* [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-19 8:50 ` Daniel Vetter 0 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-19 8:50 UTC (permalink / raw) To: DRI Development Cc: IGT development, Daniel Vetter, Intel Graphics Development, amd-gfx Hi all, This is just to collect feedback on this idea, and see whether the overall dri-devel community stands on all this. I think the past few cross-vendor uapi extensions all came with igts attached, and personally I think there's lots of value in having them: A cross-vendor interface isn't useful if every driver implements it slightly differently. I think there's 2 questions here: - Do we want to make such testcases mandatory? - If yes, are we there yet, or is there something crucially missing still? And of course there's a bunch of details to figure out. Like we probably want to also recommend the selftests/unit-tests in drivers/gpu/drm/selftest, since fairly often that's a much more effective approach to checking the details than from userspace. Feedback and thoughts very much appreciated. Cheers, Daniel --- Documentation/gpu/drm-uapi.rst | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst index 4b4bf2c5eac5..91cf6e4b6303 100644 --- a/Documentation/gpu/drm-uapi.rst +++ b/Documentation/gpu/drm-uapi.rst @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of Testing and validation ====================== +Testing Requirements for userspace API +-------------------------------------- + +New cross-driver userspace interface extensions, like new IOCTL, new KMS +properties, new files in sysfs or anything else that constitutes an API change +need to have driver-agnostic testcases in IGT for that feature. + Validating changes with IGT --------------------------- -- 2.19.1 _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel ^ permalink raw reply related [flat|nested] 27+ messages in thread
* [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-19 8:50 ` Daniel Vetter 0 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-19 8:50 UTC (permalink / raw) To: DRI Development; +Cc: IGT development, Intel Graphics Development, amd-gfx Hi all, This is just to collect feedback on this idea, and see whether the overall dri-devel community stands on all this. I think the past few cross-vendor uapi extensions all came with igts attached, and personally I think there's lots of value in having them: A cross-vendor interface isn't useful if every driver implements it slightly differently. I think there's 2 questions here: - Do we want to make such testcases mandatory? - If yes, are we there yet, or is there something crucially missing still? And of course there's a bunch of details to figure out. Like we probably want to also recommend the selftests/unit-tests in drivers/gpu/drm/selftest, since fairly often that's a much more effective approach to checking the details than from userspace. Feedback and thoughts very much appreciated. Cheers, Daniel --- Documentation/gpu/drm-uapi.rst | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst index 4b4bf2c5eac5..91cf6e4b6303 100644 --- a/Documentation/gpu/drm-uapi.rst +++ b/Documentation/gpu/drm-uapi.rst @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of Testing and validation ====================== +Testing Requirements for userspace API +-------------------------------------- + +New cross-driver userspace interface extensions, like new IOCTL, new KMS +properties, new files in sysfs or anything else that constitutes an API change +need to have driver-agnostic testcases in IGT for that feature. + Validating changes with IGT --------------------------- -- 2.19.1 _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply related [flat|nested] 27+ messages in thread
