Bug 98553

Summary: [IGT][IVB/BXT] performance failure when running kms_setmode basic
Product: DRI Reporter: cprigent <christophe.prigent>
Component: DRM/IntelAssignee: Joseph Garvey <joseph1.garvey>
Status: CLOSED DUPLICATE QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: normal    
Priority: medium CC: intel-gfx-bugs
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard: ReadyForDev
i915 platform: BXT, IVB i915 features: display/Other
Attachments:
Description Flags
IVB__kms_setmode__basic__output
none
IVB__kms_setmode__basic__kern.log
none
HSW__kms_setmode__basic__kern.log
none
HSW__kms_setmode__basic__output none

Description cprigent 2016-11-02 15:15:55 UTC
Created attachment 127691 [details]
IVB__kms_setmode__basic__output

Platform: IVB
CPU: Intel(R) Core(TM) i7-3770K CPU @ 3.50GHz (family 6, model 58, stepping 9)
Motherboard version: DH77EB
GPU: Intel® HD Graphics 4000 - Xeon E3-1200 v2/3rd Gen Core processor Graphics Controller
Memory: 4 x DDR3 8GB 1333

Software
Bios: EBH7710H.86A.0096.2012.1012.1645
Linux distribution: Ubuntu 16.04 64 bits
Kernel: 4.9.0-rc2 6a1197b from http://cgit.freedesktop.org/drm-intel/
  commit 6a1197bcb5cc18a56ad4ae8e6d706a212bc3db7d
  Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
  Date:   Mon Oct 31 16:58:45 2016 +0200
  drm-intel-nightly: 2016y-10m-31d-14h-58m-16s UTC integration manifest
libdrm-2.4.71-8 fe4579e from git://anongit.freedesktop.org/mesa/drm
mesa: mesa-12.0.0 8b06176 from git://anongit.freedesktop.org/mesa/mesa
cairo 1.15.2 db8a7f1 from git://anongit.freedesktop.org/cairo
xorg-server-1.18.99.902 512e75a from git://git.freedesktop.org/git/xorg/xserver
xf86-video-intel 2.99.917-722 714052f from git://git.freedesktop.org/git/xorg/driver/xf86-video-intel
libva-1.7.2-39 5c47c33 from git://git.freedesktop.org/git/vaapi/libva 
vaapi-intel-driver: 1.7.2-153 77ff763 from git://git.freedesktop.org/git/vaapi/intel-driver
IGT: intel-gpu-tools-1.16-105 5bfbbea from http://anongit.freedesktop.org/git/xorg/app/intel-gpu-tools.git

External screen: DELL U2311Hb

Steps:
------
1. Execute IGT test: 
# ./kms_setmode --r basic

Actual result:
---------------
1. Test is fail

Expected result:
----------------
1. Test is Pass
Comment 1 cprigent 2016-11-02 15:33:55 UTC
Created attachment 127692 [details]
IVB__kms_setmode__basic__kern.log
Comment 2 cprigent 2016-11-02 15:38:27 UTC
This is a regression, test was pass with:
Kernel: 4.9.0-rc1 17dc529 from http://cgit.freedesktop.org/drm-intel/
  commit 17dc529acb9a6a4328b419048e32df586b90646b
  Author: Chris Wilson <chris@chris-wilson.co.uk>
  Date: Mon Oct 24 14:24:26 2016 +0100
  drm-intel-nightly: 2016y-10m-24d-13h-22m-52s UTC integration manifest
libdrm-2.4.71 9e24d0c from git://anongit.freedesktop.org/mesa/drm
mesa: mesa-12.0.0 8b06176 from git://anongit.freedesktop.org/mesa/mesa
cairo 1.15.2 db8a7f1 from git://anongit.freedesktop.org/cairo
xorg-server-1.18.99.901-80 5dcb066 from git://git.freedesktop.org/git/xorg/xserver
xf86-video-intel 2.99.917-720 388fd4a from git://git.freedesktop.org/git/xorg/driver/xf86-video-intel
libva-1.7.2-38 3b7e499 from git://git.freedesktop.org/git/vaapi/libva
vaapi-intel-driver: 1.7.2-140 852cea1 from git://git.freedesktop.org/git/vaapi/intel-driver
IGT: intel-gpu-tools-1.16-96 93437cb from http://anongit.freedesktop.org/git/xorg/app/intel-gpu-tools.git
Comment 3 cprigent 2016-11-02 16:14:02 UTC
Sa
Comment 4 cprigent 2016-11-02 16:17:09 UTC
Created attachment 127693 [details]
HSW__kms_setmode__basic__kern.log

