Created attachment 97204 [details] intel_display.c:768 dmesg System Environment: -------------------------- Platform: ILK SNB IVB kernel: (drm-intel-nightly)35dc7c8af75a3e5ae52a5469786602ce083a6368 Bug detailed description: ---------------------------- CallTrace: WARNING: CPU: 0 PID: 1335 at drivers/gpu/drm/i915/intel_display.c:768 intel_wait_for_vblank+0x159/0x17c [i915]() while running kms_flip_absolute-wf_vblank on ILK and SNB, the same issue is able to reproduced on IVB while running igt_kms_flip_bcs-wf_vblank-vs-dpms. 15ba30b2ecea353a5f0d5951f21805ba0714d889 is the first bad commit commit 15ba30b2ecea353a5f0d5951f21805ba0714d889 Author: Jesse Barnes Call Trace: ---------------------------- [ 1.276787] WARNING: CPU: 0 PID: 1335 at drivers/gpu/drm/i915/intel_display.c:768 intel_wait_for_vblank+0x159/0x17c [i915]() [ 1.276787] vblank wait timed out [ 1.276789] Modules linked in: i915(+) video button drm_kms_helper drm [ 1.276791] CPU: 0 PID: 1335 Comm: udevd Not tainted 3.14.0_drm-intel-nightly_35dc7c_20140410+ #1511 [ 1.276791] Hardware name: Dell Inc. OptiPlex 990/0DXWW6, BIOS A02 02/26/2011 [ 1.276794] 00000000 c1696876 f81fdd0c c102c7ff f81bc8a7 f4e98000 00070040 000016c0 [ 1.276795] fffb710a c102c87a 00000009 f4db5a38 f81fdd0c f4db5a50 f81bc8a7 f81fc580 [ 1.276797] 00000300 f81fdd0c f4e98000 00000001 f4e98000 f5a34000 f81c1936 f53505b4 [ 1.276797] Call Trace: [ 1.276800] [<c1696876>] ? dump_stack+0x3e/0x4e [ 1.276804] [<c102c7ff>] ? warn_slowpath_common+0x61/0x74 [ 1.276821] [<f81bc8a7>] ? intel_wait_for_vblank+0x159/0x17c [i915] [ 1.276822] [<c102c87a>] ? warn_slowpath_fmt+0x29/0x2d [ 1.276836] [<f81bc8a7>] ? intel_wait_for_vblank+0x159/0x17c [i915] [ 1.276851] [<f81c1936>] ? ironlake_crtc_enable+0x1fd/0x9ce [i915] [ 1.276865] [<f81c4121>] ? __intel_set_mode+0xf10/0xff7 [i915] [ 1.276880] [<f81c5af7>] ? intel_set_mode+0x11/0x25 [i915] [ 1.276894] [<f81c654e>] ? intel_crtc_set_config+0x646/0x8f1 [i915] [ 1.276900] [<f808d2c4>] ? drm_mode_set_config_internal+0x39/0x9b [drm] [ 1.276901] [<f80bf2f2>] ? drm_fb_helper_restore_fbdev_mode+0x7c/0x91 [drm_kms_helper] [ 1.276915] [<f81b9cd7>] ? lpt_disable_clkout_dp+0xa1/0xa1 [i915] [ 1.276916] [<f80bf60a>] ? drm_fb_helper_set_par+0x3a/0x59 [drm_kms_helper] [ 1.276918] [<c12bb1de>] ? fbcon_init+0x2c3/0x3a3 [ 1.276920] [<c12fef23>] ? visual_init+0x97/0xe9 [ 1.276922] [<c1300674>] ? do_bind_con_driver+0x16e/0x275 [ 1.276923] [<c1300887>] ? do_take_over_console+0x10c/0x134 [ 1.276924] [<c12ba9a2>] ? do_fbcon_takeover+0x49/0x88 [ 1.276926] [<c169db0a>] ? notifier_call_chain+0x20/0x42 [ 1.276927] [<c10421b0>] ? __blocking_notifier_call_chain+0x34/0x4b [ 1.276928] [<c10421d0>] ? blocking_notifier_call_chain+0x9/0xc [ 1.276930] [<c12b32fa>] ? register_framebuffer+0x1f6/0x248 [ 1.276932] [<f80bedf2>] ? drm_fb_helper_initial_config+0x355/0x3d4 [drm_kms_helper] [ 1.276951] [<f81e5a52>] ? i915_driver_load+0x950/0xbba [i915] [ 1.276953] [<c1690579>] ? klist_node_init+0x3a/0x3d [ 1.276954] [<c1690601>] ? klist_add_tail+0x16/0x30 [ 1.276956] [<c1316e47>] ? device_add+0x45d/0x467 [ 1.276957] [<c100303a>] ? show_stack_log_lvl+0x6d/0x8a [ 1.276961] [<f808935a>] ? drm_minor_register+0xf5/0x149 [drm] [ 1.276965] [<f8089410>] ? drm_dev_register+0x62/0xc3 [drm] [ 1.276969] [<f808b046>] ? drm_get_pci_dev+0xd7/0x16a [drm] [ 1.276971] [<c12a513d>] ? pci_device_probe+0x4e/0x95 [ 1.276972] [<c1318803>] ? driver_probe_device+0x79/0x168 [ 1.276973] [<c1318936>] ? __driver_attach+0x44/0x5f [ 1.276975] [<c131761f>] ? bus_for_each_dev+0x37/0x61 [ 1.276976] [<c13184d6>] ? driver_attach+0x11/0x13 [ 1.276976] [<c13188f2>] ? driver_probe_device+0x168/0x168 [ 1.276977] [<c1318213>] ? bus_add_driver+0xad/0x167 [ 1.276978] [<c1318d8a>] ? driver_register+0x6b/0x99 [ 1.276979] [<f8216000>] ? 0xf8215fff [ 1.276980] [<c10003de>] ? do_one_initcall+0x6d/0xe7 [ 1.276981] [<c169db0a>] ? notifier_call_chain+0x20/0x42 [ 1.276982] [<c10421b9>] ? __blocking_notifier_call_chain+0x3d/0x4b [ 1.276984] [<c1070be3>] ? load_module+0x140b/0x16e5 [ 1.276985] [<c169dae2>] ? __do_page_fault+0x46c/0x46c [ 1.276986] [<c1070f2f>] ? SyS_init_module+0x72/0x8a [ 1.276988] [<c169f5da>] ? sysenter_do_call+0x12/0x22 [ 1.276989] ---[ end trace 6042e7a2be61cb36 ]--- [ 1.297824] usb 1-1: new high-speed USB device number 2 using ehci-pci Reproduce steps: ---------------------------- 1. ./kms_flip --run-subtest absolute-wf_vblank
commit 15ba30b2ecea353a5f0d5951f21805ba0714d889 Author: Jesse Barnes <jbarnes@virtuousgeek.org> Date: Fri Apr 4 16:12:09 2014 -0700 drm/i915: warn when a vblank wait times out This always indicates a bug somewhere. Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Jesse promised to take care of these as they happen!
Well it annoyed me too much, so patches to fix this are on the m-l now: http://patchwork.freedesktop.org/patch/24280/ http://patchwork.freedesktop.org/patch/24281/ Please test these two (need to be applied in the right order).
Should be resolved with latest dinq: commit a5c4d7bc187bd13bc11ac06bb4ea3a0d4001aa4d Author: Ville Syrjälä <ville.syrjala@linux.intel.com> Date: Fri Mar 7 18:32:13 2014 +0200 drm/i915: Disable/enable planes as the first/last thing during modeset on ILK+ and commit be6a6f8ec707f2e446e445ad4b8cc93cc85d5d54 Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Tue Apr 15 18:41:22 2014 +0200 drm/i915: Don't vblank wait on ilk-ivb after pipe enable
Tested on latest -testing(0e078e452654cf09f6342452a3ddeebc1f6edf96) and -nightly(1e771b84e47085ef9b6efea1321e7cb5a8b2c065), this bug still can be reproduce both after boot and after run testdisplay testing. Please check dmesg in attachment for details. Thanks.
Created attachment 97792 [details] dmesg
Some other subcase below are able to reproduce this issue too plane-position-covered-pipe-B-plane-1 plane-position-covered-pipe-B-plane-2 plane-position-covered-pipe-C-plane-1 plane-position-covered-pipe-C-plane-2 plane-position-hole-pipe-B-plane-1 plane-position-hole-pipe-B-plane-2 plane-position-hole-pipe-C-plane-1 plane-position-hole-pipe-C-plane-2
Some subcases of kms_plane below are able to reproduce this issue too: plane-position-covered-pipe-B-plane-1 plane-position-covered-pipe-B-plane-2 plane-position-covered-pipe-C-plane-1 plane-position-covered-pipe-C-plane-2 plane-position-hole-pipe-B-plane-1 plane-position-hole-pipe-B-plane-2 plane-position-hole-pipe-C-plane-1 plane-position-hole-pipe-C-plane-2
Who owns the next step on this issue? We have a customer escalating the resulting timeout delay in their resume path. Updating importance accordingly.
The latest dmesg has a vblank timeout warning from dp_link_down. But that should only happen on IBX platforms, the wait is gone from all others. Does your tree have commit 93c9c19b3d259a76fc2efa4b8f2478dc9f339bee Author: Jesse Barnes <jbarnes@virtuousgeek.org> Date: Fri Apr 11 14:25:41 2014 -0700 drm/i915: remove unexplained vblank wait in the DP off code ?
Please retest with latest -nightly, it should be fixed on all platforms now.
(In reply to comment #11) > Please retest with latest -nightly, it should be fixed on all platforms now. This bug still able to reproduced on latest -nightly(2be456541ea41728002ccca2de5235f48d14326e) on ILK.
Created attachment 99053 [details] dmesg
This is a different calltrace. This bug is about intel_wait_for_vblank But the calltrace you've spotted is intel_wait_for_pipe_off That's a different bug, see https://bugs.freedesktop.org/show_bug.cgi?id=54687 The dmesg triage rules in the bug filing BKM tell you how to do this.
verified
(In reply to comment #13) > Created attachment 99053 [details] > dmesg Jinxian, please file a new bug if we find new Call trace in this dmesg.
(In reply to comment #16) > (In reply to comment #13) > > Created attachment 99053 [details] > > dmesg > Jinxian, please file a new bug if we find new Call trace in this dmesg. Done, Bug 78763 was created to track intel_wait_for_pipe_off issue.
(In reply to comment #16) > (In reply to comment #13) > > Created attachment 99053 [details] > > dmesg > Jinxian, please file a new bug if we find new Call trace in this dmesg. Erhm, I' clearly said that the new backtrace is a dupe of an existing bug. I've fixed it up.
Created attachment 101151 [details] dmesg This bug still able to reproduce on latest -nightly() on ILK while run cases below: igt/kms_flip/absolute-wf_vblank igt/kms_flip/absolute-wf_vblank-interruptible Dmesg shows: [ 141.708735] WARNING: CPU: 0 PID: 3985 at drivers/gpu/drm/i915/intel_display.c:893 intel_wait_for_vblank+0x156/0x179 [i915]() [ 141.708738] vblank wait timed out [ 141.708740] Modules linked in: dm_mod snd_hda_codec_hdmi snd_hda_codec_idt snd_hda_codec_generic iTCO_wdt iTCO_vendor_support ppdev dcdbas pcspkr firewire_ohci i2c_i801 snd_hda_intel lpc_ich