Created attachment 126273 [details] log-synmark2 Platform: KABY LAKE-U Processor : Genuine Intel(R) CPU 0000 @ 1.80GHz (cpu family: 6, model: 142, stepping: 9) MCP : KBL-U J0 2+3e QDF : QL9J PCH: PCH-LP C1 CRB : KABY LAKE U DDR3L RVP7 Rework: O-16 Software BIOS: 45.1 3KBLSE2R1.R00.X045.P01.1606291634 from https://ubit-artifactory-ba.intel.com/artifactory/owr-repos/Submissions/ifwi/KBL_ORANGE_IFWI_2016_WW27_3_03_SR'17/ ME FW: 11.6.0.1065 EC FW: 1.24 KSC: 1.24 Linux distribution: Ubuntu 16.04 64 bits Kernel: 4.8.0-rc4 9baa666 from http://cgit.freedesktop.org/drm-intel/ commit 9baa666b3e48f71b46c5f63541f57d2a95a1b1c0 Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Sat Sep 3 13:12:38 2016 +0100 drm-intel-nightly: 2016y-09m-03d-12h-12m-15s UTC integration manifest libdrm-2.4.70-6 4462303 from git://anongit.freedesktop.org/mesa/drm mesa: mesa-12 from git://anongit.freedesktop.org/mesa/mesa cairo 1.15.2 db8a7f1 from git://anongit.freedesktop.org/cairo xorg-server-1.18.0-546 deae9c7 from git://git.freedesktop.org/git/xorg/xserver 2.99.917-701 205146b from git://git.freedesktop.org/git/xorg/driver/xf86-video-intel libva-1.7.0-47 2ebf897 from git://git.freedesktop.org/git/vaapi/libva vaapi-intel-driver: 1.7.0-95 1817bee from git://git.freedesktop.org/git/vaapi/intel-driver DMC 1.01 from https://01.org/sites/default/files/downloads/intelr-graphics-linux/kbldmcver101.tar.bz2 GuC 9.14 from http://rdvivi-hillsboro.jf.intel.com/firmware/kbl_guc_ver9_14.tar.bz2 Synmark2 from http://benchsrv.fi.intel.com/archive/benchmarks/ Precondition: -------------- Boot with kernel boot command lines: i915.enable_guc_loading=2 i915.enable_guc_submission=2 Steps: ------ 1. Download Synmark2: mkdir benchmark cd benchmark wget http://benchsrv.fi.intel.com/archive/benchmarks/SynMark2.tar.gz tar -xvf SynMark2.tar.gz 2. From, Terminal 1: startx 3. From Terminal 2: cd SynMark2 sudo -E ./synmark-test.sh Actual result: -------------- 3. Green square is displayed a couple of seconds (no rotation) then tool is ended and returns errors: intel_do_flush_locked failed: Input/output error libGL: OpenDriver: trying /opt/X11R7/lib/dri/tls/i965_dri.so libGL: OpenDriver: trying /opt/X11R7/lib/dri/i965_dri.so Expected result: ---------------- 3. Benchmark is launched without problem when GuC is loaded Info: ----- Benchmark is launched without problem after removing i915.enable_guc_loading=2 i915.enable_guc_submission=2 from Grub
GPU hang, let's see the error state and start pondering what more we need to see in it for debugging GuC.
Created attachment 126275 [details] kern.log
Created attachment 126293 [details] GPU crash dump
From error dump, hung is happening in render ring batch with active head at 0xff2057e4, with 0x7b000005 (3DPRIMITIVE) as IPEHR. Batch extract (around 0xff2057e4): 0xff2057c0: 0x780c0000: 3D UNKNOWN: 3d_965 opcode = 0x780c 0xff2057c4: 0x00000000: MI_NOOP Bad length 7 in (null), expected 6-6 0xff2053c8: 0x7b000005: 3DPRIMITIVE: fail sequential 0xff2053cc: 0x0000000f: vertex count 0xff2053d0: 0x00000003: start vertex 0xff2053d4: 0x00000000: instance count 0xff2053d8: 0x00000001: start instance 0xff2053dc: 0x00000000: index bias 0xff2053e0: 0x00000000: MI_NOOP 0xff2057e4: 0x05000000: MI_BATCH_BUFFER_END 0xff2057e8: 0x00000000: 0xff2057ec: 0x00000000: 0xff2057f0: 0x00000000: Should we assign to Mesa team? Or does it need investigation from GuC side?
Which SynMark tests fail, most of them or just some specific ones? Do other test-suites than SynMark fail with GuC? If not, which SynMark2 version you're using? v6.0? v7.0?
synmark-test.sh results in a GPU hang. synmark2testOgl.sh works fine. Version is 6.0
(In reply to anusha from comment #6) > synmark-test.sh results in a GPU hang. synmark2testOgl.sh works fine. > Version is 6.0 Please use v7.0 and confirm if you are seeing or not GPU hang. If it is occurring again then attach gpu crash dump.
