<div>On Mon, Jan 16, 2012 at 5:36 PM, Daniel Vetter <span dir="ltr"><<a href="mailto:daniel@ffwll.ch">daniel@ffwll.ch</a>></span> wrote:</div><div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="HOEnZb"><div class="h5">On Mon, Jan 16, 2012 at 05:18:17PM +0100, CC wrote:<br>
> Hi,<br>
><br>
> I've heard that you need users having the RC6 bug.<br>
><br>
> I have the following setup:<br>
> CPU: Intel Core i5-2500K<br>
> Mainboard: ASRock Z68 Pro3-M<br>
> Memory: Corsair Vengeance CMZ8GX3M2A1866C9<br>
><br>
> Although the CPU doesn't support VT-d, I disabled all virtualization<br>
> support in the UEFI setup.<br>
><br>
> I use Arch Linux and Gnome 3 in the fallback mode. The problem is more<br>
> drastic without fallback mode, however.<br>
><br>
> Whenever I enable RC6, I get the a few of these errors in dmesg:<br>
><br>
> [ 48.900000] WARNING: at drivers/gpu/drm/i915/i915_drv.c:387<br>
> __gen6_gt_wait_for_fifo+0x94/0xa0 [i915]()<br>
> [ 48.900002] Hardware name: To Be Filled By O.E.M.<br>
> [ 48.900002] Modules linked in: ipv6 fuse ext2 snd_hda_codec_hdmi<br>
> snd_hda_codec_realtek mei(C) joydev r8169 shpchp pci_hotplug usbhid hid<br>
> snd_hda_intel iTCO_wdt mii iTCO_vendor_support i2c_i801 snd_hda_codec<br>
> processor snd_hwdep snd_pcm snd_timer snd soundcore snd_page_alloc psmouse<br>
> serio_raw pcspkr evdev ext4 mbcache jbd2 crc16 xhci_hcd ehci_hcd usbcore<br>
> i915 drm_kms_helper drm intel_agp i2c_algo_bit button intel_gtt i2c_core<br>
> video sd_mod ahci libahci libata scsi_mod<br>
> [ 48.900019] Pid: 623, comm: Xorg Tainted: G WC 3.1.9-2-ARCH #1<br>
> [ 48.900020] Call Trace:<br>
> [ 48.900023] [<ffffffff81061bef>] warn_slowpath_common+0x7f/0xc0<br>
> [ 48.900025] [<ffffffff81061c4a>] warn_slowpath_null+0x1a/0x20<br>
> [ 48.900028] [<ffffffffa00e0764>] __gen6_gt_wait_for_fifo+0x94/0xa0<br>
> [i915]<br>
> [ 48.900032] [<ffffffffa015d2d5>] ring_write_tail+0x65/0x120 [i915]<br>
> [ 48.900036] [<ffffffffa01619bc>] render_ring_flush+0xbc/0xe0 [i915]<br>
> [ 48.900040] [<ffffffffa010b803>] i915_gem_flush_ring+0x43/0x250 [i915]<br>
> [ 48.900044] [<ffffffffa0112b50>]<br>
> i915_gem_do_execbuffer.isra.7+0x1020/0x16d0 [i915]<br>
> [ 48.900048] [<ffffffffa01136bb>] i915_gem_execbuffer2+0x8b/0x240 [i915]<br>
> [ 48.900051] [<ffffffffa0098434>] drm_ioctl+0x3e4/0x4c0 [drm]<br>
> [ 48.900053] [<ffffffff810746cb>] ? recalc_sigpending+0x1b/0x50<br>
> [ 48.900057] [<ffffffffa0113630>] ? i915_gem_execbuffer+0x430/0x430<br>
> [i915]<br>
> [ 48.900059] [<ffffffff8101e9b1>] ? fpu_finit+0x21/0x40<br>
> [ 48.900061] [<ffffffff8116fddf>] do_vfs_ioctl+0x8f/0x500<br>
> [ 48.900063] [<ffffffff81014beb>] ? sys_rt_sigreturn+0x1eb/0x200<br>
> [ 48.900064] [<ffffffff811702e1>] sys_ioctl+0x91/0xa0<br>
> [ 48.900066] [<ffffffff8140c3c2>] system_call_fastpath+0x16/0x1b<br>
> [ 48.900067] ---[ end trace 9a23b8b32b16a424 ]---<br>
<br>
</div></div>This is a known side-effect of a dying gpu. It essentially means that the<br>
gpu refuses to wake up from deep-sleep states.<br>
<div class="im"><br>
> and then<br>
><br>
> [ 53.163526] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer<br>
> elapsed... GPU hung<br>
> [ 53.165046] [drm] capturing error event; look for more information in<br>
> /debug/dri/0/i915_error_state<br>
> [ 53.177356] [drm:i915_wait_request] *ERROR* i915_wait_request returns<br>
> -11 (awaiting 1593 at 1592, next 1594)<br>
> [ 53.181979] [drm:init_ring_common] *ERROR* render ring initialization<br>
> failed ctl 00000000 head 00000000 tail 00000000 start 00000000<br>
> [ 53.185522] [drm:init_ring_common] *ERROR* gen6 bsd ring initialization<br>
> failed ctl 00000000 head 00000000 tail 00000000 start 00000000<br>
> [ 53.188558] [drm:init_ring_common] *ERROR* blt ring initialization<br>
> failed ctl 00000000 head 00000000 tail 00000000 start 00000000<br>
> [ 55.330146] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer<br>
> elapsed... GPU hung<br>
> [ 55.332202] [drm:i915_wait_request] *ERROR* i915_wait_request returns<br>
> -11 (awaiting 1594 at 1591, next 1595)<br>
> [ 55.333258] [drm:i915_reset] *ERROR* GPU hanging too fast, declaring<br>
> wedged!<br>
> [ 55.333260] [drm:i915_reset] *ERROR* Failed to reset chip.<br>
><br>
> Of course, I'd be willing to test out stuff. I'd need a bit of guide,<br>
> however.<br>
<br>
</div>Can you please attach i915_error_state from debugfs (you need to retrigger<br>
the issue)? It contains a gpu dump which is useful to diagnose the bug.<br>
<br>
Yours, Daniel<br>
<span class="HOEnZb"><font color="#888888">--<br>
Daniel Vetter<br>
Mail: <a href="mailto:daniel@ffwll.ch">daniel@ffwll.ch</a><br>
Mobile: <a href="tel:%2B41%20%280%2979%20365%2057%2048" value="+41793655748">+41 (0)79 365 57 48</a><br>
</font></span></blockquote></div><br></div><div>I attached the error state.</div><div><br></div><div>Best,</div><div>CC</div>