snd_hda_controller mfd_core snd_hda_codec snd_hwdep firewire_core crc_itu_t snd_pcm snd_timer snd soundcore wmi battery parport_pc parport tpm_tis tpm ac acpi_cpufreq joydev i915 video button drm_kms_helper drm [ 141.708776] CPU: 0 PID: 3985 Comm: kms_flip Not tainted 3.15.0-rc8_drm-intel-nightly_b438e8_20140616+ #3594 [ 141.708778] Hardware name: Dell Inc. Latitude E6510/0JKDHD, BIOS A05 08/10/2010 [ 141.708781] 00000000 c16e77a2 f4ebbc90 c102e6c8 f858528f c24d0000 00071040 000003a2 [ 141.708788] fffd94ec c102e743 00000009 f4ebbc90 f85cb42d f4ebbca8 f858528f f85c9ae4 [ 141.708794] 0000037d f85cb42d c24d0000 f596bc80 8004400c c2429400 f859c6e2 c2487000 [ 141.708801] Call Trace: [ 141.708810] [<c16e77a2>] ? dump_stack+0x3e/0x4e [ 141.708815] [<c102e6c8>] ? warn_slowpath_common+0x65/0x78 [ 141.708843] [<f858528f>] ? intel_wait_for_vblank+0x156/0x179 [i915] [ 141.708847] [<c102e743>] ? warn_slowpath_fmt+0x29/0x2d [ 141.708866] [<f858528f>] ? intel_wait_for_vblank+0x156/0x179 [i915] [ 141.708882] [<f859c6e2>] ? intel_dp_link_down+0x167/0x169 [i915] [ 141.708897] [<f859c77c>] ? g4x_post_disable_dp+0x19/0xa5 [i915] [ 141.708912] [<f858a385>] ? ironlake_crtc_disable+0xaa/0x6f8 [i915] [ 141.708920] [<f804c33d>] ? drm_ut_debug_printk+0x2c/0x30 [drm] [ 141.708936] [<f858b78d>] ? __intel_set_mode+0x1a8/0x1024 [i915] [ 141.708953] [<f858c0dc>] ? __intel_set_mode+0xaf7/0x1024 [i915] [ 141.708956] [<c10593ff>] ? up+0x9/0x2a [ 141.708959] [<c10621f9>] ? console_unlock+0x309/0x33d [ 141.708962] [<c16e4ca8>] ? printk+0x16/0x1a [ 141.708976] [<f858e0a8>] ? intel_set_mode+0x11/0x25 [i915] [ 141.708990] [<f858eb37>] ? intel_crtc_set_config+0x62f/0x92c [i915] [ 141.708993] [<c16eb129>] ? __ww_mutex_lock+0xe/0x6d [ 141.709000] [<f8050267>] ? drm_mode_set_config_internal+0x39/0x8a [drm] [ 141.709007] [<f805047b>] ? drm_framebuffer_remove+0x70/0xd6 [drm] [ 141.709014] [<f8053528>] ? drm_mode_rmfb+0xa7/0xcd [drm] [ 141.709021] [<f8053481>] ? drm_mode_addfb2+0x27/0x27 [drm] [ 141.709026] [<f80488cf>] ? drm_ioctl+0x22e/0x35f [drm] [ 141.709033] [<f8053481>] ? drm_mode_addfb2+0x27/0x27 [drm] [ 141.709036] [<c10b2a4e>] ? release_pages+0x13d/0x153 [ 141.709039] [<c10cd1f8>] ? free_pages_and_swap_cache+0x5e/0x6c [ 141.709042] [<c10c0923>] ? tlb_flush_mmu_free+0x17/0x2b [ 141.709044] [<c10c0b0e>] ? tlb_finish_mmu+0x9/0x27 [ 141.709046] [<c10c5152>] ? unmap_region+0x99/0xa1 [ 141.709051] [<f80486a1>] ? drm_copy_field+0x47/0x47 [drm] [ 141.709056] [<c10e5431>] ? do_vfs_ioctl+0x3fa/0x444 [ 141.709062] [<f804994c>] ? drm_gem_vm_close+0x1f/0x3c [drm] [ 141.709064] [<c10c5190>] ? remove_vma+0x36/0x3b [ 141.709066] [<c10c6693>] ? do_munmap+0x1eb/0x206 [ 141.709068] [<c10e54bd>] ? SyS_ioctl+0x42/0x6d [ 141.709072] [<c16f1dac>] ? sysenter_do_call+0x12/0x22