Same on HSW:
Platform HSW: Dell Inc. OptiPlex 9020
CPU: Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz (Family 6, Model 60, Stepping 3)
Motherboard: 0DNKMN
GPU: Intel® HD Graphics 4600 / Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller

Software
Bios: A07
Linux distribution: Ubuntu 16.04 64 bits
External screen: DELL E1912Hf
Comment 5 cprigent 2016-11-02 16:17:25 UTC
Created attachment 127694 [details]
HSW__kms_setmode__basic__output
Comment 6 cprigent 2016-11-02 16:18:17 UTC
Associated bug: bug 98547
Comment 7 Chris Wilson 2016-11-15 09:03:49 UTC
It's a new test. It may simply be asking for much precision. The failure is an error of +-20us, which is greater than a scanline and our accuracy goal is roughly half a scanline.
Comment 8 cprigent 2016-11-17 15:22:05 UTC
Yes, this is a new test, not a regression.
Comment 9 cprigent 2016-11-17 16:13:05 UTC
I'm reproducing it on IVB.
On HSW I now reproduce bug 98547.

Tested with:
Kernel: 4.9.0-rc4 91e164f branch drm-intel-nightly from http://cgit.freedesktop.org/drm-intel/ 
  commit 91e164fea17d3e5366048b6eae3c6eea4e14e9fe
  Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
  Date:   Mon Nov 14 16:31:06 2016 +0200
  drm-intel-nightly: 2016y-11m-14d-14h-30m-30s UTC integration manifest

libdrm-2.4.71-13 670f1e4 from git://anongit.freedesktop.org/mesa/drm
mesa: mesa-13.0.0 df1b0a5 from git://anongit.freedesktop.org/mesa/mesa
cairo 1.15.2 db8a7f1 from git://anongit.freedesktop.org/cairo
xorg-server-1.18.99.902-2 7513da4 from git://git.freedesktop.org/git/xorg/xserver
xf86-video-intel 2.99.917-731 d1d14f2 from git://git.freedesktop.org/git/xorg/driver/xf86-video-intel
libva-1.7.2-40 3a7547b from git://git.freedesktop.org/git/vaapi/libva 
vaapi-intel-driver: 1.7.2-157 55a538c from git://git.freedesktop.org/git/vaapi/intel-driver
intel-gpu-tools-1.16-132 773ac7c from http://anongit.freedesktop.org/git/xorg/app/intel-gpu-tools.git
Comment 10 Jari Tahvanainen 2017-03-29 08:24:11 UTC
Still valid for 4.11.0-rc4-drm-tip-qa-ww13-commit-5c7479a+ (IVB-1-0XR1GT)

./kms_setmode --run-subtest basic
IGT-Version: 1.18-ga27c743 (x86_64) (Linux: 4.11.0-rc4-drm-tip-qa-ww13-commit-5c7479a+ x86_64)
Testing: basic 1 connector combinations
  Test id#1 CRTC count 1
    CRTC[32] [Pipe A] Mode: 1920x1200@60Hz Connectors: VGA-1[47]
Expected frametime: 16681us; measured 16626.4us +- 2.753us accuracy 0.05% [0.61 scanlines]
Stack trace:
  #0 [__igt_fail_assert+0x101]
  #1 [test_combinations+0x10b8]
  #2 [main+0x1f5]
  #3 [__libc_start_main+0xf1]
  #4 [_start+0x29]
  #5 [<unknown>+0x29]
Subtest basic: FAIL (2.170s)