(In reply to anusha from comment #6) > synmark-test.sh results in a GPU hang. synmark2testOgl.sh works fine. If you're using the same scripts as we are, the only difference between those two should be in which order they run the 40+ SynMark tests (synmark2testOgl.sh tries also couple of OpenCL tests which naturally will fail). In which SynMark test the GPU hang happens? Is it always the same one? Do you get hangs with any other test-suite (GpuTest, GfxBench, GLBenchmark, Unigine demos...)? Is there a difference if you run tests in fullscreen or in windowed mode (i.e. with Unity composition)?
Last configuration tested: Platform: KABY LAKE-U Processor : Genuine Intel(R) CPU 0000 @ 1.80GHz (cpu family: 6, model: 142, stepping: 9) MCP : KBL-U J0 2+3e QDF : QL9J PCH: PCH-LP C1 CRB : KABY LAKE U DDR3L RVP7 Rework: O-16 DMC 1.01 from https://01.org/sites/default/files/downloads/intelr-graphics-linux/kbldmcver101.tar.bz2 GuC 9.14 from http://rdvivi-hillsboro.jf.intel.com/firmware/kbl_guc_ver9_14.tar.bz2 Synmark2 from http://benchsrv.fi.intel.com/archive/benchmarks/ ***in this case i tested with Version 7 **** Kernel: commit b34b27fe61d8fe953e8bd28695c9407082e4667e Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Date: Tue Oct 4 11:12:04 2016 +0100 drm-intel-nightly: 2016y-10m-04d-10h-10m-17s UTC integration manifest Kernel version : 4.8.0-rc8-cedbdec Architecture : source amd64 all Component : drm url : http://cgit.freedesktop.org/mesa/drm tag : libdrm-2.4.68 commit : fc09c5ab84240e9b6bd0bed01685ef004f56c4fa Component : mesa url : http://cgit.freedesktop.org/mesa/mesa tag : mesa-12.0.1 commit : 04277f058d00238937e664cf546c43b16cea7b2b Component : xf86-video-intel url : http://cgit.freedesktop.org/xorg/driver/xf86-video-intel tag : 2.99.917-701-g205146b commit : 205146b0fdc8db016e5cfeeae5a6b25df3470ebc Component : libva url : http://cgit.freedesktop.org/libva tag : libva-1.7.2.pre1 commit : 9927bd2fbbb1d923fd6a8932a3cdbb5c9185ee22 Component : intel-driver url : http://cgit.freedesktop.org/vaapi/intel-driver tag : 1.7.2.pre1 commit : ce444fb412966ca6afbb1331b7cae8ab621c1108 Component : cairo url : http://cgit.freedesktop.org/cairo tag : 1.15.2 commit : db8a7f1697c49ae4942d2aa49eed52dd73dd9c7a Component : xserver url : http://cgit.freedesktop.org/xorg/xserver tag : xorg-server-1.18.3 commit : 9454cd51da9b38b974cff7c8b7125901f6403848 Component : macros url : https://cgit.freedesktop.org/xorg/util/macros tag : util-macros-1.19.0-2-gd7acec2 commit : d7acec2 Component : intel-gpu-tools url : https://cgit.freedesktop.org/xorg/app/intel-gpu-tools tag : intel-gpu-tools-1.16 commit : a28e9e38a9efc6daf5a08d60d29adcd3e328fe6f As requested, test.conf was modified with FALSE Tested 3 different scenarios 1.- Guc parameters in grub = i915.enable_guc_loading=2 i915.enable_guc_submission=2 GPU hang on test OglBatch0 2.- Guc parameters in grub = i915.enable_guc_loading=0 i915.enable_guc_submission=0 GPU hand on test OglDrvShComp
3.- No guc parameters in grub. All test finished without problems. Attaching 3 different Synmark Logs and Dmesg
Created attachment 127033 [details] All sysmark logs and dmesg for each scenario
(In reply to Elio from comment #11) > Created attachment 127033 [details] > All sysmark logs and dmesg for each scenario So, do all (or at least most) of the SynMark tests fail with GuC? Please run them one by one. And if most of those fail, test also other benchmarks. It would be really peculiar if most SynMark tests fail, but no other benchmarks fail.
(In reply to Eero Tamminen from comment #12) > (In reply to Elio from comment #11) > > Created attachment 127033 [details] > > All sysmark logs and dmesg for each scenario > > So, do all (or at least most) of the SynMark tests fail with GuC? Please > run them one by one. Elio confirmed it happens with OglBatch0 (comment 9). > And if most of those fail, test also other benchmarks. It would be really > peculiar if most SynMark tests fail, but no other benchmarks fail. OK, we can have a try Eero, do you also reproduce it? I shared with you the GuC installation procedure.
(In reply to cprigent from comment #13) > Eero, do you also reproduce it? I shared with you the GuC installation > procedure. While I can comment on what needs to be tested in order to narrow down this bug (e.g. to what component is responsible), I'm not planning on testing it myself (don't have the time right now).
Good news, I don't reproduce it with a fresh setup. Platform: KABY LAKE-U Processor : Genuine Intel(R) CPU 0000 @ 1.80GHz (cpu family: 6, model: 142, stepping: 9) MCP : KBL-U J0 2+3e QDF : QL9J PCH: PCH-LP C1 CRB : KABY LAKE U DDR3L RVP7 Software BIOS: 45.1 3KBLSE2R1.R00.X045.P01.1606291634 from https://ubit-artifactory-ba.intel.com/artifactory/owr-repos/Submissions/ifwi/KBL_ORANGE_IFWI_2016_WW27_3_03_SR'17/ ME FW: 11.6.0.1065 EC FW: 1.24 KSC: 1.24 Linux distribution: Ubuntu 16.04 64 bits DMC 1.01 from https://01.org/sites/default/files/downloads/intelr-graphics-linux/kbldmcver101.tar.bz2 GuC 9.14 from http://rdvivi-hillsboro.jf.intel.com/firmware/kbl_guc_ver9_14.tar.bz2 Kernel: 4.8.0 f35ed31 from http://cgit.freedesktop.org/drm-intel/ commit f35ed31aea66b3230c366fcba5f3456ae2cb956e Author: Jani Nikula <jani.nikula@intel.com> Date: Mon Oct 10 14:29:09 2016 +0300 drm-intel-nightly: 2016y-10m-10d-11h-28m-51s UTC integration manifest libdrm-2.4.71 a44c9c3 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-1.18.99.901-76 97a8353 from git://git.freedesktop.org/git/xorg/xserver xf86-video-intel 2.99.917-712 696f58f 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: 1.7.2-133 dd73514 from git://git.freedesktop.org/git/vaapi/intel-driver SynMark2 from http://benchsrv.fi.intel.com/archive/benchmarks/SynMark2-7.0.tar.gz External screen: DELL U2311Hb (DP)
So closed
(In reply to cprigent from comment #15) > Good news, I don't reproduce it with a fresh setup. What changed with the "fresh setup" compared to the other setup?
Bad news. I just reproduced it. I didn't save the logs associated to comment 15 but I'm quite sure the GuC was not loaded.
(In reply to Eero Tamminen from comment #12) > And if most of those fail, test also other benchmarks. It would be really > peculiar if most SynMark tests fail, but no other benchmarks fail. GuC Loaded: Heaven: Not reproduced (launched twice) Valley: DUT freezed (need to unplug charger, logs lost after reboot), then not reproduced the 2nd try gfxbench3_desktop: DUT also freezed (motherboard indicates a crash, need to unplug charger, logs lost after reboot) GuC not Loaded: Synmark2: no problem Heaven: no problem Valley: no problem gfxbench3_desktop: no problem
Re-tested with latest nightly kernel, same behavior, should we exclude or blacklist those chrashing tests? commit 21787266bd182df4c0d2067cf1b5c2379f61c24d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Oct 19 16:40:09 2016 +0100 drm-intel-nightly: 2016y-10m-19d-15h-39m-22s UTC integration manifest Kernel version : 4.9.0-rc1-fb0d8ec Architecture : source amd64 all
Created attachment 127451 [details] attachment-17355-0.html Hi Thank you for your email. I am OoO returning 24th of Oct 2016. If any urgent contact Kimmo Nikkanen. Br, Jani
Latest configuration for GFX is working without problems + latest patches for firmware. Guc 9.14 and 2.00.18 Huc: (Graphic Stack) Intel® Graphics for Linux* | 01.org ============================================ Software information ============================================ Kernel version : 4.9.0-rc8latestfirmware+ Linux distribution : Ubuntu 16.04.1 LTS Architecture : 64-bit Gfx stack code : 2504012391 Mesa version : 13.0.2 (git-c9e993b xf86-video-intel version : Xorg-Xserver version : 1.19.0 DRM version : 2.4.74 VAAPI version : Intel i965 driver for Intel(R) Kabylake - 1.7.3 Cairo version : 1.15.2 Intel GPU Tools version : Tag [intel-gpu-tools-1.17] / Commit [e631bb5] Kernel driver in use : i915 Hardware acceleration : Enabled Bios revision : 52.10 Bios release date : 10/05/2016 KSC revision : 1.24 ============================================ Firmwares information ============================================ DMC fw loaded : yes DMC version : 1.1 GUC fw loaded : SUCCESS GUC version wanted : 9.14 GUC version found : 9.14 Kernel parameters ============================================ drm.debug=0xe i915.enable_guc_loading=2 i915.enable_guc_submission=0 quiet splash Works for me without problems including performance behavior.
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.