[0], num_heads=[0], num_cylinders=[0] OTHSBI AHTCN0F23BM A init_drive sectors_per_track=[32], num_heads=[4], num_cylinders=[496] IDE2 4/496/32 cap: 0200 init_drive sectors_per_track=[0], num_heads=[0], num_cylinders=[0] Initialized controller... IDE_BOOT_DRIVE = 2 .ide_read_sector drive[2], buffer[0000b820], block[00000020], offset[0], n_bytes [512] ide_read_sector block(00000020) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[20], block[ 20] .ide_read_sector drive[2], buffer[0000b820], block[00000021], offset[0], n_bytes [512] ide_read_sector block(00000021) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[21], block[ 21] .ide_read_sector drive[2], buffer[0000b820], block[00000022], offset[0], n_bytes [512] ide_read_sector block(00000022) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[22], block[ 22] .ide_read_sector drive[2], buffer[0000b820], block[00000023], offset[0], n_bytes [512] ide_read_sector block(00000023) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[23], block[ 23] .ide_read_sector drive[2], buffer[0000b820], block[00000024], offset[0], n_bytes [512] ide_read_sector block(00000024) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[24], block[ 24] .ide_read_sector drive[2], buffer[0000b820], block[00000025], offset[0], n_bytes [512] ide_read_sector block(00000025) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[25], block[ 25] .ide_read_sector drive[2], buffer[0000b820], block[00000026], offset[0], n_bytes [512] ide_read_sector block(00000026) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[26], block[ 26] .ide_read_sector drive[2], buffer[0000b820], block[00000027], offset[0], n_bytes [512] ide_read_sector block(00000027) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[27], block[ 27] .ide_read_sector drive[2], buffer[0000b820], block[00000028], offset[0], n_bytes [512] ide_read_sector block(00000028) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[28], block[ 28] .ide_read_sector drive[2], buffer[0000b820], block[00000029], offset[0], n_bytes [512] ide_read_sector block(00000029) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[29], block[ 29] .ide_read_sector drive[2], buffer[0000b820], block[0000002a], offset[0], n_bytes [512] ide_read_sector block(0000002a) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2a], block[ 2a] .ide_read_sector drive[2], buffer[0000b820], block[0000002b], offset[0], n_bytes [512] ide_read_sector block(0000002b) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2b], block[ 2b] .ide_read_sector drive[2], buffer[0000b820], block[0000002c], offset[0], n_bytes [512] ide_read_sector block(0000002c) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2c], block[ 2c] .ide_read_sector drive[2], buffer[0000b820], block[0000002d], offset[0], n_bytes [512] ide_read_sector block(0000002d) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2d], block[ 2d] .ide_read_sector drive[2], buffer[0000b820], block[0000002e], offset[0], n_bytes [512] ide_read_sector block(0000002e) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2e], block[ 2e] .ide_read_sector drive[2], buffer[0000b820], block[0000002f], offset[0], n_bytes [512] ide_read_sector block(0000002f) to addr(0000b820)ide_read_sector LBA: drivehead[e0], cylinder[0000], sector[2f], block[ 2f] header_offset is -1 Cannot Load ELF Image ***************************************************************** -Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Shubhangi Jadhav Sent: Tuesday, April 29, 2003 7:30 PM To: ron minnich; Richard Smith Cc: Linuxbios at clustermatic.org Subject: RE: Using ADLO with Linuxbios Hi, I got dd working for my CF, it was the buffer misalignment problem with ext2. Now with ext3 it works. But I still have problem with Linuxbios,it detects the CF now but still gives me the error: Cannot Load ELF Image Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Shubhangi Jadhav Sent: Tuesday, April 29, 2003 5:52 PM To: ron minnich; Richard Smith Cc: Linuxbios at clustermatic.org Subject: RE: Using ADLO with Linuxbios Hi, dd if=elfimage of=/dev/hdc bs=4096 seek=1 gives me an error "/dev/hdc invalid argument". But "dd if=elfimage of=/dev/hdc" works as in it shows me the in/out records. However this is overwriting the partition table. when I do a fdisk -l /dev/hdc it says "disk /dev/hdc doesn't contain a valid partition table" Any suggestions on what could be done?? Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of ron minnich Sent: Tuesday, April 29, 2003 2:00 AM To: Richard Smith Cc: Linuxbios at clustermatic.org Subject: Re: Using ADLO with Linuxbios On Mon, 28 Apr 2003, Richard Smith wrote: > On my system if I add the seek option to dd it whines about /dev/hde > being an invalid argument as well so I don't think its the partition > number thats causing the issue. > > 'dd if=elfimage of=/dev/hde bs=4096 seek=1' I've never had that happen. Weird. ron _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Tue Apr 29 09:48:07 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:48:07 2003 Subject: Bios flashing continued In-Reply-To: <933A3E57-7A4D-11D7-9304-0003931B4D6A@gnat.ca> Message-ID: On Tue, 29 Apr 2003, Nathanael Noblet wrote: > This program segfaults when I try to load the module for the board, > though it works fine on some of my other machines. did you see anything at all via dmesg? ron From nathanael at gnat.ca Tue Apr 29 09:57:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 29 09:57:00 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: <3DFB88E6-7A4F-11D7-9304-0003931B4D6A@gnat.ca> On Tuesday, April 29, 2003, at 08:21 AM, ron minnich wrote: > On Tue, 29 Apr 2003, Nathanael Noblet wrote: > >> This program segfaults when I try to load the module for the board, >> though it works fine on some of my other machines. > > did you see anything at all via dmesg? The thing is I think devbios supports the chip I have, but that something weird is happening (obviously) that I don't know about. Most of the code I've seen assumes that the sis chip is coupled with a superio chip that mine isn't either the sis950 or the sis 5595 or something like that, when mine is a ITE8661 (to which I have the spec sheet, but don't know what they are trying to do with the other chips so don't know what to change to make it work for mine). BIOS: Probing system firmware with 2048k rom area @0xffe00000 (1039:0008) Unable to handle kernel NULL pointer dereference at virtual address 00000051 printing eip: c484397e *pde = 00000000 Oops: 0000 CPU: 0 EIP: 0010:[] Not tainted EFLAGS: 00010202 eax: 000000c8 ebx: c4849004 ecx: 00000cf8 edx: 00000cfe esi: c4849000 edi: c10dc800 ebp: c484b000 esp: c2e8bec8 ds: 0018 es: 0018 ss: 0018 Process insmod (pid: 669, stackpage=c2e8b000) Stack: c10dc000 00000076 c484b000 c48450c9 c2e8bf10 10390008 c10dc800 00000060 c48434ba ffe00000 00200000 00000000 c48468e0 c2e8bf10 c2e8bf0c 00000008 10390000 ffe00000 00200000 c4846948 00000000 00000000 00000060 c48430f5 Call Trace: [] [] [] [] [] [] [] [] Code: a0 51 00 00 00 a2 e5 8f 84 c4 a1 c4 87 84 c4 6a 40 8d 04 40 -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From public at demiansinclair.com Tue Apr 29 09:58:00 2003 From: public at demiansinclair.com (demian sinclair) Date: Tue Apr 29 09:58:00 2003 Subject: can't get 2.4.19 to compile Message-ID: <20030429093237.6b7501b1.public@demiansinclair.com> I am trying to get linux bios to work on my m758lmr+ for my carputer. Unfortunately I am stuck at the first stage. When I try make bzImage after patching with the 2.4.19-sis patch, I get this: ld -m elf_i386 -T /usr/src/linux-2.4.19/arch/i386/vmlinux.lds -e stext arch/i386/kernel/head.o arch/i386/kernel/init_task.o init/main.o init/version.o init/do_mounts.o \ --start-group \ arch/i386/kernel/kernel.o arch/i386/mm/mm.o kernel/kernel.o mm/mm.o fs/fs.o ipc/ipc.o \ drivers/char/char.o drivers/block/block.o drivers/misc/misc.o drivers/net/net.o drivers/media/media.o drivers/ide/idedriver.o drivers/cdrom/driver.o drivers/sound/sounddrivers.o drivers/pci/driver.o drivers/video/video.o \ net/network.o \ /usr/src/linux-2.4.19/arch/i386/lib/lib.a /usr/src/linux-2.4.19/lib/lib.a /usr/src/linux-2.4.19/arch/i386/lib/lib.a \ --end-group \ -o vmlinux drivers/video/video.o(.data+0x1bd4): undefined reference to `local symbols in discarded section .text.exit' make: *** [vmlinux] Error 1 This is on debian woddy with gcc 2.95. I have also tried 2.4.17 and I get this: ld -m elf_i386 -T /usr/src/linux/arch/i386/vmlinux.lds -e stext arch/i386/kernel/head.o arch/i386/kernel/init_task.o init/main.o init/version.o \ --start-group \ arch/i386/kernel/kernel.o arch/i386/mm/mm.o kernel/kernel.o mm/mm.o fs/fs.o ipc/ipc.o \ drivers/char/char.o drivers/block/block.o drivers/misc/misc.o drivers/net/net.o drivers/media/media.o drivers/ide/idedriver.o drivers/cdrom/driver.o drivers/sound/sounddrivers.o drivers/pci/driver.o drivers/mtd/mtdlink.o drivers/video/video.o drivers/usb/usbdrv.o \ net/network.o \ /usr/src/linux/arch/i386/lib/lib.a /usr/src/linux/lib/lib.a /usr/src/linux/arch/i386/lib/lib.a \ --end-group \ -o vmlinux drivers/sound/sounddrivers.o(.data+0x1d4): undefined reference to `local symbols in discarded section .text.exit' drivers/mtd/mtdlink.o(.text.lock+0xf6): undefined reference to `local symbols in discarded section .text.exit' drivers/video/video.o(.data+0x1bd4): undefined reference to `local symbols in discarded section .text.exit' make: *** [vmlinux] Error 1 Would someone plz point me in the right direction. Thanks, DS From rminnich at lanl.gov Tue Apr 29 09:59:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 09:59:00 2003 Subject: Bios flashing continued In-Reply-To: <3DFB88E6-7A4F-11D7-9304-0003931B4D6A@gnat.ca> Message-ID: oh boy, that looks fun. Let's just try to get flash_rom working -- I don't have time to debug this one :-) ron From shubhangi.jadhav at