Messages in this thread | | | From | "Rafael J. Wysocki" <> | Subject | Re: [GIT PATCH] Driver Core patches for 2.6.18 | Date | Wed, 27 Sep 2006 08:58:42 +0200 |
| |
On Wednesday, 27 September 2006 10:47, Mike Galbraith wrote: > On Tue, 2006-09-26 at 13:39 -0700, Greg KH wrote: > > On Tue, Sep 26, 2006 at 12:34:05PM +0000, Mike Galbraith wrote: > > > On Mon, 2006-09-25 at 22:37 -0700, Greg KH wrote: > > > > Here are a bunch of driver core and sysfs patches and fixes for 2.6.18. > > > > > > Hi, > > > > > > Just as an fyi, these patches cause a ~regression on my P4/HT box. > > > > > > Suspend stopped here working after 2.6.17. > > > > Does 2.6.18 also have these problems? > > Suspend failure, yes, kill box on failure, no. > > > If so, it's not due to these patches :) > > The 'Mikie's box' killer is definitely somewhere in these patches. I > decided to test them after having just happened to have tried > 2.6.18-rc7-mm1 to see if it the mm tree had fixed my suspend problem for > me, and then seeing your post. Since 2.6.18 with these patches went > from ho-hum broke to ding-dong dead, I figured I should mention it. > > > > I haven't dug into why, but > > > since then, I get a message "Class driver suspend failed for cpu0", and > > > the suspend fails, but everything works fine afterward. If I ignore the > > > return of drv->suspend(), the box will suspend and resume just fine, > > > both with this patch set and without. (which is what I've been doing > > > while waiting for it to fix itself or for my round toit to turn up) > > > > What driver is controling cpu0? > > What driver is failing the suspend? > > Dunno, see below. > > > What line did you have to change? > > drivers/base/sys.c:409 > > Adding a WARN_ON() there wasn't particularly illuminating. I enabled > debugging (this is virgin 2.6.18), and get the following. I see a > problem I didn't notice before, see 'Hmm' below. I have no idea if it > has anything to do with box killer problem with patchset applied.
Please try to remove the acpi_cpufreq driver before the suspend.
If that works, please add your system configuration to the bugzilla entry at http://bugzilla.kernel.org/show_bug.cgi?id=7188
Greetings, Rafael
-- You never change things by fighting the existing reality. R. Buckminster Fuller - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |