Skip to content
Snippets Groups Projects
New look: Off

[GLK] all tests - dmesg-warn / dmesg-fail/fail - *ERROR* CPU pipe [AB] FIFO underrun

    • Open Issue created by Bugzilla Migration User @bugzilla-migration

    Submitted by Tomi Sarvela @tsa

    Assigned to Intel GFX Bugs mailing list

    Link to original bug (#106538)

    Description

    FIFO underruns made an appearance on Gemini Lake NUCs. Examples of tests it hits:

    https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4182/shard-glk4/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html

    https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4182/shard-glk4/igt@kms_cursor_legacy@flip-vs-cursor-crc-atomic.html

    https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4182/shard-glk4/igt@kms_busy@extended-pageflip-modeset-hang-oldfb-render-b.html

    This might be residual issues from 4.17-rc5 pull, and happened only on one run on host, so we'll follow if it happens again.

    Blocking

    Edited
    To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information

    Child items ...

    • Activity

    • Bugzilla Migration User added CI feature: display/atomic platform: GLK priority::medium severity::normal + 1 deleted label

      added CI feature: display/atomic platform: GLK priority::medium severity::normal + 1 deleted label

    • Bugzilla Migration User
      Bugzilla Migration User @bugzilla-migration ·
      Author

      :speech_balloon: Tomi Sarvela @tsa said:

      Happened again on another same kind of NUC:

      https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4187/shard-glk8/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html

      https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4187/shard-glk8/igt@kms_concurrent@pipe-a.html

    • Bugzilla Migration User
      Bugzilla Migration User @bugzilla-migration ·
      Author

      :speech_balloon: Martin Peres @mupuf said:

      https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_1821/shard-glk5/igt@kms_rotation_crc@sprite-rotation-270.html

      <3> [58.669164] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] ERROR CPU pipe B FIFO underrun
      <3> [58.852204] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] ERROR CPU pipe A FIFO underrun

    • Bugzilla Migration User
      Bugzilla Migration User @bugzilla-migration ·
      Author

      :speech_balloon: Vít Ondruch said:

      I observer similar error. Actually my computer typically freezes due to this :/ It typically happens when my computer is a sleep and some gnome nofitication appears on the lock screen. I also suspect it has something to do with my external monitory, which has from time to time hard time to wake up. Also, I suspect it started to happen with Kernel 4.17. 4.16 was almost rock stable for me.

      I am running Lenovo t470s

      $ uname -a<br>
Linux localhost.localdomain 4.18.5-300.fc29.x86_64 #1 SMP Fri Aug 24 17:16:35 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux<br>
<br>
$ rpm -q kernel<br>
kernel-4.17.6-200.fc28.x86_64<br>
kernel-4.17.14-200.fc28.x86_64<br>
kernel-4.18.5-300.fc29.x86_64<br>
    • Bugzilla Migration User
      Bugzilla Migration User @bugzilla-migration ·
      Author

      :paperclip: Behrang Saeedzadeh uploaded an attachment:

      I think this issue has been affecting me for a few years now but i just recently started to look into it.

      my current system is ubuntu 19.04 but this issue was present even when i was using 18.04 and 18.10 and even before that.

      os: ubuntu 19.04
      cpu: intel core i7 6700
      graphics (integrated only): hd graphics 530 (skylake gt2)

      anyway, from time to time only one of my monitors turn on. when that happens dmesg | grep i915 prints this:

      ➜  ~ dmesg -e | grep i915<br>
[  +0.000545] i915 0000:00:02.0: vgaarb: changed vga decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem<br>
[  +0.000627] [drm] finished loading dmc firmware i915/skl_dmc_ver1_27.bin (v1.27)<br>
[  +0.003052] [drm] initialized i915 1.6.0 20181204 for 0000:00:02.0 on minor 0<br>
[  +0.000036] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])<br>
[  +0.000047] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device<br>
[  +0.249846] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *error* cpu pipe a fifo underrun<br>
<br>
➜  ~ dmesg -t | grep i915<br>
[sat apr 27 17:47:08 2019] i915 0000:00:02.0: vgaarb: changed vga decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem<br>
[sat apr 27 17:47:08 2019] [drm] finished loading dmc firmware i915/skl_dmc_ver1_27.bin (v1.27)<br>
[sat apr 27 17:47:09 2019] [drm] initialized i915 1.6.0 20181204 for 0000:00:02.0 on minor 0<br>
[sat apr 27 17:47:09 2019] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])<br>
[sat apr 27 17:47:09 2019] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device<br>
[sat apr 27 17:47:09 2019] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *error* cpu pipe a fifo underrun<br>
```<br>
<br>
when i switch to wayland, sometimes none of my monitors turn on at all. once when both turned on, i turned off one<br>
and tried to turn it back on again, but it went into energy saving mode. then `dmesg` was showing this:<br>
<br>
<br>
```<br>
➜  ~ dmesg -t | grep i915<br>
[sat apr 27 18:14:17 2019] i915 0000:00:02.0: vgaarb: changed vga decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem<br>
[sat apr 27 18:14:17 2019] [drm] finished loading dmc firmware i915/skl_dmc_ver1_27.bin (v1.27)<br>
[sat apr 27 18:14:18 2019] [drm] initialized i915 1.6.0 20181204 for 0000:00:02.0 on minor 0<br>
[sat apr 27 18:14:18 2019] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])<br>
[sat apr 27 18:14:18 2019] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device<br>
[sat apr 27 18:14:56 2019] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *error* cpu pipe b fifo underrun<br>
<br>
➜  ~ dmesg -e | grep i915<br>
[  +0.000664] i915 0000:00:02.0: vgaarb: changed vga decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem<br>
[  +0.000861] [drm] finished loading dmc firmware i915/skl_dmc_ver1_27.bin (v1.27)<br>
[  +0.016298] [drm] initialized i915 1.6.0 20181204 for 0000:00:02.0 on minor 0<br>
[  +0.000047] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])<br>
[  +0.000047] i915 0000:00:02.0: fb0: inteldrmfb frame buffer device<br>
[ +23.689059] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *error* cpu pipe b fifo underrun<br>
<br>

      Attachment 144100, "dmesg":
      dmesg.txt

    • CI Bug Log
      CI Bug Log @cibuglog ·
      Reporter

      A CI Bug Log filter associated to this bug has been updated by Lakshmi Vudum:

      Description: GLK: all tests - dmesg-warn / dmesg-fail - *ERROR* CPU pipe [AB] FIFO underrun

      Equivalent query: runconfig_tag IS IN ["DRM-TIP"] AND (machine_name IS IN ["shard-glk8", "shard-glk6", "shard-glk7", "fi-glk-1", "fi-glk-dsi", "fi-glk-j4005", "shard-glk9", "shard-glk", "shard-glk1", "shard-glk3", "shard-glk4", "shard-glk5", "shard-glk2"] OR machine_tag IS IN ["GLK"]) AND ((testsuite_name = "IGT" AND status_name IS IN ["fail", "dmesg-warn", "dmesg-fail"])) AND dmesg ~= '\*ERROR\* CPU pipe [AB] FIFO underrun'

      No new failures caught with the new filter

    • CI Bug Log
      CI Bug Log @cibuglog ·
      Reporter

      A CI Bug Log filter associated to this bug has been updated by Lakshmi Vudum:

      Description: GLK: all tests - dmesg-warn / dmesg-fail/fail - *ERROR* CPU pipe [AB] FIFO underrun

      Equivalent query: runconfig_tag IS IN ["DRM-TIP"] AND (machine_name IS IN ["shard-glk8", "shard-glk6", "shard-glk7", "fi-glk-1", "fi-glk-dsi", "fi-glk-j4005", "shard-glk9", "shard-glk", "shard-glk1", "shard-glk3", "shard-glk4", "shard-glk5", "shard-glk2"] OR machine_tag IS IN ["GLK"]) AND ((testsuite_name = "IGT" AND status_name IS IN ["fail", "dmesg-warn", "dmesg-fail"])) AND dmesg ~= '\*ERROR\* CPU pipe [AB] FIFO underrun'

      No new failures caught with the new filter

    • Yann Soubeyrand
      Yann Soubeyrand @yann-soubeyrand ·

      I'm also affected: I'm using a T480 running Fedora 32 (kernel 5.8.4-200.fc32.x86_64) and my external monitors are regularly flickering.

    • SAI NANDAN
      SAI NANDAN @ysainan ·
      Reporter

      Hi Reporter, please confirm if this issue is still seen on latest kernel also.

      If yes, provide the steps to reproduce it for further debugging.

    • Tejasree Illipilli
      Tejasree Illipilli @tejasree ·
      Reporter

      Issue is still reproducing on GLK with CI_DRM_9488 and IGT_5901 in shard runs.

      https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_9488/shard-glk7/igt@gem_exec_whisper@basic-queues-all.html

    • SAI NANDAN
      SAI NANDAN @ysainan ·
      Reporter

      Issue not reproducing on GLK Test case has passed with CI_DRM_9532 , IGT_5928.

      https://intel-gfx-ci.01.org/tree/drm-tip/IGT_5928/shard-glk8/igt@gem_exec_whisper@basic-queues-all.html

      Hence closing the ticket.

    • SAI NANDAN closed

      closed

    • SAI NANDAN
      SAI NANDAN @ysainan ·
      Reporter

      Issue is reproducing on GLK. Issue is observed consistently.