(kms_setmode:17155) CRITICAL: Test assertion failure function check_timings, file kms_setmode.c:521:
(kms_setmode:17155) CRITICAL: Failed assertion: fabs(mean - expected) < 1.718 * stddev
(kms_setmode:17155) CRITICAL: vblank interval differs from modeline! expected 16680.5us, measured 16626us +- 2.753us, difference 54.1us (19.6 sigma)
Subtest basic failed.
**** DEBUG ****
(kms_setmode:17155) INFO: Testing: basic 1 connector combinations
(kms_setmode:17155) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1200, format=0x34325258, tiling=0x0, size=0)
(kms_setmode:17155) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=7680)
(kms_setmode:17155) INFO:   Test id#1 CRTC count 1
(kms_setmode:17155) INFO:     CRTC[32] [Pipe A] Mode: 1920x1200@60Hz Connectors: VGA-1[47]
(kms_setmode:17155) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1200, format=0x34325258, tiling=0x0, size=0)
(kms_setmode:17155) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=2, pitch=7680)
(kms_setmode:17155) INFO: Expected frametime: 16681us; measured 16626.4us +- 2.753us accuracy 0.05% [0.61 scanlines]
(kms_setmode:17155) CRITICAL: Test assertion failure function check_timings, file kms_setmode.c:521:
(kms_setmode:17155) CRITICAL: Failed assertion: fabs(mean - expected) < 1.718 * stddev
(kms_setmode:17155) CRITICAL: vblank interval differs from modeline! expected 16680.5us, measured 16626us +- 2.753us, difference 54.1us (19.6 sigma)
****  END  ****
Comment 11 Ricardo Madrigal 2017-07-20 18:57:34 UTC
The following tests fail on BYT with latest configuration

====================================================
Test list
====================================================
igt@kms_setmode@basic

====================================================
Graphic Stack
====================================================
Component: drm
     tag: libdrm-2.4.81-31-g23e234a
     commit: 23e234a3503f51b9d9c585123d33b936f522808d
Component: cairo
    tag: 1.15.6-2-g57b4050
    commit: 57b40507dda3f58dfc8635548d606b86dc7bcf51
Component: intel-gpu-tools
    tag: intel-gpu-tools-1.19-96-gfb1ddc4
    commit: fb1ddc47003ad6a683db79beeb81b6cbab1feb7c
Component: piglit
    tag: piglit-v1
    commit: 56e7e5583cd4a3ca15a8cda154d46d168959dd25

======================================
             Hardware
======================================
motherboard model          : .................................
motherboard id             : DN2820FYK
form factor                : Desktop
manufacturer               : .................................
cpu family                 : Celeron
cpu family id              : 6
cpu information            : Intel(R) Celeron(R) CPU  N2830  @ 2.16GHz
gpu card                   : Intel Corporation Atom Processor Z36xxx/Z37xxx Series Graphics & Display (rev 0e) (prog-if 00 [VGA controller])
memory ram                 : 7.66 GB
max memory ram             : 8 GB
cpu thread                 : 2
cpu core                   : 2
cpu model                  : 55
cpu stepping               : 8
socket                     : <OUT OF SPEC>
signature                  : Type 0, Family 6, Model 55, Stepping 8
hard drive                 : 111GiB (120GB)
current cd clock frequency : 266667 kHz
maximum cd clock frequency : 400000 kHz
displays connected         : HDMI-A-1

=============================
IGT-Version: 1.19-gfb1ddc4 (x86_64) (Linux: 4.13.0-rc1-drm-tip-ww29-commit-f1c32d6+ x86_64)
(kms_setmode:31795) igt-core-DEBUG: Test requirement passed: !igt_run_in_simulation()
(kms_setmode:31795) drmtest-DEBUG: Test requirement passed: !(fd<0)
(kms_setmode:31795) igt-debugfs-DEBUG: Opening debugfs directory '/sys/kernel/debug/dri/0'
(kms_setmode:31795) drmtest-DEBUG: Test requirement passed: drmSetMaster(fd) == 0
(kms_setmode:31795) igt-kms-DEBUG: VT: graphics mode set (mode was 0x0)
(kms_setmode:31795) igt-core-DEBUG: Starting subtest: basic
Testing: basic 1 connector combinations
(kms_setmode:31795) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1080, format=0x34325258, tiling=0x0, size=0)
(kms_setmode:31795) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=1, pitch=7680)
  Test id#1 CRTC count 1
    CRTC[34] [Pipe A] Mode: 1920x1080@60Hz Connectors: HDMI-A-1[49]
(kms_setmode:31795) igt-fb-DEBUG: igt_create_fb_with_bo_size(width=1920, height=1080, format=0x34325258, tiling=0x0, size=0)
(kms_setmode:31795) igt-fb-DEBUG: igt_create_fb_with_bo_size(handle=2, pitch=7680)
Expected frametime: 16667us; measured 16673.3us +- 1.615us accuracy 0.03% [0.33 scanlines]
Stack trace:
  #0 [__igt_fail_assert+0x101]
  #1 [test_one_combination+0xd7f]
  #2 [main+0x4c6]
  #3 [__libc_start_main+0xf1]
  #4 [_start+0x29]
  #5 [<unknown>+0x29]
Subtest basic: FAIL (2.564s)
(kms_setmode:31795) igt-core-DEBUG: Exiting with status code 99
(kms_setmode:31795) igt-kms-DEBUG: VT: original mode 0x0 restored
(kms_setmode:31795) igt-debugfs-DEBUG: Opening debugfs directory '/sys/kernel/debug/dri/0'
=======================
Comment 12 Ricardo Madrigal 2017-07-24 19:07:59 UTC
The following tests fail on BXT with latest configuration

