Bug 149791 – ati-drivers 8.28.8 won't compile against *-sources-2.6.18

• Status: RESOLVED NEEDINFO
• Product: Gentoo Linux

[Description Iain Buchanan 2006-10-01 17:18:07 UTC]

wow there are a lot of ati bugs atm...

anyway,
8.29.6 compiles against suspend2-sources-2.6.18
8.28.8 compiles against suspend2-sources-2.6.17-r6

but
8.28.8 doesn't compile against suspend2-sources-2.6.18.

I need to go back to 8.28.8 because 8.29.6 doesn't suspend (at least I suspect that's the reason - I'm trying to test it :)

attachments to follow...

[Comment 1 Iain Buchanan 2006-10-01 17:19:32 UTC]

Created attachment 98572 [details]
output from emerge --info

[Comment 2 Iain Buchanan 2006-10-01 17:30:50 UTC]

Created attachment 98573 [details]
output from emerge -u ati-drivers

[Comment 3 Iain Buchanan 2006-10-01 18:10:44 UTC]

Created attachment 98574 [details, diff]
patch to overcome undefined VM_SHM.

This is a patch from [1] that fixes the compile problem, but only the patch concerning VM_SHM.  Another patch there concerning UTS_RELEASE I understand is fixed already.

I added this patch after the powermode patch in the 8.28.8 ebuild, and the module seems to compile, load and run correctly.

[1] http://www.jikos.cz/jikos/dev/fglrx_8.12.26_kernel_2.6.18-rc4.patch

[Comment 4 Iain Buchanan 2006-10-01 18:37:51 UTC]

FYI the suspend issue isn't the ati-driver, it seems to be suspend2-sources.

Also, the patch I attached seems to be a bit funny in setting VM_SHM to 0x00000000, but it's working fine - same frame rates as before.

thanks.

[Comment 5 Jakub Moc (RETIRED) 2006-10-02 00:01:32 UTC]

suspend2 has its own bugzilla, please report such issues there. Thanks.

http://bugzilla.suspend2.net/index.cgi

[Comment 6 Iain Buchanan 2006-10-02 00:12:23 UTC]

:D trigger happy!

perhaps I wasn't clear, but this doesn't depend on suspend2 (that's just what I use) it depends on 2.6.18 of any flavour, according to the various posts I can find on the web about it.

The suspend2 folks would be a little suprised if I submitted an ebuild patch to their bugzilla :)

If I am still wrong, then I apologise profusely :)

[Comment 7 Jakub Moc (RETIRED) 2006-10-02 00:14:59 UTC]

> I need to go back to 8.28.8 because 8.29.6 doesn't suspend 

So, can you explain why are we supposed to patch the thing when 8.29.6 works just fine?

[Comment 8 Iain Buchanan 2006-10-02 00:24:05 UTC]

> > I need to go back to 8.28.8 because 8.29.6 doesn't suspend 
> FYI the suspend issue isn't the ati-driver, it seems to be suspend2-sources.

so far 8.29.6 works just fine. I gather that a whole lot of modular-xorg-ers haven't tried it yet.  I'm just trying to be helpful incase 8.28.8 makes it stable and 8.29.6 doesn't for some reason.  Otherwise, why is 8.28.8 still in portage?  If it's in portage, someone is bound to use it.

Anyway, do whatever you want - at the very least hopefully someone will stumble across this from google now that it's written down somewhere :)