[coreboot] patch: add support for blobs; example SMM blob included
c-d.hailfinger.devel.2006 at gmx.net
Fri Nov 28 04:18:23 CET 2008
On 26.11.2008 23:41, ron minnich wrote:
> On Wed, Nov 26, 2008 at 2:19 PM, Peter Stuge <peter at stuge.se> wrote:
>> ron minnich wrote:
>>> +++ arch/x86/Makefile (working copy)
>>> @@ -48,6 +48,13 @@
>>> COMPRESSFLAG := -C nrv2b
>>> +# all depends on coreboot.rom and BLOBS.
>>> +# BLOBS may be empty. They are things such as smm that are searched for at runtime.
>>> +# They depend on the coreboot rom existing so that they can be placed there.
>>> +# The name of the blob is determined by the various blob rules. There is so much possible
>>> +# variation in them that it is not really possible to put a standard rule in this file.
>>> +rom:: $(obj)/coreboot.rom $(BLOBS)
>>> $(obj)/coreboot.rom $(obj)/coreboot.map: $(obj)/coreboot.bootblock $(obj)/util/lar/lar lzma nrv2b $(obj)/coreboot.initram $(obj)/coreboot.stage2 $(obj)/option_table
>>> $(Q)printf " LAR $(subst $(shell pwd)/,,$(@))\n"
>>> $(Q)rm -f $(obj)/coreboot.rom
>>> @@ -306,4 +313,4 @@
>>> $(Q)cp cscope.proj $(obj)/mainboard/$(MAINBOARDDIR)/kscope
>>> $(Q)sh util/mkdep $@ $(INITCFLAGS) "--" $(ALLSRC)
>>> # $(Q)sort -u -o $@ $@
>>> +.PHONY: rom
>> I don't like to add PHONY targets, nor special purpose toplevel
>> dependencies. Could the boards that need them simply depend on
> we have a bunch already. I don't see a big deal with rom as the
> target. I can take the PHONY out however.
>> I remember discussing blobs being sort-of stage3/4. Do we call them
>> that instead?
> no, that makes no sense. Blobs are all over. VSA blob is in stage1.
> SMM is installed in stage2 phase 6. They are not classifiable.
Do we want fallback/normal SMM? From a BIOS update POV, it would make sense.
More information about the coreboot