xorg/gnome freezing, possible kernel regression

[Sérgio Basto]

On Qui, 2013-10-17 at 16:57 -0600, Chris Murphy wrote: 
> I'm getting something of a regression with kernel 3.11.5-300 today that doesn't happen with 3.11.4 but I don't know what's instigating it. The result is Gnome freezes, including the mouse arrow. Capslock works, and I still have access via ssh. I get a bunch of kernel messages, it starts with Xorg page allocation failure. selinux is in the trace, but I have no AVC messages other than the already reported tmpfile and systemd readahead alerts.
> 
> Should I file a bug against the kernel or xorg? Or?
> 
> 
> 
> Oct 17 16:42:20 f20s.localdomain kernel: Xorg: page allocation failure: order:4, mode:0x10c0d0
> Oct 17 16:42:20 f20s.localdomain kernel: CPU: 0 PID: 1201 Comm: Xorg Not tainted 3.11.5-300.fc20.x86_64 #1
> Oct 17 16:42:20 f20s.localdomain kernel: Hardware name: Apple Inc. MacBookPro4,1/Mac-F42C89C8, BIOS    MBP41.88Z.00C1.B03.0802271651 02/27/08
> Oct 17 16:42:20 f20s.localdomain kernel:  0000000000000000 ffff88012e6c9978 ffffffff8164871b 000000000010c0d0
> Oct 17 16:42:20 f20s.localdomain kernel:  ffff88012e6c9a00 ffffffff81141850 0000004000000000 ffff88013ffe5b38
> Oct 17 16:42:20 f20s.localdomain kernel:  ffffffff81143e36 000000000000000f 00000004810c7013 0010c0d000000010
> Oct 17 16:42:20 f20s.localdomain kernel: Call Trace:
> Oct 17 16:42:20 f20s.localdomain kernel:  [<ffffffff8164871b>] dump_stack+0x45/0x56
> Oct 17 16:42:20 f20s.localdomain kernel:  [<ffffffff81141850>] warn_alloc_failed+0xf0/0x160
> Oct 17 16:42:20 f20s.localdomain kernel:  [<ffffffff81143e36>] ? drain_local_pages+0x16/0x20
> Oct 17 16:42:20 f20s.localdomain kernel:  [<ffffffff81145bb7>] __alloc_pages_nodemask+0x827/0xa30
> Oct 17 16:42:20 f20s.localdomain kernel:  [<ffffffff81183139>] alloc_pages_current+0xa9/0x170
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811406b1>] __get_free_pages+0x21/0x70
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff8118cc7e>] kmalloc_order_trace+0x2e/0xa0
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811d7051>] ? __block_commit_write.isra.19+0x81/0xb0
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff8118f09a>] __kmalloc+0x1ca/0x250
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811ac45c>] ? cdev_get+0x2c/0x50
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff814b2e29>] evdev_open+0x79/0x1d0
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811acbac>] chrdev_open+0x9c/0x180
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811a605b>] do_dentry_open+0x1eb/0x280
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811acb10>] ? cdev_put+0x30/0x30
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811a6121>] finish_open+0x31/0x40
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811b69ca>] do_last+0x4ca/0xe00
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff8129ea6c>] ? selinux_file_alloc_security+0x3c/0x60
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811b73bb>] path_openat+0xbb/0x690
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811c584e>] ? mntput_no_expire+0x3e/0x120
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811b805a>] do_filp_open+0x3a/0x90
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811c3d7d>] ? __alloc_fd+0x7d/0x110
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811a74fe>] do_sys_open+0x12e/0x210
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff811a75fe>] SyS_open+0x1e/0x20
> Oct 17 16:42:21 f20s.localdomain kernel:  [<ffffffff81657919>] system_call_fastpath+0x16/0x1b
> Oct 17 16:42:21 f20s.localdomain kernel: Mem-Info:
> Oct 17 16:42:21 f20s.localdomain kernel: Node 0 DMA per-cpu:
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    0: hi:    0, btch:   1 usd:   0
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    1: hi:    0, btch:   1 usd:   0
> Oct 17 16:42:21 f20s.localdomain kernel: Node 0 DMA32 per-cpu:
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    0: hi:  186, btch:  31 usd:   0
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    1: hi:  186, btch:  31 usd:   0
> Oct 17 16:42:21 f20s.localdomain kernel: Node 0 Normal per-cpu:
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    0: hi:  186, btch:  31 usd:   0
> Oct 17 16:42:21 f20s.localdomain kernel: CPU    1: hi:  186, btch:  31 usd:  47

I saw yesterday this bug 
https://bugzilla.kernel.org/show_bug.cgi?id=63101#c0

have the coincide to be a regression on kernel 3.11.5 
-- 
Sérgio M. B.