[LinuxBIOS] #31: Do proper checking for flash erase for SST FWH parts

[Segher Boessenkool]

>> The patch never made it to the mailing list.  To review the
>> patch I now have to:
>
> I contacted the trac folks for that problem. Attachments should  
> make it
> into notification mails.

Thanks!

>> Don't take this as a flame, I just need a working working
>> environment :-)
>
> Of course not. ;-) There was public discussion on the symposium about
> which tracker to use, and we all participated, so I am of course glad
> about your late feedback.
>
> Nobody wanted bugzilla nor roundup.

I wanted (a heavily patched) Bugzilla, but I of course know all
the reasons people hate it.  It "works for me" though ;-)

> Nobody opposed to trac either,
> so we worked around some problems, and bought us into some others.

Yeah, it seems trac isn't too good a fit for us yet.  Some
patching will help, I have no idea how much effort it is.  Do
we have contact with some friendly trac developer(s) who could
help us out?

> Patch management on the mailing list just does not work. It never  
> will.

It works fine (and has worked fine since basically forever) for:
-- linux-kernel and all its children lists;
-- gcc-patches;
-- binutils;
-- libc-alpha;
-- many other GNU mailing lists;
-- and who knows what else.

The only thing that's needed are some "maintainers" (or whatever
you want to call them, "gatekeepers" or whatever) who review
patches and ACK or NAK them for inclusion into the main tree.  And
hey, we have that already.

Even if patch review is done on the mailing list a) all actual
checkins still go over SVN so trac sees them; and b) there's nothing
that stops people from associating a mailing list mail with a
trac ticket (either manually write a comment saying "patch posted
at this-or-that-ml-archive-address", or have trac snoop the list
and recognise certain tags and attach the stuff by itself (for
example, "Closes: #21" and "Patch-for: #21", etc.)

> Here's the golden open source rule that applies to trac as well:
> If you want it fixed, send a fix ;-)

Trust me I would if I could.  But I a) don't have the time; and
b) I don't have the configuration files or the exact source code
for our installed version of trac.


Segher