Fedora Core 5 Test Update: kernel-2.6.17-1.2171_FC5

[Rodd Clarkson]

On Tue, 2006-08-15 at 18:43 -0400, Dave Jones wrote: 
> On Wed, Aug 16, 2006 at 08:24:17AM +1000, Rodd Clarkson wrote:
> 
>  > Suspend and resume is still busted in this kernel on my Dell Inspiron
>  > 9300.
>  > 
>  > Seems to suspend okay, but wont resume.  Keyboard is dead and the only
>  > way to fix is to hold down the power button.
> 
> Some of the rawhide testers found that things broke a while ago
> and recently started working again.  The problem is I'm not entirely
> sure which changes are responsible.

Allow me to point a finger if I may.  I'm currently using
kernel-2.6.17-1.2174_FC5

I (like many fedora users) use livna and in particular, since I have a
nvidia graphics card, I use the binary nvidia supplied using livna. I
also use updates-testing.  Recently (the last few days) I did a yum
update and livna offered up a package for my kernel (which is currently
in updates-testing).  This doesn't happen that often.

I booted into kernel-2.6.17-1.2174_FC5 to give it a try (since the
nvidia driver had also been upgraded) and for a lark I tried out
suspend/resume, only to find it worked.  I've been using
kernel-2.6.17-1.2139_FC5 for some time since suspend/resume works and
this makes using my laptop a lot nicer.

This got me to thinking (since you complained that this suspend and
resume problems seems to come and go and you can't seem to pin down what
was causing it) that maybe the problem isn't the kernel, maybe its the
xorg-x11-drv-nv package. 

In the true spirit of a good tester, i disables the nvidia-glx service
(which toggles xorg.conf files depending on whether or not the nvidia
kernel module exist) and then restarted X, which start, but using the
'nv' driver (and not the 'nvidia' driver).  Suspend and (particularly)
resume failed.

Restart the system, turn back on the nvidia-glx driver support and
suspend and resume works again.

> As a worst case, hopefully a rebase to .18 when its released will fix it,
> but hopefully before then we can figure out which patch broke things
> and just revert that change.

So, to cut a long story short, I don't think it's a kernel related
patch, but something to do with xorg-x11-drv-nv

Can someone confirm this and (regardless) do I file a bug report under
this driver?


R.
-- 
"It's a fine line between denial and faith.
 It's much better on my side"