If nvidia-drivers are installed with +acpi flag, after resuming from sleep (on kde-4.6.2), random 2-3 s freezes are experienced. X seems to be the process which actually freezes. The problem goes away after recompiling nvidia-drivers with -acpi. This problem has been already reported several times: https://bugs.archlinux.org/task/22731 http://www.nvnews.net/vbulletin/showthread.php?t=158248 The workaround reported in bug #363363 (to disable acpid) is not enough. The freezes are only gone after recompiling nvidia-drivers with -acpi. I would suggest to add a message in the nvidia-drivers ebuild to warn against those freezes, or to disable the acpi flag altogether until the problem is solved. Reproducible: Always
Not sure I have the exact same issue but using x11-drivers/nvidia-drivers-285.05.09-r1 I'm experiencing a similar issue: The system crashes systematically after resume, I don't have this problem with x11-drivers/nvidia-drivers-275.09.07. In both cases, it has been compiled with +acpi. I will recompile the 285.05.09 version with -acpi to see if this is related to this issue or if I should open a new bug report.
I can still confirm this issue. Suspending (from KDE) then resuming systematically produces a crash. Confirmed with x11-drivers/nvidia-drivers 285.05.09-r1 and 290.10. Using -acpi solves the issue. What will I miss by disabling the acpi flag? Shouldn't it be the default until this issue if fixed?
USE=acpi simply RDEPENDS on the acpid daemon. It doesn't affect the code in anyway. If you have the acpid daemon running then the driver will connect to it no matter what if the acpi USE flag is on or off. So it doesn't make sense that disabling the acpid daemon didn't work but recompiling without USE=acpi did.
Please use nvidia-bug-report.sh and submit the issue to linux-bugs@nvidia.com along with the output of nvidia-bug-report.sh The issue you're having is in the closed portion of the code so there's not much we can do.