patni.com Tue Apr 29 09:59:09 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 09:59:09 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: >payload /usr/src/linux/arch/i386/boot/bzImage Can this be the same kernel image from which I created my elf image? -Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of ron minnich Sent: Tuesday, April 29, 2003 7:48 PM To: Shubhangi Jadhav Cc: Steve Gehlbach; Eric W. Biederman; Deepak Kotian; Adam Agnew; bendany; Linuxbios at clustermatic.org Subject: RE: Using ADLO with Linuxbios On Tue, 29 Apr 2003, Shubhangi Jadhav wrote: > >I believe you will see this error if you do not set "linux > >/usr/src/linux" or where ever your linux kernel source tree is. Put > >this in, and I think this error will go away. > > Tried it and it got the following error - > > make: *** No rule to make target `/usr/src/linux2.4.18-3/vmlinux', needed by > `linux.bin'. Stop. payload /usr/src/linux/arch/i386/boot/bzImage the 'linux' command is deprecated > Another question, if I'm directly booting an elf image from linuxbios would > I need to set linux in the Config file? payload, and yes you need to do that. ron _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From rminnich at lanl.gov Tue Apr 29 10:00:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 10:00:01 2003 Subject: Using ADLO with Linuxbios In-Reply-To: Message-ID: On Tue, 29 Apr 2003, Shubhangi Jadhav wrote: > Can this be the same kernel image from which I created my elf image? stupid me. Sorry. you mkelfImage and then use that as the payload. Memory glitch on my part. ron From rminnich at lanl.gov Tue Apr 29 10:11:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 10:11:01 2003 Subject: can't get 2.4.19 to compile In-Reply-To: <20030429093237.6b7501b1.public@demiansinclair.com> Message-ID: On Tue, 29 Apr 2003, demian sinclair wrote: > I am trying to get linux bios to work on my m758lmr+ for my carputer. > Unfortunately I am stuck at the first stage. When I try make bzImage > after patching with the 2.4.19-sis patch, I get this: did the kernel make before the patch? If so, you need to reset and try each thing one at a time. This looks like some weird binutils problem. ron From dkotian3 at vsnl.net Tue Apr 29 10:22:01 2003 From: dkotian3 at vsnl.net (dkotian3 at vsnl.net) Date: Tue Apr 29 10:22:01 2003 Subject: Using ADLO with Linuxbios Message-ID: <20030429152927.63D304FE0F@bom6.vsnl.net.in> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From shubhangi.jadhav at patni.com Tue Apr 29 10:26:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Tue Apr 29 10:26:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: Message-ID: Hi, what version of etherboot are you using, I tried with etherboot5.0.9 and etherboot5.0.10 but "make ide_disk.elf" doesn't work, neither does "make bin32/ide_disk.elf" Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Yap Ee Huey Sent: Thursday, April 24, 2003 6:11 PM To: linuxbios at clustermatic.org Cc: etherboot-users at lists.sourceforge.net Subject: Linuxbios + Etherboot + tagged Kernel Image Hi hackers : I am a little bit confused here after building linuxbios, etherboot and making a tagged kernel image with mknbi. Hopefully I can get some help here : I have made ide_disk.elf, (22k) I tag bzImage with mknbi-linux, (888k) I have built linuxbios.rom (64k) I have 256k bios, so the kernel should be put somewhere on my harddisk. If i am not wrong my ide_disk.elf should be placed at the first sector of my bios rom. My question is: where should I put my Kernel Image and linuxbios.rom so that ide_disk.elf will know where to get and load them ? Do I need to modify the -DDEFAULT_BOOTFILE option which defaulted to tftp:///tftpboot/kernel ? Thanks a lot ! best regards, eehuey _________________________________________________________________ Are you in love? Find a date on MSN Personals http://match.msn.com.my/ _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From nathanael at gnat.ca Tue Apr 29 10:33:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 29 10:33:00 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: <5243C7E5-7A54-11D7-9304-0003931B4D6A@gnat.ca> On Tuesday, April 29, 2003, at 07:42 AM, ron minnich wrote: > Do you have a plain old flash part you can put in there? SST 28SF040 is > ideal. So I ordered the flash parts that were already in the machine when I bought it, the chances of another type of flash working is slim to none right? Were I to purchase the one above, how would I know if it would work? -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From bendany at mistdl.com Tue Apr 29 10:45:01 2003 From: bendany at mistdl.com (bendany) Date: Tue Apr 29 10:45:01 2003 Subject: Another way of using ADLO References: Message-ID: <001301c30edf$8040c490$376376ca@benbook> ----- Original Message ----- From: "ron minnich" To: "bendany" Cc: Sent: Tuesday, April 29, 2003 6:49 AM Subject: Re: Another way of using ADLO > On Thu, 1 May 2003, bendany wrote: > > > I am trying to use the VGA with LinuxBIOS. since I am trying to use the > > Xwindows. if you want to start the Xserver, you must have the vgabios. I > > think when the Xserver startup, it will probe the videocard. but since > > the LinuxBIOS does not have the videobios. so the Xserver crash. > > no, the problem is that the Xserver emulation of an 8086 has bugs, and so > it crashes. There are some real problems with the X11 VGA support. That you means the Xserver can run without vgabios? or we can modify the Xserver, to make it support LinuxBIOS? I am not clear with this. may be you can explain more. > > Actually, this is very helpful but not completely right. I am glad you got > your vga to work however. I have an idea. that is , since we don't have time or documents to write the vga driver. we can make LinuxBIOS to support the origianl vgabios. Just like ADLO, but somewhat simplify just only init the vgabios. and then jump into protect mode again. because the vgabios is distributed with mb. we can get it legally. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From bendany at mistdl.com Tue Apr 29 10:47:00 2003 From: bendany at mistdl.com (bendany) Date: Tue Apr 29 10:47:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image References: Message-ID: <003301c30ee0$d69b7000$376376ca@benbook> ----- Original Message ----- From: "Shubhangi Jadhav" To: "Yap Ee Huey" ; Cc: Sent: Tuesday, April 29, 2003 8:03 AM Subject: RE: Linuxbios + Etherboot + tagged Kernel Image > Hi, > > what version of etherboot are you using, I tried with etherboot5.0.9 and > etherboot5.0.10 but "make ide_disk.elf" doesn't work, neither does "make > bin32/ide_disk.elf" mostly etherboot version that we use is 5.0.6 . plus the AA patch. or if you want to get ide_disk.elf. you may try the development vesion. that is 5.1.8 now. > > Regards, > Shubhangi > > -----Original Message----- > From: linuxbios-admin at clustermatic.org > [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Yap Ee Huey > Sent: Thursday, April 24, 2003 6:11 PM > To: linuxbios at clustermatic.org > Cc: etherboot-users at lists.sourceforge.net > Subject: Linuxbios + Etherboot + tagged Kernel Image > > > Hi hackers : > > I am a little bit confused here after building linuxbios, > etherboot and making a tagged kernel image with mknbi. > Hopefully I can get some help here : > > I have made ide_disk.elf, (22k) > I tag bzImage with mknbi-linux, (888k) > I have built linuxbios.rom (64k) > > I have 256k bios, so the kernel should be put somewhere on > my harddisk. If i am not wrong my ide_disk.elf should be > placed at the first sector of my bios rom. > > My question is: where should I put my Kernel Image and > linuxbios.rom so that ide_disk.elf will know where to get > and load them ? > > Do I need to modify the -DDEFAULT_BOOTFILE option which > defaulted to tftp:///tftpboot/kernel ? > > Thanks a lot ! > > best regards, > eehuey > > _________________________________________________________________ > Are you in love? Find a date on MSN Personals http://match.msn.com.my/ > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From rminnich at lanl.gov Tue Apr 29 11:34:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 11:34:00 2003 Subject: Bios flashing continued In-Reply-To: <5243C7E5-7A54-11D7-9304-0003931B4D6A@gnat.ca> Message-ID: On Tue, 29 Apr 2003, Nathanael Noblet wrote: > So I ordered the flash parts that were already in the machine when I > bought it, the chances of another type of flash working is slim to none > right? Were I to purchase the one above, how would I know if it would > work? better yet, let's just try to fix flash_rom so it works for you. Tell me the flash part # again. ron From rminnich at lanl.gov Tue Apr 29 11:35:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 11:35:00 2003 Subject: Another way of using ADLO In-Reply-To: <001301c30edf$8040c490$376376ca@benbook> Message-ID: On Tue, 29 Apr 2003, bendany wrote: > I have an idea. that is , since we don't have time or documents to write > the vga driver. we can make LinuxBIOS to support the origianl vgabios. > Just like ADLO, but somewhat simplify just only init the vgabios. and > then jump into protect mode again. because the vgabios is distributed > with mb. we can get it legally. well, that support is in there already. ron From rminnich at lanl.gov Tue Apr 29 11:38:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 11:38:01 2003 Subject: Another way of using ADLO In-Reply-To: <001301c30edf$8040c490$376376ca@benbook> Message-ID: On Tue, 29 Apr 2003, bendany wrote: > That you means the Xserver can run without vgabios? or we can modify the > Xserver, to make it support LinuxBIOS? I am not clear with this. may be > you can explain more. it means you need to fix the x86 emulation in x11, as it has errors. Linuxbios emulator fixed the errors. > I have an idea. that is , since we don't have time or documents to write > the vga driver. we can make LinuxBIOS to support the origianl vgabios. > Just like ADLO, but somewhat simplify just only init the vgabios. and > then jump into protect mode again. because the vgabios is distributed > with mb. we can get it legally. Already done. ron From nathanael at gnat.ca Tue Apr 29 11:44:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Tue Apr 29 11:44:00 2003 Subject: Bios flashing continued In-Reply-To: Message-ID: <374F650A-7A5E-11D7-82A7-0003931B4D6A@gnat.ca> On Tuesday, April 29, 2003, at 10:07 AM, ron minnich wrote: > On Tue, 29 Apr 2003, Nathanael Noblet wrote: > >> So I ordered the flash parts that were already in the machine when I >> bought it, the chances of another type of flash working is slim to >> none >> right? Were I to purchase the one above, how would I know if it would >> work? > > better yet, let's just try to fix flash_rom so it works for you. > > Tell me the flash part # again. w29c020c (p90b) spec sheet I just found is available at http://www.winbond.com/e-winbondhtm/partner/b_2_h_1_1.htm the p90b is for plcc 90ns and b I can't remember... -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From aip at cwlinux.com Tue Apr 29 12:36:01 2003 From: aip at cwlinux.com (Andrew Ip) Date: Tue Apr 29 12:36:01 2003 Subject: Bios flashing continued In-Reply-To: <374F650A-7A5E-11D7-82A7-0003931B4D6A@gnat.ca>; from Nathanael Noblet on Tue, Apr 29, 2003 at 10:18:30AM -0600 References: <374F650A-7A5E-11D7-82A7-0003931B4D6A@gnat.ca> Message-ID: <20030430011056.A5655@mail.cwlinux.com> > > Tell me the flash part # again. > w29c020c (p90b) It is already supported. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From pyro at linuxlabs.com Tue Apr 29 12:57:00 2003 From: pyro at linuxlabs.com (steven james) Date: Tue Apr 29 12:57:00 2003 Subject: can't get 2.4.19 to compile In-Reply-To: <20030429093237.6b7501b1.public@demiansinclair.com> Message-ID: Greetings, I'm not certain, but this is the sort of thing that can come up when you don't make mrproper after patching. copy your .config somewhere safe since mrproper will wipe it out, then make mrproper now copy .config back into kernel tree and make oldconfig build kernel as usual. G'day, sjames On Tue, 29 Apr 2003, demian sinclair wrote: > I am trying to get linux bios to work on my m758lmr+ for my carputer. > Unfortunately I am stuck at the first stage. When I try make bzImage > after patching with the 2.4.19-sis patch, I get this: > > ld -m elf_i386 -T /usr/src/linux-2.4.19/arch/i386/vmlinux.lds -e stext > arch/i386/kernel/head.o arch/i386/kernel/init_task.o init/main.o > init/version.o init/do_mounts.o \ --start-group \ > arch/i386/kernel/kernel.o arch/i386/mm/mm.o kernel/kernel.o > mm/mm.o fs/fs.o ipc/ipc.o \ drivers/char/char.o > drivers/block/block.o drivers/misc/misc.o drivers/net/net.o > drivers/media/media.o drivers/ide/idedriver.o drivers/cdrom/driver.o > drivers/sound/sounddrivers.o drivers/pci/driver.o drivers/video/video.o > \ net/network.o \ > /usr/src/linux-2.4.19/arch/i386/lib/lib.a > /usr/src/linux-2.4.19/lib/lib.a > /usr/src/linux-2.4.19/arch/i386/lib/lib.a \ --end-group \ > -o vmlinux > drivers/video/video.o(.data+0x1bd4): undefined reference to `local > symbols in discarded section .text.exit' make: *** [vmlinux] Error 1 > > This is on debian woddy with gcc 2.95. I have also tried 2.4.17 and I > get this: > > ld -m elf_i386 -T /usr/src/linux/arch/i386/vmlinux.lds -e stext > arch/i386/kernel/head.o arch/i386/kernel/init_task.o init/main.o > init/version.o \ --start-group \ > arch/i386/kernel/kernel.o arch/i386/mm/mm.o kernel/kernel.o > mm/mm.o fs/fs.o ipc/ipc.o \ drivers/char/char.o > drivers/block/block.o drivers/misc/misc.o drivers/net/net.o > drivers/media/media.o drivers/ide/idedriver.o drivers/cdrom/driver.o > drivers/sound/sounddrivers.o drivers/pci/driver.o drivers/mtd/mtdlink.o > drivers/video/video.o drivers/usb/usbdrv.o \ net/network.o \ > /usr/src/linux/arch/i386/lib/lib.a /usr/src/linux/lib/lib.a > /usr/src/linux/arch/i386/lib/lib.a \ --end-group \ > -o vmlinux > drivers/sound/sounddrivers.o(.data+0x1d4): undefined reference to `local > symbols in discarded section .text.exit' > drivers/mtd/mtdlink.o(.text.lock+0xf6): undefined reference to `local > symbols in discarded section .text.exit' > drivers/video/video.o(.data+0x1bd4): undefined reference to `local > symbols in discarded section .text.exit' make: *** [vmlinux] Error 1 > > > Would someone plz point me in the right direction. > > Thanks, > DS > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > -- -------------------------steven james, director of research, linux labs ... ........ ..... .... 230 peachtree st nw ste 2701 the original linux labs atlanta.ga.us 30303 -since 1995 http://www.linuxlabs.com office 404.577.7747 fax 404.577.7743 ----------------------------------------------------------------------- From dkotian3 at vsnl.net Tue Apr 29 14:26:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 29 14:26:01 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT Message-ID: <014a01c30e81$8d930b20$303e41db@vsnl.net> Hi, My enviorment : I have a system, which has 256/512K BIOS FLASH ROMS with 32MB Compact flash IDE. And the kernel image is residing on Compact flash. I can change the partioning as per the needs for LinuxBIOS to work. I just want BOOT Linux and no other O.S for now. I wish to burn the LinuxBIOS on 256K/512K BIOS FLASHROMS. Question: 1. What should I use, ADLO,etherboot,LinuxBIOS itself supports? 2. If LinuxBIOS supports, could some give a link to a sample conifg file for the same. Please elaborate, if possible. 3. If ADLO or etherboot is to be used, can a sample config file be refered for the same.. Could someone please give some pointer on this one. Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Tue Apr 29 14:34:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 14:34:00 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT In-Reply-To: <014a01c30e81$8d930b20$303e41db@vsnl.net> Message-ID: On Wed, 30 Apr 2003, Deepak Kotian wrote: > I have a system, which has 256/512K BIOS FLASH ROMS with 32MB Compact > flash IDE. And the kernel image is residing on Compact flash. I can > change the partioning as per the needs for LinuxBIOS to work. I just > want BOOT Linux and no other O.S for now. What we do here: burn linuxbios on flash with an etherboot payload burn the linux kernel onto the Compact Flash. ron From dkotian3 at vsnl.net Tue Apr 29 14:39:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 29 14:39:00 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT References: Message-ID: <019301c30e83$5a190680$303e41db@vsnl.net> Thanks. Can I please get a sample config as well. Or the config file settings in this situation. Thanks and Regards Deepak ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: Sent: Wednesday, April 30, 2003 12:37 AM Subject: Re: BIOS FLASHROMS AND COMPACT FLASH BOOT > On Wed, 30 Apr 2003, Deepak Kotian wrote: > > > I have a system, which has 256/512K BIOS FLASH ROMS with 32MB Compact > > flash IDE. And the kernel image is residing on Compact flash. I can > > change the partioning as per the needs for LinuxBIOS to work. I just > > want BOOT Linux and no other O.S for now. > > What we do here: > > burn linuxbios on flash with an etherboot payload > > burn the linux kernel onto the Compact Flash. > > ron > From nathanael at gnat.ca Tue Apr 29 14:49:01 2003 From: nathanael at gnat.ca (Nathanael D. Noblet) Date: Tue Apr 29 14:49:01 2003 Subject: Bios flashing continued In-Reply-To: <20030430011056.A5655@mail.cwlinux.com> References: <374F650A-7A5E-11D7-82A7-0003931B4D6A@gnat.ca> <20030430011056.A5655@mail.cwlinux.com> Message-ID: <1051644213.26392.2.camel@platinum.gnat.ca> On Tue, 2003-04-29 at 11:10, Andrew Ip wrote: > > > Tell me the flash part # again. > > w29c020c (p90b) > It is already supported. > > -Andrew ok so it may be supported but I get nothing when I run ./flash_rom with no arguments. So if it is supported and I'm not even to the point of trying to write to the flash. How come it is not detected? What can I start doing to fix it? Where do I look, what do I look at? etc... Here's the output of ./flash_rom Try 2097152 timeusec is 44995 Try 20971520 timeusec is 449132 Try 209715200 timeusec is 4491720 one us is 466 count Trying Am29F040B, 512 KB probe_29f040bid1 37, id2 241 Trying At29C040A, 512 KB probe_jedecid1 37, id2 241 Trying Mx29f002, 256 KB probe_29f002id1 37, id2 241 Trying SST29EE020A, 256 KB probe_jedecid1 37, id2 241 Trying SST28SF040A, 512 KB probe_28sf040id1 37, id2 241 Trying SST39SF020A, 256 KB probe_39sf020id1 37, id2 241 Trying SST39VF020, 256 KB probe_39sf020id1 37, id2 241 Trying W29C020C, 256 KB probe_jedecid1 37, id2 241 Trying W49F002U, 256 KB probe_49f002id1 37, id2 241 Trying M29F400BT, 512 KB probe_m29f400bt id1 25, id2 2d Trying 82802ab, 512 KB probe_82802abid1 37, id2 241 EEPROM not found -- Nathanael D. Noblet Gnat Solutions From dkotian3 at vsnl.net Tue Apr 29 14:51:01 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Tue Apr 29 14:51:01 2003 Subject: Question on Storage Area Networks(SAN) with LINUX, what is the BEST forrum Message-ID: <01a201c30e85$11df6c40$303e41db@vsnl.net> Hi , Sorry, off topic/may if off forum.. But any help would be appreciated. Where would be best place to put queries for SAN Linux HOSTS. As LinuxBIOS is close with clusters, I am just guessing some people on this list may have good idea about this. Thanks and Regards Deepak -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Tue Apr 29 15:03:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 15:03:00 2003 Subject: Bios flashing continued In-Reply-To: <1051644213.26392.2.camel@platinum.gnat.ca> Message-ID: On 29 Apr 2003, Nathanael D. Noblet wrote: look in flash.h. id1 is the vendor, and i2 is the part. 37 is 25 hex, which is an unknown vendor to me. I think we may still be at 'no write enable to flash :-(' ron From spirit at reactor.ru Tue Apr 29 15:29:01 2003 From: spirit at reactor.ru (Alexander Amelkin) Date: Tue Apr 29 15:29:01 2003 Subject: Bios flashing continued In-Reply-To: <933A3E57-7A4D-11D7-9304-0003931B4D6A@gnat.ca> References: <933A3E57-7A4D-11D7-9304-0003931B4D6A@gnat.ca> Message-ID: <184198419462.20030430000237@reactor.ru> Hello Nathanael, Tuesday, April 29, 2003, 6:19:23 PM, you wrote: >> >> (1)UNIFLASH (DOS program) >> http://www.pppr.sk/rainbow/programs.html NN> Interesting... I don't have floppy access to the board so unfortunately NN> this won't work either. Can you make a DOS bootable ide disk on another machine, put uniflash and the rom image on the disk and then boot from it on your target board? So far I found that uniflash is the only working thing (which means 'no dancing around with a voodoo tambourine') out there. flash_rom never worked for me, not even on the sis630 board for which it was originally designed. With best regards, Alexander mailto:spirit at reactor.ru --8<--just-a-cookie---------------------------------------------- Real Programmers love Windows. Whenever you make a mistake you can blame Microsoft for it. --8<------------------------------------------------------------- From nathanael at gnat.ca Tue Apr 29 15:44:00 2003 From: nathanael at gnat.ca (Nathanael D. Noblet) Date: Tue Apr 29 15:44:00 2003 Subject: Bios flashing continued In-Reply-To: References: Message-ID: <1051647544.26923.7.camel@platinum.gnat.ca> On Tue, 2003-04-29 at 13:36, ron minnich wrote: > On 29 Apr 2003, Nathanael D. Noblet wrote: > > look in flash.h. id1 is the vendor, and i2 is the part. > 37 is 25 hex, which is an unknown vendor to me. I removed the current flash chip that was in there and replaced it with one of the ones I bought. Ran ./flash_rom again and got... Try 2097152 timeusec is 44994 Try 20971520 timeusec is 449133 Try 209715200 timeusec is 4491595 one us is 466 count Trying Am29F040B, 512 KB probe_29f040bid1 255, id2 255 Trying At29C040A, 512 KB probe_jedecid1 255, id2 255 Trying Mx29f002, 256 KB probe_29f002id1 255, id2 255 Trying SST29EE020A, 256 KB probe_jedecid1 255, id2 255 Trying SST28SF040A, 512 KB probe_28sf040id1 255, id2 255 Trying SST39SF020A, 256 KB probe_39sf020id1 255, id2 255 Trying SST39VF020, 256 KB probe_39sf020id1 255, id2 255 Trying W29C020C, 256 KB probe_jedecid1 255, id2 255 Trying W49F002U, 256 KB probe_49f002id1 255, id2 255 Trying M29F400BT, 512 KB probe_m29f400bt id1 ff, id2 ff Trying 82802ab, 512 KB probe_82802abid1 255, id2 255 EEPROM not found > I think we may still be at 'no write enable to flash :-(' so to be able to detect the flash part I need to enable flash write? I have the data sheet and found the sequence of commands to enable flash write to it. but that still doesn't explain the wrong return for the id1 & id2... I'll look into it a little more. I think with the bit of reading I've found in the flash parts spec sheets I may be able to make a little more sense of the code found in flash_and_burn... hopefully ;) -- Nathanael D. Noblet Gnat Solutions From rodmur at maybe.org Tue Apr 29 15:51:01 2003 From: rodmur at maybe.org (Dale Harris) Date: Tue Apr 29 15:51:01 2003 Subject: Socket 370 boards... Message-ID: <20030429202545.GP17267@maybe.org> Hi, I'm kind of looking at a situation where I'm probably going to purchase some dual CPU Socket 370 boards, cause I have bunch of Pentium III - 1Ghz chips laying around. What is the best board at this point to purchase, that is supported by LinuxBIOS? Need to be able to support at least 2Gb RAM. -- Dale Harris rodmur at maybe.org /.-) From rminnich at lanl.gov Tue Apr 29 15:56:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Tue Apr 29 15:56:01 2003 Subject: Bios flashing continued In-Reply-To: <1051647544.26923.7.camel@platinum.gnat.ca> Message-ID: On 29 Apr 2003, Nathanael D. Noblet wrote: > so to be able to detect the flash part I need to enable flash write? I > have the data sheet and found the sequence of commands to enable flash > write to it. but that still doesn't explain the wrong return for the id1 > & id2... flash write is not enabled, most likely. Do you have a scope handy? ron From ollie at sis.com.tw Tue Apr 29 21:26:01 2003 From: ollie at sis.com.tw (ollie lho) Date: Tue Apr 29 21:26:01 2003 Subject: Another way of using ADLO In-Reply-To: References: Message-ID: <1051667415.1635.4.camel@ollie> On Tue, 2003-04-29 at 21:49, ron minnich wrote: > On Thu, 1 May 2003, bendany wrote: > > > I am trying to use the VGA with LinuxBIOS. since I am trying to use the > > Xwindows. if you want to start the Xserver, you must have the vgabios. I > > think when the Xserver startup, it will probe the videocard. but since > > the LinuxBIOS does not have the videobios. so the Xserver crash. > > no, the problem is that the Xserver emulation of an 8086 has bugs, and so > it crashes. There are some real problems with the X11 VGA support. > Well, for some version of Xserver for SiS chipsets, it requires register table embedded in VGABIOS image. So you need the image in the C segment anyway. -- ollie lho From ollie at sis.com.tw Tue Apr 29 21:41:01 2003 From: ollie at sis.com.tw (ollie lho) Date: Tue Apr 29 21:41:01 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT In-Reply-To: <019301c30e83$5a190680$303e41db@vsnl.net> References: <019301c30e83$5a190680$303e41db@vsnl.net> Message-ID: <1051667624.1635.6.camel@ollie> On Wed, 2003-04-30 at 03:13, Deepak Kotian wrote: > Thanks. > Can I please get a sample config as well. Or the config file settings > in this situation. > See the winfast-elf-etherboot.config -- ollie lho From yapeehuey at hotmail.com Tue Apr 29 22:17:00 2003 From: yapeehuey at hotmail.com (Yap Ee Huey) Date: Tue Apr 29 22:17:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image Message-ID: Hi, I am using etherboot 5.1.7 , any later versions will have it. regards, eehuey >Hi, > > what version of etherboot are you using, I tried with etherboot5.0.9 and >etherboot5.0.10 but "make ide_disk.elf" doesn't work, neither does "make >bin32/ide_disk.elf" > >Regards, >Shubhangi _________________________________________________________________ Download the latest MSN Messenger http://messenger.msn.com.my From bendany at mistdl.com Wed Apr 30 02:01:01 2003 From: bendany at mistdl.com (bendany) Date: Wed Apr 30 02:01:01 2003 Subject: Another way of using ADLO References: <1051667415.1635.4.camel@ollie> Message-ID: <001301c31075$8c09dae0$2a00a8c0@ben> ----- Original Message ----- From: "ollie lho" To: "Ronald G Minnich" Cc: "bendany" ; "LinuxBIOS Mailing List" Sent: Wednesday, April 30, 2003 9:50 AM Subject: Re: Another way of using ADLO > On Tue, 2003-04-29 at 21:49, ron minnich wrote: > > On Thu, 1 May 2003, bendany wrote: > > > > > I am trying to use the VGA with LinuxBIOS. since I am trying to use the > > > Xwindows. if you want to start the Xserver, you must have the vgabios. I > > > think when the Xserver startup, it will probe the videocard. but since > > > the LinuxBIOS does not have the videobios. so the Xserver crash. > > > > no, the problem is that the Xserver emulation of an 8086 has bugs, and so > > it crashes. There are some real problems with the X11 VGA support. > > > > Well, for some version of Xserver for SiS chipsets, it requires register > table embedded in VGABIOS image. So you need the image in the C segment > anyway. > Today I try the K7SEM mb. And doesn't work with the Correct VGA display. But the Xserver works. maybe other mbs can do this. > -- > ollie lho > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From yenneo at oreka.com Wed Apr 30 03:48:01 2003 From: yenneo at oreka.com (Philippe CABANNES) Date: Wed Apr 30 03:48:01 2003 Subject: EEPROM not found In-Reply-To: Message-ID: <200304300822.KAA87643@mailhub5.isdnet.net> I think the flash write enabled, I use this : setpci -s 0:11.0 40.b=54 I've read that this should enable writting, but i'm not sure. Here is my flash_rom output: Calibrating timer since microsleep sucks ... takes a second Setting up microsecond timing loop 184M loops per second Ok, calibrated, now do the deed Trying Am29F040B, 512 KB probe_29f040b: id1 0xff, id2 0xff Trying At29C040A, 512 KB probe_jedec: id1 0xff, id2 0xff Trying Mx29f002, 256 KB probe_29f002: id1 255, id2 255 Trying SST29EE020A, 256 KB probe_jedec: id1 0xff, id2 0xff Trying SST28SF040A, 512 KB probe_28sf040: id1 0xff, id2 0xff Trying SST39SF020A, 256 KB probe_39sf020: id1 0xff, id2 0xff Trying SST39VF020, 256 KB probe_39sf020: id1 0xff, id2 0xff Trying W29C011, 128 KB probe_jedec: id1 0xff, id2 0xff Trying W29C020C, 256 KB probe_jedec: id1 0xff, id2 0xff Trying W49F002U, 256 KB probe_49f002: id1 0xff, id2 0xff Trying M29F400BT, 512 KB probe_m29f400bt: id1 0xff, id2 0xff Trying 82802ab, 512 KB probe_82802ab: id1 0xff, id2 0xff Trying 82802ac, 1024 KB probe_82802ab: id1 0xff, id2 0xff EEPROM not found Mine is a 39SF020A (it's written on the chip) thanks. -------------------------------------------------- Oreka ! Nous sommes l'internet moins cher ! Surfez 25% moins cher avec http://www.oreka.com From ts1 at cma.co.jp Wed Apr 30 04:48:00 2003 From: ts1 at cma.co.jp (SONE Takeshi) Date: Wed Apr 30 04:48:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> <3EAD866C.3020508@nexpath.com> Message-ID: <20030430092248.GA16057@cma.co.jp> On Mon, Apr 28, 2003 at 01:55:16PM -0600, Eric W. Biederman wrote: > Steve Gehlbach writes: > > > Eric W. Biederman wrote: > > > > > ``Write Protect'' in the mtrrs does not mean write protected. It is a strange > > > > > messed up form of write-through. In particular it dumps the cache on writes > > > it does not forbid writes. > > > Unless you know of a situation where write protect is more appropriate please > > > use write-through. It is less confusing, and since no one is writing to that > > > area anyway it gives the exact same result, reads are cached. > > > Eric So why did you use WP cache for "XIP" area? Using same type for both area makes sense. -- Takeshi From shubhangi.jadhav at patni.com Wed Apr 30 05:13:00 2003 From: shubhangi.jadhav at patni.com (Shubhangi Jadhav) Date: Wed Apr 30 05:13:00 2003 Subject: Linuxbios + Etherboot + tagged Kernel Image In-Reply-To: <003301c30ee0$d69b7000$376376ca@benbook> Message-ID: Thanks, for the information. I got Etherboot 5.1.8 to build now. Do I need to change anything in the etherboot Config file before compilation, as it was required with the etherboot 5.0.x versions. Regards, Shubhangi -----Original Message----- From: linuxbios-admin at clustermatic.org [mailto:linuxbios-admin at clustermatic.org]On Behalf Of bendany Sent: Wednesday, April 30, 2003 11:52 AM To: Shubhangi Jadhav Cc: linuxbios at clustermatic.org Subject: Re: Linuxbios + Etherboot + tagged Kernel Image ----- Original Message ----- From: "Shubhangi Jadhav" To: "Yap Ee Huey" ; Cc: Sent: Tuesday, April 29, 2003 8:03 AM Subject: RE: Linuxbios + Etherboot + tagged Kernel Image > Hi, > > what version of etherboot are you using, I tried with etherboot5.0.9 and > etherboot5.0.10 but "make ide_disk.elf" doesn't work, neither does "make > bin32/ide_disk.elf" mostly etherboot version that we use is 5.0.6 . plus the AA patch. or if you want to get ide_disk.elf. you may try the development vesion. that is 5.1.8 now. > > Regards, > Shubhangi > > -----Original Message----- > From: linuxbios-admin at clustermatic.org > [mailto:linuxbios-admin at clustermatic.org]On Behalf Of Yap Ee Huey > Sent: Thursday, April 24, 2003 6:11 PM > To: linuxbios at clustermatic.org > Cc: etherboot-users at lists.sourceforge.net > Subject: Linuxbios + Etherboot + tagged Kernel Image > > > Hi hackers : > > I am a little bit confused here after building linuxbios, > etherboot and making a tagged kernel image with mknbi. > Hopefully I can get some help here : > > I have made ide_disk.elf, (22k) > I tag bzImage with mknbi-linux, (888k) > I have built linuxbios.rom (64k) > > I have 256k bios, so the kernel should be put somewhere on > my harddisk. If i am not wrong my ide_disk.elf should be > placed at the first sector of my bios rom. > > My question is: where should I put my Kernel Image and > linuxbios.rom so that ide_disk.elf will know where to get > and load them ? > > Do I need to modify the -DDEFAULT_BOOTFILE option which > defaulted to tftp:///tftpboot/kernel ? > > Thanks a lot ! > > best regards, > eehuey > > _________________________________________________________________ > Are you in love? Find a date on MSN Personals http://match.msn.com.my/ > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > _______________________________________________ Linuxbios mailing list Linuxbios at clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios From aip at cwlinux.com Wed Apr 30 06:01:00 2003 From: aip at cwlinux.com (Andrew Ip) Date: Wed Apr 30 06:01:00 2003 Subject: EEPROM not found In-Reply-To: <200304300822.KAA87643@mailhub5.isdnet.net>; from Philippe CABANNES on Wed, Apr 30, 2003 at 10:13:13AM +0100 References: <200304300822.KAA87643@mailhub5.isdnet.net> Message-ID: <20030430183557.A17331@mail.cwlinux.com> Philippe, > I think the flash write enabled, I use this : > setpci -s 0:11.0 40.b=54 > I've read that this should enable writting, but i'm not sure. > Here is my flash_rom output: It could be something else. I haven't figured out how to enable flash write so far. I'm using flash programmer or dos flash program at the moment. -Andrew -- Andrew Ip Email: aip at cwlinux.com Tel: (852) 2542 2046 Fax: (852) 2542 2036 Mobile: (852) 9201 9866 Cwlinux Limited Unit 202B 2/F Lai Cheong Factory Building, 479-479A Castle Peak Road, Lai Chi Kok, Kowloon, Hong Kong. Tel: (852)2542 2046 Fax: (852)2542 2036 For public pgp key, please obtain it from http://www.keyserver.net/en. From ebiederman at lnxi.com Wed Apr 30 06:28:00 2003 From: ebiederman at lnxi.com (Eric W. Biederman) Date: Wed Apr 30 06:28:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <20030430092248.GA16057@cma.co.jp> References: <3EA80024.1080401@nexpath.com> <005a01c30a8d$a2f15fa0$373e41db@vsnl.net> <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> <3EAD866C.3020508@nexpath.com> <20030430092248.GA16057@cma.co.jp> Message-ID: SONE Takeshi writes: > On Mon, Apr 28, 2003 at 01:55:16PM -0600, Eric W. Biederman wrote: > > Steve Gehlbach writes: > > > > > Eric W. Biederman wrote: > > > > > > > ``Write Protect'' in the mtrrs does not mean write protected. It is a > strange > > > > > > > > messed up form of write-through. In particular it dumps the cache on > writes > > > > > it does not forbid writes. > > > > Unless you know of a situation where write protect is more appropriate > please > > > > > use write-through. It is less confusing, and since no one is writing to > that > > > > > area anyway it gives the exact same result, reads are cached. > > > > Eric > > > So why did you use WP cache for "XIP" area? > Using same type for both area makes sense. Yep. Good catch. It looks like the XIP area needs to be fixed as well. Actually now that I think about it, is there any reason to not use XIP infrastructure for running the rom at 0xf000 as well as at higher addresses? Sure it's a waste but we have mtrr registers to burn at that point. Eric From ts1 at cma.co.jp Wed Apr 30 07:25:01 2003 From: ts1 at cma.co.jp (SONE Takeshi) Date: Wed Apr 30 07:25:01 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: References: <3EA8421F.3040805@nexpath.com> <20030425091117.GA5673@cma.co.jp> <3EA9770D.4040507@nexpath.com> <20030428183028.GA11030@tsn.or.jp> <3EAD866C.3020508@nexpath.com> <20030430092248.GA16057@cma.co.jp> Message-ID: <20030430115916.GA19257@cma.co.jp> On Wed, Apr 30, 2003 at 05:07:10AM -0600, Eric W. Biederman wrote: > Actually now that I think about it, is there any reason to > not use XIP infrastructure for running the rom at 0xf000 > as well as at higher addresses? Sure it's a waste but > we have mtrr registers to burn at that point. Do you mean the configuration like this? option XIP_ROM_SIZE=65536 option XIP_ROM_BASE=0xf0000 Actually, I've tried this at first, and it was not successful. I guess it is because fixed MTRR overrides the variable MTRR setting (so states intel manual). Totally disabling fixed MTRR (do everything in variable MTRR) looks cleaner solution, but wasteful. -- Takeshi From NEWBELL7 at magicn.com Wed Apr 30 09:06:01 2003 From: NEWBELL7 at magicn.com (NEWBELL7 at magicn.com) Date: Wed Apr 30 09:06:01 2003 Subject: linuxbios with ADLO [EPIA] Message-ID: <8474d01c30f1e$1c28b050$82cda8c0@magicn.net> Hello! Long time no see. I am trying to use framebuffer using linuxbios. So I am about to use ADLO. linuxbios -> etherboot -> ADLO -> Bochs -> VGA BIOS -> LiLo -> Linux ( flash rom ) ( remote server ) (MBR) Is that sequence right ? And I adopt the code which Andrew lp wrote on mailing list to enabel VGA device. But etherboot didn't work. ( Before enabling VGA device , it worked. ) .. The debug message is below. why didn't etherboot worked ? I think that the difference between enabling VGA and disabling VGA is PCI bus scanning and resource allocation part. Who does know this problem ? Thanks for your attention and your help. ======================================================================== ===== 0 LinuxBIOS-1.0.0 Wed Apr 30 15:46:33 KST 2003 starting... Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.0.0 Wed Apr 30 15:46:33 KST 2003 booting... Finding PCI configuration type. PCI: Using configuration type 1 Scanning PCI bus...PCI: pci_scan_bus for bus 0 PCI: 00:00.0 [1106/0601] PCI: 00:01.0 [1106/8601] PCI: 00:11.0 [1106/8231] PCI: 00:11.1 [1106/0571] PCI: 00:11.2 [1106/3038] PCI: 00:11.3 [1106/3038] PCI: 00:11.4 [1106/8235] PCI: 00:11.5 [1106/3058] PCI: 00:11.6 [1106/3068] PCI: 00:12.0 [1106/3065] PCI: pci_scan_bus for bus 1 PCI: 01:00.0 [1023/8500] PCI: pci_scan_bus returning with max=01 PCI: pci_scan_bus returning with max=01 done Allocating PCI resources... PCI: 00:00.0 register 10(00000008), read-only ignoring it PCI: 00:00.0 register 10(00000008), read-only ignoring it PCI: 00:00.0 register 10(00000008), read-only ignoring it PCI: 00:00.0 register 10(00000008), read-only ignoring it ASSIGN RESOURCES, bus 0 PCI: 00:01.0 1c <- [0x00001000 - 0x00000fff] bus 1 io PCI: 00:01.0 24 <- [0xfe900000 - 0xfe8fffff] bus 1 prefmem PCI: 00:01.0 20 <- [0xfd800000 - 0xfe8fffff] bus 1 mem ASSIGN RESOURCES, bus 1 PCI: 01:00.0 10 <- [0xfd800000 - 0xfdffffff] mem PCI: 01:00.0 14 <- [0xfe800000 - 0xfe81ffff] mem PCI: 01:00.0 18 <- [0xfe000000 - 0xfe7fffff] mem ASSIGNED RESOURCES, bus 1 PCI: 00:11.1 20 <- [0x00001c40 - 0x00001c4f] io PCI: 00:11.2 20 <- [0x00001c00 - 0x00001c1f] io PCI: 00:11.3 20 <- [0x00001c20 - 0x00001c3f] io PCI: 00:11.5 10 <- [0x00001000 - 0x000010ff] io PCI: 00:11.5 14 <- [0x00001c50 - 0x00001c53] io PCI: 00:11.5 18 <- [0x00001c60 - 0x00001c63] io PCI: 00:11.6 10 <- [0x00001400 - 0x000014ff] io PCI: 00:12.0 10 <- [0x00001800 - 0x000018ff] io PCI: 00:12.0 14 <- [0xfe900000 - 0xfe9000ff] mem ASSIGNED RESOURCES, bus 0 Allocating VGA resource done. Enabling PCI resourcess...PCI: 00:00.0 cmd <- 06 PCI: 00:01.0 cmd <- 07 PCI: 00:11.0 cmd <- 87 PCI: 00:11.1 cmd <- 81 PCI: 00:11.2 cmd <- 01 PCI: 00:11.3 cmd <- 01 PCI: 00:11.4 cmd <- 00 PCI: 00:11.5 cmd <- 01 PCI: 00:11.6 cmd <- 01 PCI: 00:12.0 cmd <- 83 PCI: 01:00.0 cmd <- 03 done. Initializing PCI devices... PCI devices initialized sizeram: returning 0x3e000 KB sizeram: NOT returning 0x3e000 KB sizeram: there are still some SPD problems ... sizeram: SO we return only 0x10000 KB totalram: 64M Initializing CPU #0 Enabling cache... Setting fixed MTRRs(0-88) type: UC Setting fixed MTRRs(0-16) type: WB DONE fixed MTRRs Setting variable MTRR 0, base: 0MB, range: 64MB, type WB DONE variable MTRRs Clear out the extra MTRR's call intel_enable_fixed_mtrr() call intel_enable_var_mtrr() Leave setup_mtrrs done. Max cpuid index : 1 Vendor ID : CentaurHauls Processor Type : 0x00 Processor Family : 0x06 Processor Model : 0x07 Processor Mask : 0x00 Processor Stepping : 0x08 Feature flags : 0x00803035 MTRR check Fixed MTRRs : Enabled Variable MTRRs: Enabled Disabling local apic...done. CPU #0 Initialized Mainboard fixup IDE enable in reg. 50 is 0x7 set IDE reg. 50 to 0x7 IRQs in reg. 4c are 0x4 setting reg. 4c to 0x4 enables in reg 0x40 0x8 enables in reg 0x40 read back as 0xb enables in reg 0x9 0x8f enables in reg 0x9 read back as 0x8a command in reg 0x4 0x81 command in reg 0x4 reads back as 0x7 Ethernet fixup Configuring VIA LAN RTC Init Invalid CMOS LB checksum Final mainboard fixup Southbridge fixup setting southbridge Assigning IRQ 12 to 0:11.2 Readback = 12 Assigning IRQ 12 to 0:11.3 Readback = 12 Assigning IRQ 10 to 0:11.5 Readback = 10 Assigning IRQ 10 to 0:11.6 Readback = 10 setting ethernet Assigning IRQ 11 to 0:12.0 Readback = 11 setting pci slot Checking IRQ routing tables... /home/newbell/LinuxBios/freebios/src/arch/i386/lib/pirq_routing.c: 24:check_0 done. Copying IRQ routing tables to 0xf0000...done. Verifing priq routing tables copy at 0xf0000...failed Wrote linuxbios table at: 00000500 - 00000664 checksum 2a26 Welcome to elfboot, the open sourced starter. January 2002, Eric Biederman. Version 1.2 203:init_bytes() - zkernel_start:0xfff00000 zkernel_mask:0x0000ffff Searching for 16 byte tags 64:rom_read_bytes() - overflowed source buffer. max_block = 15 init_bytes found 0 tags Found ELF candiate at offset 0 New segment addr 0x94000 size 0x74e8 offset 0x60 filesize 0x396c (cleaned up) New segment addr 0x94000 size 0x74e8 offset 0x60 filesize 0x396c Loading Segment: addr: 0x0000000000094000 memsz: 0x00000000000074e8 filesz: 0x0c Clearing Segment: addr: 0x000000000009796c memsz: 0x0000000000003b7c Jumping to boot code at 0x94000 -------------- next part -------------- An HTML attachment was scrubbed... URL: From rminnich at lanl.gov Wed Apr 30 09:46:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 09:46:00 2003 Subject: EEPROM not found In-Reply-To: <200304300822.KAA87643@mailhub5.isdnet.net> Message-ID: On Wed, 30 Apr 2003, Philippe CABANNES wrote: > I think the flash write enabled, I use this : > > setpci -s 0:11.0 40.b=54 send me an lspci. Are you sure this is the right chip? The bad news is there may be extra hardware on the mobo which makes this step only one of the steps you have to do. ron From rminnich at lanl.gov Wed Apr 30 09:53:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 09:53:00 2003 Subject: Delay in copying Linuxbios to ram In-Reply-To: <20030430115916.GA19257@cma.co.jp> Message-ID: On Wed, 30 Apr 2003, SONE Takeshi wrote: > Totally disabling fixed MTRR (do everything in variable MTRR) > looks cleaner solution, but wasteful. I thought we tried this at one point but there were machines that it had trouble on. ron From nathanael at gnat.ca Wed Apr 30 11:44:00 2003 From: nathanael at gnat.ca (Nathanael Noblet) Date: Wed Apr 30 11:44:00 2003 Subject: Flashing Bios Success... Message-ID: <66119B33-7B27-11D7-BEA3-0003931B4D6A@gnat.ca> Hello all, So I've been trying for quite some time to gain access to the bios of the SBC I'm using so I can start placing linuxbios on it. Unable to get flash_rom to work and devbios segfaulting in some pci detection functions I was a little stuck. I didn't have a floppy drive on that SBC so couldn't use the uniflash dos programmer. Well I have found a way to get a dos boot and the uniflasher working. I have saved the original bios to file, and then flashed a fresh bios with the saved bios and its all working great. So though I would love to get flash_rom working or devbios for that matter. So now I'd like to know how to proceed. I have a build that builds I really have no idea how functional it is, and I fully expect some bugs. What I would LIKE to do right now is flash the flash chip with linuxbios (ignoring the kernel for the moment) just to see if it initializes the serial output etc as I want it to. So I've looked over the HOWTOs for the other boards. My board has a DoC, but I don't want to use it yet. So I assumed that I need to flash linuxbios or linuxbios.strip onto the flash. I tried, but uniflash gives me an error #100 unable to read file??? As well, my flash is 256K and the linuxbios is 24K... do I need to pad it? As well when I do the STD_FLASH,ROM_SIZE options and such the romimage is still 900K. I though originally that I would put the romimage on the flash.. it is obviously too big... The last question I have is the boot block in the flash chip itself there are two 8K boot blocks at the beginning and end of the flash, does something specific go in that area of the flash? So can anyone enlighten me on the next step? -- Nathanael Noblet Gnat Solutions 4604 Monterey Ave NW Calgary, AB T3B 5K4 T/F 403.288.5360 C 403.809.5368 http://www.gnat.ca/ From rminnich at lanl.gov Wed Apr 30 12:05:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 12:05:01 2003 Subject: Flashing Bios Success... In-Reply-To: <66119B33-7B27-11D7-BEA3-0003931B4D6A@gnat.ca> Message-ID: the file you want is called romimage. Try that. ron From rsmith at bitworks.com Wed Apr 30 13:11:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 13:11:01 2003 Subject: errors compiling the vga bios stuff In-Reply-To: References: Message-ID: <3EB00B70.5020505@bitworks.com> ron minnich wrote: > dammit. Version creep. OK, I'll take a look. Last time I tried this it > worked. I figured out how to get pcibios.o into the makefile. Looks like freebios/src/config/Config needs a dir/bioscall added to the list of std directories to parse. This gets pcibios.o in the works based on the CONFIG_PCIBIOS setting. The problem now is that pcibios won't compile. pcibios() has the following in it. busdevfn = dev->bus->number | (dev->devfn & 0xff) where dev is a pci_dev structure. bus is also a pci_dev struct. According to include/pci.h pci_dev dosen't have a member named 'number' and thus it errors. The the layout of pci_dev changed any reciently ? -- Richard A. Smith rsmith at bitworks.com From rminnich at lanl.gov Wed Apr 30 13:28:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 13:28:00 2003 Subject: errors compiling the vga bios stuff In-Reply-To: <3EB00B70.5020505@bitworks.com> Message-ID: OK, I am currently trying to sync my tree back up with cvs, then will look at this. Owners of mainboards, please see if I'm getting close. Eric how is the LNXI tree syncup doing? ron From rminnich at lanl.gov Wed Apr 30 13:37:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 13:37:00 2003 Subject: errors compiling the vga bios stuff In-Reply-To: <3EB00B70.5020505@bitworks.com> Message-ID: On Wed, 30 Apr 2003, Richard Smith wrote: > I figured out how to get pcibios.o into the makefile. Looks like > freebios/src/config/Config needs a dir/bioscall added to the list of std > directories to parse. dir/bioscall is very architecture-dependent, so this will have to go somewhere else. I'll look. I'll look at that other weird problem too. ron From rminnich at lanl.gov Wed Apr 30 13:43:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 13:43:00 2003 Subject: SERIAL_POST_TSC Message-ID: this option now works again. Very handy for timing. ron From rminnich at lanl.gov Wed Apr 30 13:48:46 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 13:48:46 2003 Subject: version freeze Message-ID: I'm pushing it off a week until I get these compile bugs worked out. So next target date for the freeze is may 8. We're close. ron From dkotian3 at vsnl.net Wed Apr 30 13:55:00 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 30 13:55:00 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT References: Message-ID: <006001c30f45$5128e320$3b3e41db@vsnl.net> I was reading the FAQ on etherboot, It says insmod bios.o dd if=natsemi.elf of=/dev/bios bs=64k dd if=linuxbios.rom of=/dev/bios bs=64k seek=1 looks to be valid for TFTP , could someone kindly please elaborate the etherboot procedure in my case. Regards Deepak ----- Original Message ----- From: "ron minnich" To: "Deepak Kotian" Cc: Sent: Wednesday, April 30, 2003 12:37 AM Subject: Re: BIOS FLASHROMS AND COMPACT FLASH BOOT > On Wed, 30 Apr 2003, Deepak Kotian wrote: > > > I have a system, which has 256/512K BIOS FLASH ROMS with 32MB Compact > > flash IDE. And the kernel image is residing on Compact flash. I can > > change the partioning as per the needs for LinuxBIOS to work. I just > > want BOOT Linux and no other O.S for now. > > What we do here: > > burn linuxbios on flash with an etherboot payload > > burn the linux kernel onto the Compact Flash. > > ron > > _______________________________________________ > Linuxbios mailing list > Linuxbios at clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios > From dkotian3 at vsnl.net Wed Apr 30 14:00:46 2003 From: dkotian3 at vsnl.net (Deepak Kotian) Date: Wed Apr 30 14:00:46 2003 Subject: BIOS FLASHROMS AND COMPACT FLASH BOOT References: <019301c30e83$5a190680$303e41db@vsnl.net> <1051667624.1635.6.camel@ollie> Message-ID: <006101c30f45$51e184c0$3b3e41db@vsnl.net> Also, why can we use ADLO or direct elfimage boot in my case. Could I get some more details on this one. ----- Original Message ----- From: "ollie lho" To: "Deepak Kotian" Cc: "Ronald G Minnich" ; "LinuxBIOS Mailing List" Sent: Wednesday, April 30, 2003 7:39 AM Subject: Re: BIOS FLASHROMS AND COMPACT FLASH BOOT > On Wed, 2003-04-30 at 03:13, Deepak Kotian wrote: > > Thanks. > > Can I please get a sample config as well. Or the config file settings > > in this situation. > > > > See the winfast-elf-etherboot.config > > -- > ollie lho > > From nathanael at gnat.ca Wed Apr 30 14:07:01 2003 From: nathanael at gnat.ca (Nathanael D. Noblet) Date: Wed Apr 30 14:07:01 2003 Subject: Flashing Bios Success... In-Reply-To: References: Message-ID: <1051727474.32216.7.camel@platinum.gnat.ca> On Wed, 2003-04-30 at 10:38, ron minnich wrote: > the file you want is called romimage. Ok but the file romimage is 900K and my flash is but a measly 256K here is my config file... # Sample config file for Arbor PIA 671 SBC with DoC Millennium (as root) # This will make a target directory of ./arbour target arbor mainboard arbor/pia-671 # Enable Serial Console for debugging option SERIAL_CONSOLE=1 option TTYS0_BAUD=115200 option DEFAULT_CONSOLE_LOGLEVEL=9 #option DEBUG=1 # Use 256KB Standard Flash as Normal BIOS #option RAMTEST=1 option USE_GENERIC_ROM=1 option STD_FLASH=1 #option ZKERNEL_START=0xfffc0000 option ROM_SIZE=262144 # use DOC MIL #option USE_DOC_2000=1 #docipl northsouthbridge/sis/530/ipl.S # Use the internal VGA frame buffer device option HAVE_FRAMEBUFFER=1 # Path to your kernel (vmlinux) linux /usr/src/linux # Kernel command line parameters commandline root=/dev/hda1 console=tty0 console=ttyS0 The other question I have is this, the linux /usr/src/linux line points to my kernel, but I'm wondering, when I normally build a kernel I copy from /usr/src/linux/arch/i386/boot/bzImage and not /usr/src/linux/vmlinux so what exactly is the vmlinux part in /usr/src and what is it used for? I mean I can see by using file that vmlinux us a ELF File and all that, and bzImage is a x86 boot sector. but well, I normally copy bzImage over and that is it... so don't know what the vmlinux file is used for to tell you the truth, and up until now didn't know it existed... Is that *supposed* to be the kernel I use to boot or the vmlinux file? -- Nathanael D. Noblet Gnat Solutions From rminnich at lanl.gov Wed Apr 30 14:54:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 14:54:01 2003 Subject: Flashing Bios Success... In-Reply-To: <1051727474.32216.7.camel@platinum.gnat.ca> Message-ID: On 30 Apr 2003, Nathanael D. Noblet wrote: > On Wed, 2003-04-30 at 10:38, ron minnich wrote: > > the file you want is called romimage. > Ok but the file romimage is 900K and my flash is but a measly 256K here > is my config file... then you can't put a linux payload in there. You have to use etherboot. We don't have the quantum storage option working yet. :-) ron From rminnich at lanl.gov Wed Apr 30 15:08:00 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 15:08:00 2003 Subject: errors compiling the vga bios stuff In-Reply-To: <3EB00B70.5020505@bitworks.com> Message-ID: On Wed, 30 Apr 2003, Richard Smith wrote: First, I just fixed the silly typo (not sure where it came from) with IRQ_SLOT_COUNT Second, as regards VGA, no build errors here. Here is my .config for an epia. See the last two lines. And vgabios.o and idt.o get build and linked in. weird. on # # LinuxBIOS config file for: VIA epia mini-itx # target epia # via epia mainboard via/epia # Enable Serial Console for debugging option SERIAL_CONSOLE=1 option TTYS0_BAUD=115200 option DEFAULT_CONSOLE_LOGLEVEL=9 option DEBUG=1 # Use 256KB Standard Flash as Normal BIOS option RAMTEST=1 option USE_GENERIC_ROM=1 option STD_FLASH=1 #option ZKERNEL_START=0xfffc0000 option ROM_SIZE=262144 # payload size = 192KB option PAYLOAD_SIZE=196608 # use ELF Loader to load Etherboot option USE_ELF_BOOT=1 # Use Etherboot as our payload payload ../lnxieepro100.ebi option CONFIG_VGABIOS=1 option CONFIG_REALMODE_IDT=1 From nathanael at gnat.ca Wed Apr 30 15:35:01 2003 From: nathanael at gnat.ca (Nathanael D. Noblet) Date: Wed Apr 30 15:35:01 2003 Subject: Flashing Bios Success... In-Reply-To: References: Message-ID: <1051733367.393.4.camel@platinum.gnat.ca> On Wed, 2003-04-30 at 13:27, ron minnich wrote: > then you can't put a linux payload in there. You have to use etherboot. We > don't have the quantum storage option working yet. :-) Right. What I wanted to do was simply see if the initializations were working, I guess it is probably smarter to just get it working properly in the first place. This board has both the PLCC flash and a DoC. Most of the configs I've seen replace the flash part with a DoC. So since I'm not doing that, how do I go about flashing the flash part? Do I still use etherboot to get the kernel and such from the DoC? If so where are the docs for configuring etherboot for such a setup? I don't think I've seen a config with both the standard flash AND a DoC... Are there any? -- Nathanael D. Noblet Gnat Solutions From rminnich at lanl.gov Wed Apr 30 15:47:01 2003 From: rminnich at lanl.gov (ron minnich) Date: Wed Apr 30 15:47:01 2003 Subject: Flashing Bios Success... In-Reply-To: <1051733367.393.4.camel@platinum.gnat.ca> Message-ID: On 30 Apr 2003, Nathanael D. Noblet wrote: > What I wanted to do was simply see if the initializations were working, > I guess it is probably smarter to just get it working properly in the > first place. This board has both the PLCC flash and a DoC. Most of the > configs I've seen replace the flash part with a DoC. So since I'm not > doing that, how do I go about flashing the flash part? Do I still use > etherboot to get the kernel and such from the DoC? If so where are the > docs for configuring etherboot for such a setup? I don't think I've seen > a config with both the standard flash AND a DoC... Are there any? nowadays there are fewer and fewer DoC users. Let's get linuxbios working first for you and then talk about options. First we can try to solve the flash problem. I'm printing the manual now. ron From rsmith at bitworks.com Wed Apr 30 17:40:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 17:40:01 2003 Subject: ADLO no go Message-ID: <3EB04A5E.9030303@bitworks.com> I'm trying to get ADLO up but once it jumps to adlo nothing happens. I've applied the serial-debug patch to ADLO but still no results. Are there any specific memory region options that I need to have enabled /disabled? Heres the snipit from my serial output. Welcome to elfboot, the open sourced starter. January 2002, Eric Biederman. Version 1.2 Using stream address 0x9c00 37:init_bytes() - zkernel_start:0xfff00000 zkernel_mask:0x0000ffff Found ELF candiate at offset 0 New segment addr 0x7c00 size 0x20400 offset 0x100 filesize 0x20400 (cleaned up) New segment addr 0x7c00 size 0x20400 offset 0x100 filesize 0x20400 Loading Segment: addr: 0x0000000007f67368 memsz: 0x0000000000020400 filesz: 0x0000000000020400 Jumping to boot code at 0x7c00 -- Richard A. Smith rsmith at bitworks.com From adam at cfar.umd.edu Wed Apr 30 17:51:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 17:51:01 2003 Subject: ADLO no go In-Reply-To: <3EB04A5E.9030303@bitworks.com> Message-ID: <20030430183228.H53522-100000@www.missl.cs.umd.edu> what motherboard? what chipset? did you set registers that control shadow ram right? (at loader.s) On Wed, 30 Apr 2003, Richard Smith wrote: > I'm trying to get ADLO up but once it jumps to adlo nothing happens. > > I've applied the serial-debug patch to ADLO but still no results. > > Are there any specific memory region options that I need to have enabled > /disabled? > > Heres the snipit from my serial output. > > Welcome to elfboot, the open sourced starter. > January 2002, Eric Biederman. > Version 1.2 > > Using stream address 0x9c00 > 37:init_bytes() - zkernel_start:0xfff00000 zkernel_mask:0x0000ffff > Found ELF candiate at offset 0 > New segment addr 0x7c00 size 0x20400 offset 0x100 filesize 0x20400 > (cleaned up) New segment addr 0x7c00 size 0x20400 offset 0x100 filesize > 0x20400 > Loading Segment: addr: 0x0000000007f67368 memsz: 0x0000000000020400 > filesz: 0x0000000000020400 > Jumping to boot code at 0x7c00 > > > > > -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rsmith at bitworks.com Wed Apr 30 20:02:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 20:02:01 2003 Subject: ADLO is GO In-Reply-To: <20030430194741.Q53522-100000@www.missl.cs.umd.edu> References: <20030430194741.Q53522-100000@www.missl.cs.umd.edu> Message-ID: <3EB06BC2.1040406@bitworks.com> Ok. that fixed it.. ADLO is booting for me. For the list: I had to enable read and write in the shadow memory. I don't think you had to but I enabled the full range from 0c0000 to 0ffffff Adam: I added a macro that does a PCI config space write. Saves you from haveing to do all the calulations. Would you like a diff? It doesn't appear however that my video bios is getting called. I rememember seeing somewhere that I have to enable that? -- Richard A. Smith rsmith at bitworks.com From adam at cfar.umd.edu Wed Apr 30 20:09:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 20:09:00 2003 Subject: ADLO is GO In-Reply-To: <3EB06BC2.1040406@bitworks.com> Message-ID: <20030430205053.P54030-100000@www.missl.cs.umd.edu> > Ok. that fixed it.. ADLO is booting for me. sweet. > For the list: > > I had to enable read and write in the shadow memory. I don't think you > had to but I enabled the full range from 0c0000 to 0ffffff FWIW, need for doing this mentioned in the "HACKING" file. > Adam: > > I added a macro that does a PCI config space write. Saves you from > haveing to do all the calulations. > > Would you like a diff? sure. can you send it to me and list so that it gets archived. Though, as I said in the long term we probalby want to have this sort of stuff done in the LinuxBIOS rather than ADLO so that it can stay (err become :-) architecture independent. > It doesn't appear however that my video bios is getting called. I > rememember seeing somewhere that I have to enable that? I guess it is differnet for each case. Not sure what's it like in your particular case. What's is this gfx card? overall BOCHS BIOS should find the binary only vga bios in memory and try to initalize it. there may be some stuff in linuxbios to enable too. not sure. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From adam at cfar.umd.edu Wed Apr 30 20:14:46 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 20:14:46 2003 Subject: ADLO is GO In-Reply-To: <3EB06BC2.1040406@bitworks.com> Message-ID: <20030430204455.N53522-100000@www.missl.cs.umd.edu> On Wed, 30 Apr 2003, Richard Smith wrote: > Ok. that fixed it.. ADLO is booting for me. sweet. > For the list: > > I had to enable read and write in the shadow memory. I don't think you > had to but I enabled the full range from 0c0000 to 0ffffff FWIW, need for doing this mentioned in the "HACKING" file. > Adam: > > I added a macro that does a PCI config space write. Saves you from > haveing to do all the calulations. > > Would you like a diff? sure. can you send it to me and list so that it gets archived. Though, as I said in the long term we probalby want to have this sort of stuff done in the LinuxBIOS rather than ADLO so that it can stay (err become :-) architecture independent. > It doesn't appear however that my video bios is getting called. I > rememember seeing somewhere that I have to enable that? I guess it is differnet for each case. Not sure what's it like in your particular case. What's is this gfx card? overall BOCHS BIOS should find the binary only vga bios in memory and try to initalize it. there may be some stuff in linuxbios to enable too. not sure. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rsmith at bitworks.com Wed Apr 30 20:21:12 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 20:21:12 2003 Subject: ADLO is GO In-Reply-To: <20030430204455.N53522-100000@www.missl.cs.umd.edu> References: <20030430204455.N53522-100000@www.missl.cs.umd.edu> Message-ID: <3EB06FC8.2030503@bitworks.com> Adam Sulmicki wrote: >>I had to enable read and write in the shadow memory. I don't think you >>had to but I enabled the full range from 0c0000 to 0ffffff > > FWIW, need for doing this mentioned in the "HACKING" file. Ah. RTFM then. Figures the one file I didn't read. >>It doesn't appear however that my video bios is getting called. I >>rememember seeing somewhere that I have to enable that? I guess it is differnet for each case. Not sure what's it like in your > particular case. > > What's is this gfx card? Asiliant (used to be C&T) 69000. > overall BOCHS BIOS should find the binary only vga bios in memory and try > to initalize it. Should there be any serial debug output? I don't see any thing that relates to video bios init. > there may be some stuff in linuxbios to enable too. not sure. Heres what came out the serial port: rombios.c,v 1.1 2002/11/25 02:07:53 rminnich Exp $ FATAL: Keyboard RESET error:993 FATAL: Keyboard RESET error:31 FATAL: Keyboard RESET error:994 FATAL: Keyboard RESET error:995 FATAL: Keyboard RESET error:996 Bochs BIOS, 1 cpu, $Revision: 1.1 $ $Date: 2002/11/25 02:07:53 $ [BOCHS BIOS VER:1.79] [COMPILE DATE:Apr 30 2003 TIME:16:50:43] DEVICE:0 ata0 master: PCHS=246/2/32 translation=lba LCHS=15/16/63 ata0 master: Hitachi CV 7.1.1 ATA-0 Hard-Disk (7 MBytes) Booting from Hard Disk... rombios.c,v 1.1 2002/11/25 02:07:53 rminnich Exp $ FATAL: Keyboard RESET error:993 FATAL: Keyboard RESET error:31 FATAL: Keyboard RESET error:994 FATAL: Keyboard RESET error:995 FATAL: Keyboard RESET error:996 KBD: int09h_handler(): unknown scancode read! -- Richard A. Smith rsmith at bitworks.com From adam at cfar.umd.edu Wed Apr 30 20:27:01 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 20:27:01 2003 Subject: ADLO is GO In-Reply-To: <3EB06FC8.2030503@bitworks.com> Message-ID: <20030430210201.E54030-100000@www.missl.cs.umd.edu> > > What's is this gfx card? > > Asiliant (used to be C&T) 69000. > > > overall BOCHS BIOS should find the binary only vga bios in memory and try > > to initalize it. > > Should there be any serial debug output? I don't see any thing that > relates to video bios init. nope. it is binary only vga bios. we have no control over it. it is why I don't like this solution. It is bit of black magic to get it to work. Did you try to boot with normal pc bios and then warm-boot into linuxbios/adlo? see if it helps. It might give you some tips. > > there may be some stuff in linuxbios to enable too. not sure. > Heres what came out the serial port: > > Bochs BIOS, 1 cpu, $Revision: 1.1 $ $Date: 2002/11/25 02:07:53 $ > [BOCHS BIOS VER:1.79] > [COMPILE DATE:Apr 30 2003 TIME:16:50:43] > > DEVICE:0 > ata0 master: PCHS=246/2/32 translation=lba LCHS=15/16/63 > ata0 master: Hitachi CV 7.1.1 ATA-0 Hard-Disk (7 MBytes) > > Booting from Hard Disk... > rombios.c,v 1.1 2002/11/25 02:07:53 rminnich Exp $ > FATAL: Keyboard RESET error:993 > FATAL: Keyboard RESET error:31 > FATAL: Keyboard RESET error:994 > FATAL: Keyboard RESET error:995 > FATAL: Keyboard RESET error:996 > KBD: int09h_handler(): unknown scancode read! umm are keyboard and mice connected? if no, I would connect them. umm what's at /dev/hda did you put there linuxbios or something? overall BOCHS BIOS will try to read MBR from /dev/hda and boot it. I would put there lilo and configure it for serial console. Then u should be able ot se if it reads MBR correctly and boot lilo. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rsmith at bitworks.com Wed Apr 30 20:44:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 20:44:01 2003 Subject: I got VSYNC but no text In-Reply-To: <20030430210201.E54030-100000@www.missl.cs.umd.edu> References: <20030430210201.E54030-100000@www.missl.cs.umd.edu> Message-ID: <3EB0759B.2040506@bitworks.com> Adam Sulmicki wrote: Ok... Looks like it was actually working but I just didn't notice since the screen was blank. But VSYNCS start happening so I know the chip is being initialized. The question is why is it blank. Does that serial debug patch route everything out the serial port. Or does is go to both the screen and the serial port? -- Richard A. Smith rsmith at bitworks.com From rsmith at bitworks.com Wed Apr 30 20:56:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 20:56:01 2003 Subject: ADLO Video is up In-Reply-To: <20030430212645.M54030-100000@www.missl.cs.umd.edu> References: <20030430212645.M54030-100000@www.missl.cs.umd.edu> Message-ID: <3EB07880.8010909@bitworks.com> Adam Sulmicki wrote: > can you just diasable/remove debug patch and see what happen? i actually > never used that patch. Thats what it was. I commented out SERIAL_DEBUG and now I have text on my monitor. Dude this is COOL. Ok... Now. This board is a multi-head board so there are actually 2 69000 on the PCI bus. I also need to init the second chip as well. The way we did this under General software was to activate chip 2. run bios. Deactivate chip 2. Activate chip 1 and then re run the bios. Would I have to do such a thing for BOCHS as well or is there a better way. -- Richard A. Smith rsmith at bitworks.com From adam at cfar.umd.edu Wed Apr 30 21:03:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 21:03:00 2003 Subject: ADLO Video is up In-Reply-To: <3EB07880.8010909@bitworks.com> Message-ID: <20030430214050.C54030-100000@www.missl.cs.umd.edu> > > can you just diasable/remove debug patch and see what happen? i actually > > never used that patch. > > Thats what it was. I commented out SERIAL_DEBUG and now I have text on > my monitor. Dude this is COOL. Sweet. That makes 3hrs to get ADLO to work. A new record I suppose :-) (but the truth is that it depends how much trouble is the vga bios). > Ok... Now. This board is a multi-head board so there are actually 2 > 69000 on the PCI bus. I also need to init the second chip as well. > The way we did this under General software was to activate chip 2. run > bios. Deactivate chip 2. Activate chip 1 and then re run the bios. > Would I have to do such a thing for BOCHS as well or is there a better > way. well, I don't see much problem doing this way. the bochs code is fairly "user friendly". Shouldn't be hard to do, especially that you already did that before. But then maybe someone else knows better way? -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers From rsmith at bitworks.com Wed Apr 30 22:44:01 2003 From: rsmith at bitworks.com (Richard Smith) Date: Wed Apr 30 22:44:01 2003 Subject: ADLO Video is up In-Reply-To: <20030430214050.C54030-100000@www.missl.cs.umd.edu> References: <20030430214050.C54030-100000@www.missl.cs.umd.edu> Message-ID: <3EB091AF.7040008@bitworks.com> Adam Sulmicki wrote: > well, I don't see much problem doing this way. the bochs code is fairly > "user friendly". Shouldn't be hard to do, especially that you already did > that before. > > But then maybe someone else knows better way? How does one go about accessing the PCI config space withing bochs? I see that it provides pcibios functionality but I can't find an example where any section reads or writes from PCI config space. -- Richard A. Smith rsmith at bitworks.com From adam at cfar.umd.edu Wed Apr 30 22:47:00 2003 From: adam at cfar.umd.edu (Adam Sulmicki) Date: Wed Apr 30 22:47:00 2003 Subject: ADLO Video is up In-Reply-To: <3EB091AF.7040008@bitworks.com> Message-ID: <20030430232840.X54516-100000@www.missl.cs.umd.edu> > > well, I don't see much problem doing this way. the bochs code is fairly > > "user friendly". Shouldn't be hard to do, especially that you already did > > that before. > > > > But then maybe someone else knows better way? > > How does one go about accessing the PCI config space withing bochs? maybe it is best if you just do it the same way you did it in loader.s > I see that it provides pcibios functionality but I can't find an example > where any section reads or writes from PCI config space. I'm not even sure if it is complete thing. I was under impression it is more a "stub code", and not much more. -- Adam Sulmicki http://www.eax.com The Supreme Headquarters of the 32 bit registers