      Hence reopening #118

      Edited by SAI NANDAN
    • SAI NANDAN reopened

      reopened

    • SAI NANDAN assigned to @tejaskux

      assigned to @tejaskux

    • LAKSHMINARAYANA VUDUM changed title from [CI][GLK] drm:intel_cpu_fifo_underrun_irq_handler [i915] ERROR CPU pipe B FIFO underrun to [GLK] all tests - dmesg-warn / dmesg-fail/fail - ERROR CPU pipe [AB] FIFO underrun

      changed title from [CI][GLK] drm:intel_cpu_fifo_underrun_irq_handler [i915] ERROR CPU pipe B FIFO underrun to [GLK] all tests - dmesg-warn / dmesg-fail/fail - ERROR CPU pipe [AB] FIFO underrun

    • CI Bug Log
      CI Bug Log @cibuglog ·
      Reporter

      A CI Bug Log filter associated to this bug has been updated by Lakshmi Vudum:

      Description: SKL GLK: all tests - dmesg-warn / dmesg-fail/fail - *ERROR* CPU pipe [ABCD] FIFO underrun

      Equivalent query: runconfig_tag IS IN ["DRM-TIP"] AND (machine_name IS IN ["shard-skl", "shard-skl1", "fi-glk-1", "shard-skl2", "shard-skl3", "shard-skl4", "shard-apl6", "fi-skl-6770hq", "shard-skl5", "shard-skl6", "shard-skl7", "fi-skl-6260u", "fi-apl-guc", "shard-apl2", "shard-skl10", "shard-skl8", "shard-skl9", "fi-skl-lmem", "fi-skl-6700hq", "fi-skl-6600u", "shard-apl5", "shard-apl7", "shard-apl8", "shard-apl1", "fi-glk-dsi", "shard-apl4", "shard-apl3", "fi-skl-6700k2", "fi-skl-gvtdvm", "fi-skl-guc", "fi-glk-j4005", "shard-apl", "shard-glk", "shard-glk1", "shard-glk3", "shard-glk4", "shard-glk5", "shard-glk2", "shard-glk8", "shard-glk6", "shard-glk7", "shard-glk9", "fi-skl-iommu", "fi-skl-caroline", "fi-apl-nasher"] OR machine_tag IS IN ["GLK", "APL", "SKL"]) AND ((testsuite_name = "IGT" AND status_name IS IN ["fail", "dmesg-warn", "dmesg-fail"])) AND dmesg ~= '\*ERROR\* CPU pipe [ABCD] FIFO underrun'

      No new failures caught with the new filter

    • LAKSHMINARAYANA VUDUM added platform: APL platform: SKL labels

      added platform: APL platform: SKL labels

    • CI Bug Log
      CI Bug Log @cibuglog ·
      Reporter

      A CI Bug Log filter associated to this bug has been updated by Lakshmi Vudum:

      Description: APL SKL GLK: all tests - dmesg-warn / dmesg-fail/fail - *ERROR* CPU pipe [ABCD] FIFO underrun

      Equivalent query: runconfig_tag IS IN ["DRM-TIP"] AND (machine_name IS IN ["shard-skl", "shard-skl1", "fi-glk-1", "shard-skl2", "shard-skl3", "shard-skl4", "shard-apl6", "fi-skl-6770hq", "shard-skl5", "shard-skl6", "shard-skl7", "fi-skl-6260u", "fi-apl-guc", "shard-apl2", "shard-skl10", "shard-skl8", "shard-skl9", "fi-skl-lmem", "fi-skl-6700hq", "fi-skl-6600u", "shard-apl5", "shard-apl7", "shard-apl8", "shard-apl1", "fi-glk-dsi", "shard-apl4", "shard-apl3", "fi-skl-6700k2", "fi-skl-gvtdvm", "fi-skl-guc", "fi-glk-j4005", "shard-apl", "shard-glk", "shard-glk1", "shard-glk3", "shard-glk4", "shard-glk5", "shard-glk2", "shard-glk8", "shard-glk6", "shard-glk7", "shard-glk9", "fi-skl-iommu", "fi-skl-caroline", "fi-apl-nasher"] OR machine_tag IS IN ["GLK", "APL", "SKL"]) AND ((testsuite_name = "IGT" AND status_name IS IN ["fail", "dmesg-warn", "dmesg-fail"])) AND dmesg ~= '\*ERROR\* CPU pipe [ABCD] FIFO underrun'

      No new failures caught with the new filter

    • Ville Syrjälä
      Ville Syrjälä @vsyrjala ·
      Developer

      Please stop piling all underruns across different platforms into the same bug.

    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    • Loading
    Please register or sign in to reply