====================================================
Test list
====================================================
igt@kms_setmode@basic

====================================================
Graphic Stack
====================================================
Component: drm
    tag: libdrm-2.4.81-27-gac21401
    commit: ac214017904b31bc5f80f802d748d5f4f3149d22

Component: cairo
    tag: 1.15.6-2-g57b4050
    commit: 57b40507dda3f58dfc8635548d606b86dc7bcf51

Component: intel-gpu-tools
    tag: intel-gpu-tools-1.19-88-ga844ccb
    commit: a844ccbdbab9fd16c37de81281c6281bc800e97a

Component: piglit
    tag: piglit-v1
    commit: 8717219b07c0cf2853540c373725a23654467d45


======================================
             Hardware
======================================
motherboard model          : NUC6CAYS
motherboard id             : NUC6CAYB
form factor                : Desktop
manufacturer               : Intelcorporation
cpu family                 : Celeron
cpu family id              : 6
cpu information            : Intel(R) Celeron(R) CPU J3455 @ 1.50GHz
gpu card                   : Intel Corporation Device 5a85 (rev 0b) (prog-if 00 [VGA controller])
memory ram                 : 15.54 GB
max memory ram             : 32 GB
cpu thread                 : 4
cpu core                   : 4
cpu model                  : 92
cpu stepping               : 9
socket                     : Other
signature                  : Type 0, Family 6, Model 92, Stepping 9
hard drive                 : 279GiB (300GB)
current cd clock frequency : 288000 kHz
maximum cd clock frequency : 624000 kHz
displays connected         : DP-1 DP-2
Comment 13 Elizabeth 2017-07-25 17:21:03 UTC
Adding tag into "Whiteboard" field - ReadyForDev
*Status is correct
*Platform is included
*Feature is included
*Priority and Severity correctly set
*Logs included
Comment 14 Elizabeth 2017-08-24 15:03:37 UTC
Still failed on BXT with the following configuration:
Gfx stack:
Component: drm
    tag: libdrm-2.4.81-46-g9d133dd
    commit: 9d133dd08720d80dfc8ce098bf0972ebd05c454d
Component: cairo
    tag: 1.15.6-14-gf351595
    commit: f3515954e0bc9c9ca413a84a8dd842145cec83a7
Component: intel-gpu-tools
    tag: intel-gpu-tools-1.19-193-g5a17ee2
    commit: 5a17ee2c8f9013f5db852d27564b837f9f2c5a9f
Component: piglit
    tag: piglit-v1
    commit: 2a7047a32bb3a9a4fad5bea0a8185d63af6bddf3
Kernel:
commit ad6ab9f96437f0cb1f4d8a7840fd9eeb894eb12e
Author:     Gustavo Padovan <gustavo.padovan@collabora.com>
AuthorDate: Fri Aug 18 15:26:42 2017 -0300
Commit:     Gustavo Padovan <gustavo.padovan@collabora.com>
CommitDate: Fri Aug 18 15:26:42 2017 -0300
    drm-tip: 2017y-08m-18d-18h-21m-46s UTC integration manifest
Comment 15 Joseph Garvey 2017-08-25 17:42:11 UTC
(In reply to Chris Wilson from comment #7)
> It's a new test. It may simply be asking for much precision. The failure is
> an error of +-20us, which is greater than a scanline and our accuracy goal
> is roughly half a scanline.

The first assertion, that "99.7% samples within one scanline on each side of mean", is also failing (on BDW and HSW from what I've tried). If you relax it to 95%, the test passes - but that's a pretty low MTBF. What makes 5.6s the lowest acceptable MTBF? Also, how much does vblank accuracy depend on the hardware and should the test take that into consideration?
Comment 16 Rodrigo Vivi 2018-01-09 00:32:53 UTC

*** This bug has been marked as a duplicate of bug 99912 ***

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.