This failure is able to reproduce on ILK on latest -fixes(2a592bec50994597716c633191ed6bf7af14defc) while running tests below: igt/kms_cursor_crc/cursor-128-onscreen igt/kms_cursor_crc/cursor-256-onscreen igt/kms_cursor_crc/cursor-64-onscreen igt/kms_cursor_crc/cursor-size-change [root@x-e6510 tests]# ./kms_cursor_crc --run-subtest cursor-size-change IGT-Version: 1.7-gf473a55 (i686) (Linux: 3.17.0-rc2_drm-intel-fixes_2a592b_20140902+ i686) Beginning cursor-size-change on pipe A, connector eDP-1 cursor-size-change on pipe A, connector eDP-1: PASSED Beginning cursor-size-change on pipe B, connector eDP-1 cursor-size-change on pipe B, connector eDP-1: PASSED Subtest cursor-size-change: SUCCESS [root@x-e6510 tests]# dmesg -r|egrep "<[1-4]>"|grep drm <4>[ 2110.919802] WARNING: CPU: 2 PID: 4067 at drivers/gpu/drm/i915/intel_display.c:902 intel_wait_for_vblank+0x156/0x179 [i915]() <4>[ 2110.919805] Modules linked in: iTCO_wdt b43 mac80211 cfg80211 iTCO_vendor_support ppdev rfkill dm_mod snd_hda_codec_hdmi snd_hda_codec_idt snd_hda_codec_generic dcdbas pcspkr i2c_i801 lpc_ich mfd_core firewire_ohci snd_hda_intel firewire_core snd_hda_controller crc_itu_t snd_hda_codec snd_hwdep snd_pcm snd_timer bcma snd soundcore wmi parport_pc parport battery tpm_tis tpm ac acpi_cpufreq joydev i915 button video drm_kms_helper drm cfbfillrect cfbimgblt cfbcopyarea <4>[ 2110.919836] CPU: 2 PID: 4067 Comm: kms_cursor_crc Tainted: G W 3.17.0-rc2_drm-intel-fixes_2a592b_20140902+ #2054 <4>[ 2110.920101] [<f805019d>] ? drm_mode_set_config_internal+0x39/0x8a [drm] <4>[ 2110.920112] [<f80503b1>] ? drm_framebuffer_remove+0x70/0xd6 [drm] <4>[ 2110.920123] [<f805367a>] ? drm_mode_rmfb+0xa7/0xcd [drm] <4>[ 2110.920133] [<f80535d3>] ? drm_mode_addfb2+0x27/0x27 [drm] <4>[ 2110.920141] [<f8049dba>] ? drm_ioctl+0x233/0x35c [drm] <4>[ 2110.920152] [<f80535d3>] ? drm_mode_addfb2+0x27/0x27 [drm] <4>[ 2110.920172] [<f8049b87>] ? drm_getstats+0xe/0xe [drm]
Presumed fixed commit 08aff3fe26ae7a0d6f302ac2e1b7e2eb9933cd42 Author: Ville Syrjälä <ville.syrjala@linux.intel.com> Date: Mon Aug 18 22:16:09 2014 +0300 drm/i915: Move DP port disable to post_disable for pch platforms At least we don't have a g4x_post_disable_dp any more, so defintely need a new backtrace if this is still an issue.
Can you state what kernel release(s) this is in? Thanks.
The failure unable to reproduce on latest -nightly(3b6da75500fb10e59dfb153eba9845c05f8d82d6) [root@x-ivb9 tests]# ./kms_flip --run-subtest absolute-wf_vblank IGT-Version: 1.8-gaa63fc7 (x86_64) (Linux: 3.18.0-rc4_drm-intel-nightly_3b6da7_20141116_debug+ x86_64) Using monotonic timestamps Beginning absolute-wf_vblank on crtc 8, connector 18 1024x768 60 1024 1048 1184 1344 768 771 777 806 0xa 0x40 65000 ............................................................ absolute-wf_vblank on crtc 8, connector 18: PASSED Beginning absolute-wf_vblank on crtc 12, connector 18 1024x768 60 1024 1048 1184 1344 768 771 777 806 0xa 0x40 65000 ............................................................ absolute-wf_vblank on crtc 12, connector 18: PASSED Beginning absolute-wf_vblank on crtc 16, connector 18 1024x768 60 1024 1048 1184 1344 768 771 777 806 0xa 0x40 65000 ............................................................ absolute-wf_vblank on crtc 16, connector 18: PASSED Subtest absolute-wf_vblank: SUCCESS (30.382s)
Closing (>2 years) old Verified+fixed.
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.