Created attachment 108672 [details] dmesg ==System Environment== -------------------------- Regression: yes good commit: 3a9effecae83dcdac4b48972ec24539c6e9e4e7c bad commit: 421d82916774361a4840018fc3ab3a15751c3ea6 Non-working platforms: BYT/BDW/BSW ==kernel== -------------------------- drm-intel-nightly/bd21cf795cf5dc278f1451d0f7a597cb1d13c2ba ==Bug detailed description== Run it more than 30 minutes, it doesn't exit testing. Following cases also have this issue. igt/pm_rpm/legacy-planes igt/pm_rpm/legacy-planes-dpms igt/pm_rpm/universal-planes igt/pm_rpm/universal-planes-dpms output: IGT-Version: 1.8-g15972aa (x86_64) (Linux: 3.18.0-rc2_drm-intel-nightly_bd21cf_20141030+ x86_64) Runtime PM support: 1 PC8 residency support: 1 [ 82.894312] ------------[ cut here ]------------ [ 82.894330] WARNING: CPU: 0 PID: 4689 at drivers/gpu/drm/i915/i915_gem.c:4144 i915_gem_object_unpin_fence+0x44/0x60 [i915]() [ 82.894331] Modules linked in: dm_mod iTCO_wdt iTCO_vendor_support ppdev snd_hda_codec_hdmi pcspkr i2c_i801 lpc_ich mfd_core snd_hda_intel snd_hda_controller snd_hda_codec sn d_hwdep snd_pcm snd_timer snd soundcore battery parport_pc parport ac acpi_cpufreq i915 button video drm_kms_helper drm cfbfillrect cfbimgblt cfbcopyarea [ 82.894348] CPU: 0 PID: 4689 Comm: pm_rpm Not tainted 3.18.0-rc2_drm-intel-nightly_bd21cf_20141030+ #998 [ 82.894349] 0000000000000000 0000000000000009 ffffffff8178ef42 0000000000000000 [ 82.894352] ffffffff8103a8cc ffff8801444df240 ffffffffa0095c1a ffff0000ffff0000 [ 82.894355] ffff880000040000 ffff880002e4dc00 ffff880144bfd860 0000000000000000 [ 82.894357] Call Trace: [ 82.894363] [<ffffffff8178ef42>] ? dump_stack+0x41/0x51 [ 82.894367] [<ffffffff8103a8cc>] ? warn_slowpath_common+0x78/0x90 [ 82.894377] [<ffffffffa0095c1a>] ? i915_gem_object_unpin_fence+0x44/0x60 [i915] [ 82.894386] [<ffffffffa0095c1a>] ? i915_gem_object_unpin_fence+0x44/0x60 [i915] [ 82.894398] [<ffffffffa00c1237>] ? intel_unpin_fb_obj+0x28/0x31 [i915] [ 82.894409] [<ffffffffa00d12e9>] ? intel_disable_plane+0xad/0xff [i915] [ 82.894417] [<ffffffffa0018c75>] ? __setplane_internal+0x41/0x26f [drm] [ 82.894423] [<ffffffffa002347b>] ? drm_modeset_lock_all_crtcs+0x27/0x43 [drm] [ 82.894428] [<ffffffffa002391b>] ? __drm_modeset_lock_all+0xc8/0xf2 [drm] [ 82.894435] [<ffffffffa001b52a>] ? drm_mode_setplane+0x1aa/0x1ce [drm] [ 82.894442] [<ffffffffa0010738>] ? drm_ioctl+0x279/0x3bc [drm] [ 82.894445] [<ffffffff8133eb32>] ? sprintf+0x46/0x4b [ 82.894451] [<ffffffffa001b380>] ? drm_mode_getplane+0xd0/0xd0 [drm] [ 82.894455] [<ffffffff813e8ace>] ? rtpm_status_show+0x6a/0x78 [ 82.894458] [<ffffffff813e8a64>] ? control_show+0x37/0x37 [ 82.894460] [<ffffffff813e194b>] ? dev_attr_show+0x17/0x48 [ 82.894464] [<ffffffff8116b156>] ? kernfs_seq_start+0x24/0x7a [ 82.894467] [<ffffffff8116c355>] ? sysfs_kf_seq_show+0x79/0xc3 [ 82.894470] [<ffffffff8116b128>] ? kernfs_seq_stop+0x10/0x1a [ 82.894474] [<ffffffff81120cdb>] ? do_vfs_ioctl+0x3ec/0x435 [ 82.894477] [<ffffffff81112f97>] ? vfs_read+0x101/0x13f [ 82.894479] [<ffffffff81120d6d>] ? SyS_ioctl+0x49/0x78 [ 82.894483] [<ffffffff817945d2>] ? system_call_fastpath+0x12/0x17 [ 82.894484] ---[ end trace fffcab6ee7a3fbd7 ]--- [ 82.894494] ------------[ cut here ]------------ [ 82.894527] kernel BUG at drivers/gpu/drm/i915/i915_gem.c:4116! ==Reproduce steps== ---------------------------- 1. ./pm_rpm --run-subtest universal-planes
It also impacts HSW.
According to my bisection the first bad commit is: commit ec82cb793c9224e0692eed904f43490cf70e8258 Author: Gustavo Padovan <gustavo.padovan@collabora.co.uk> Date: Fri Oct 24 14:51:32 2014 +0100 drm/i915: create a prepare phase for sprite plane updates It would be really nice if QA could provide bisection results in their regression bug reports :)
This patch should fix it: http://patchwork.freedesktop.org/patch/35960/
(In reply to Paulo Zanoni from comment #3) > This patch should fix it: http://patchwork.freedesktop.org/patch/35960/ Fixed by this patch. output IGT-Version: 1.8-ge622850 (x86_64) (Linux: 3.18.0-rc2_kcloud_476aa3_20141103+ x86_64) Runtime PM support: 1 PC8 residency support: 0 Subtest universal-planes: SUCCESS (30.562s) real 0m32.800s user 0m0.084s sys 0m0.100s
Fix from Paulo landed in dinq: commit e4e676f9850903c160c7f5467c04342152b147d2 Author: Paulo Zanoni <paulo.r.zanoni@intel.com> Date: Mon Nov 10 14:47:30 2014 -0200 drm/i915: use the correct obj when preparing the sprite plane
Verified.Fixed. root@x-bdw05:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# time ./pm_rpm --run-subtest universal-planes IGT-Version: 1.8-gc049c39 (x86_64) (Linux: 3.18.0-rc3_drm-intel-nightly_9a7620_20141112+ x86_64) Runtime PM support: 1 PC8 residency support: 1 Subtest universal-planes: SUCCESS (22.161s) real 0m22.488s user 0m0.029s sys 0m0.049s root@x-byt06:/GFX/Test/Intel_gpu_tools/intel-gpu-tools/tests# time ./pm_rpm --run-subtest universal-planes IGT-Version: 1.8-gc049c39 (x86_64) (Linux: 3.18.0-rc3_drm-intel-nightly_9a7620_20141112+ x86_64) Runtime PM support: 1 PC8 residency support: 0 Subtest universal-planes: SUCCESS (32.951s) real 0m35.963s user 0m0.091s sys 0m0.130s
Closing old verified.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.