* ✗ Fi.CI.CHECKPATCH: warning for RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter (?) @ 2018-10-22 10:27 ` Patchwork -1 siblings, 0 replies; 27+ messages in thread From: Patchwork @ 2018-10-22 10:27 UTC (permalink / raw) To: Daniel Vetter; +Cc: intel-gfx == Series Details == Series: RFC: Make igts for cross-driver stuff mandatory? URL : https://patchwork.freedesktop.org/series/51239/ State : warning == Summary == $ dim checkpatch origin/drm-tip dc8067b1819e RFC: Make igts for cross-driver stuff mandatory? -:48: ERROR:MISSING_SIGN_OFF: Missing Signed-off-by: line(s) total: 1 errors, 0 warnings, 0 checks, 13 lines checked _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* ✓ Fi.CI.BAT: success for RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter (?) (?) @ 2018-10-22 10:52 ` Patchwork -1 siblings, 0 replies; 27+ messages in thread From: Patchwork @ 2018-10-22 10:52 UTC (permalink / raw) To: Daniel Vetter; +Cc: intel-gfx == Series Details == Series: RFC: Make igts for cross-driver stuff mandatory? URL : https://patchwork.freedesktop.org/series/51239/ State : success == Summary == = CI Bug Log - changes from CI_DRM_5015 -> Patchwork_10513 = == Summary - SUCCESS == No regressions found. External URL: https://patchwork.freedesktop.org/api/1.0/series/51239/revisions/1/mbox/ == Known issues == Here are the changes found in Patchwork_10513 that come from known issues: === IGT changes === ==== Issues hit ==== igt@drv_selftest@live_evict: fi-bsw-kefka: PASS -> DMESG-WARN (fdo#107709) ==== Possible fixes ==== igt@gem_exec_suspend@basic-s4-devices: fi-blb-e6850: INCOMPLETE (fdo#107718) -> PASS igt@kms_frontbuffer_tracking@basic: fi-icl-u2: FAIL (fdo#103167) -> PASS igt@kms_pipe_crc_basic@read-crc-pipe-a: fi-byt-clapper: FAIL (fdo#107362) -> PASS fdo#103167 https://bugs.freedesktop.org/show_bug.cgi?id=103167 fdo#107362 https://bugs.freedesktop.org/show_bug.cgi?id=107362 fdo#107709 https://bugs.freedesktop.org/show_bug.cgi?id=107709 fdo#107718 https://bugs.freedesktop.org/show_bug.cgi?id=107718 == Participating hosts (52 -> 46) == Missing (6): fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-snb-2520m fi-ctg-p8600 fi-icl-u == Build changes == * Linux: CI_DRM_5015 -> Patchwork_10513 CI_DRM_5015: ec78bedc9006e9e56067eb6524bbd3064475a055 @ git://anongit.freedesktop.org/gfx-ci/linux IGT_4685: 78619fde4008424c472906041edb1d204e014f7c @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools Patchwork_10513: dc8067b1819ea163a17d8d81ed3ed478718ea341 @ git://anongit.freedesktop.org/gfx-ci/linux == Linux commits == dc8067b1819e RFC: Make igts for cross-driver stuff mandatory? == Logs == For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10513/issues.html _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* ✗ Fi.CI.IGT: failure for RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter ` (2 preceding siblings ...) (?) @ 2018-10-22 12:59 ` Patchwork -1 siblings, 0 replies; 27+ messages in thread From: Patchwork @ 2018-10-22 12:59 UTC (permalink / raw) To: Daniel Vetter; +Cc: intel-gfx == Series Details == Series: RFC: Make igts for cross-driver stuff mandatory? URL : https://patchwork.freedesktop.org/series/51239/ State : failure == Summary == = CI Bug Log - changes from CI_DRM_5015_full -> Patchwork_10513_full = == Summary - FAILURE == Serious unknown changes coming with Patchwork_10513_full absolutely need to be verified manually. If you think the reported changes have nothing to do with the changes introduced in Patchwork_10513_full, please notify your bug team to allow them to document this new failure mode, which will reduce false positives in CI. == Possible new issues == Here are the unknown changes that may have been introduced in Patchwork_10513_full: === IGT changes === ==== Possible regressions ==== igt@pm_rps@reset: shard-skl: PASS -> FAIL ==== Warnings ==== igt@kms_frontbuffer_tracking@fbc-2p-indfb-fliptrack: shard-hsw: SKIP -> PASS igt@perf_pmu@rc6: shard-kbl: SKIP -> PASS igt@pm_rc6_residency@rc6-accuracy: shard-snb: SKIP -> PASS == Known issues == Here are the changes found in Patchwork_10513_full that come from known issues: === IGT changes === ==== Issues hit ==== igt@gem_exec_basic@gtt-bsd1: shard-kbl: PASS -> DMESG-WARN (fdo#105602, fdo#103558) igt@gem_ppgtt@blt-vs-render-ctx0: shard-kbl: PASS -> INCOMPLETE (fdo#103665, fdo#106023) igt@gem_workarounds@suspend-resume: shard-kbl: PASS -> INCOMPLETE (fdo#103665) igt@kms_busy@extended-pageflip-modeset-hang-oldfb-render-b: shard-kbl: PASS -> DMESG-WARN (fdo#107956) igt@kms_cursor_crc@cursor-64x21-sliding: shard-apl: PASS -> FAIL (fdo#103232) +2 igt@kms_cursor_crc@cursor-size-change: shard-glk: PASS -> FAIL (fdo#103232) igt@kms_flip@flip-vs-expired-vblank: shard-skl: PASS -> FAIL (fdo#105363) igt@kms_flip@plain-flip-fb-recreate-interruptible: shard-skl: PASS -> FAIL (fdo#100368) igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen: shard-apl: PASS -> FAIL (fdo#103167) +2 igt@kms_plane@pixel-format-pipe-b-planes: shard-skl: NOTRUN -> DMESG-FAIL (fdo#103166, fdo#106885) igt@kms_plane_alpha_blend@pipe-c-alpha-transparant-fb: shard-skl: NOTRUN -> FAIL (fdo#108145) igt@kms_plane_alpha_blend@pipe-c-coverage-7efc: shard-skl: NOTRUN -> FAIL (fdo#108146) igt@kms_plane_multiple@atomic-pipe-a-tiling-y: shard-glk: PASS -> FAIL (fdo#103166) igt@kms_plane_multiple@atomic-pipe-b-tiling-x: shard-apl: PASS -> FAIL (fdo#103166) ==== Possible fixes ==== igt@gem_exec_await@wide-contexts: shard-apl: FAIL (fdo#106680) -> PASS igt@kms_cursor_crc@cursor-128x128-suspend: shard-glk: FAIL (fdo#103232) -> PASS +2 igt@kms_cursor_crc@cursor-256x256-dpms: shard-apl: FAIL (fdo#103232) -> PASS +1 igt@kms_draw_crc@draw-method-xrgb2101010-render-untiled: shard-skl: FAIL (fdo#103184) -> PASS igt@kms_flip@plain-flip-fb-recreate-interruptible: shard-glk: FAIL (fdo#100368) -> PASS igt@kms_plane_alpha_blend@pipe-b-constant-alpha-max: shard-glk: FAIL (fdo#108145) -> PASS +2 igt@kms_plane_multiple@atomic-pipe-b-tiling-y: shard-glk: FAIL (fdo#103166) -> PASS +3 igt@kms_setmode@basic: shard-apl: FAIL (fdo#99912) -> PASS igt@pm_rpm@gem-mmap-cpu: shard-skl: INCOMPLETE (fdo#107807) -> PASS igt@pm_rps@reset: shard-apl: FAIL (fdo#102250) -> PASS fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368 fdo#102250 https://bugs.freedesktop.org/show_bug.cgi?id=102250 fdo#103166 https://bugs.freedesktop.org/show_bug.cgi?id=103166 fdo#103167 https://bugs.freedesktop.org/show_bug.cgi?id=103167 fdo#103184 https://bugs.freedesktop.org/show_bug.cgi?id=103184 fdo#103232 https://bugs.freedesktop.org/show_bug.cgi?id=103232 fdo#103558 https://bugs.freedesktop.org/show_bug.cgi?id=103558 fdo#103665 https://bugs.freedesktop.org/show_bug.cgi?id=103665 fdo#105363 https://bugs.freedesktop.org/show_bug.cgi?id=105363 fdo#105602 https://bugs.freedesktop.org/show_bug.cgi?id=105602 fdo#106023 https://bugs.freedesktop.org/show_bug.cgi?id=106023 fdo#106680 https://bugs.freedesktop.org/show_bug.cgi?id=106680 fdo#106885 https://bugs.freedesktop.org/show_bug.cgi?id=106885 fdo#107807 https://bugs.freedesktop.org/show_bug.cgi?id=107807 fdo#107956 https://bugs.freedesktop.org/show_bug.cgi?id=107956 fdo#108145 https://bugs.freedesktop.org/show_bug.cgi?id=108145 fdo#108146 https://bugs.freedesktop.org/show_bug.cgi?id=108146 fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912 == Participating hosts (6 -> 6) == No changes in participating hosts == Build changes == * Linux: CI_DRM_5015 -> Patchwork_10513 CI_DRM_5015: ec78bedc9006e9e56067eb6524bbd3064475a055 @ git://anongit.freedesktop.org/gfx-ci/linux IGT_4685: 78619fde4008424c472906041edb1d204e014f7c @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools Patchwork_10513: dc8067b1819ea163a17d8d81ed3ed478718ea341 @ git://anongit.freedesktop.org/gfx-ci/linux piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit == Logs == For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_10513/shards.html _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <20181019085049.25482-1-daniel.vetter-/w4YWyX8dFk@public.gmane.org>]
* Re: [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter @ 2018-10-25 9:58 ` Liviu Dudau -1 siblings, 0 replies; 27+ messages in thread From: Liviu Dudau @ 2018-10-25 9:58 UTC (permalink / raw) To: Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, DRI Development On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > Hi all, Hi, (Replying from my personal address as the work email seems to have let this one go to /dev/null) > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. I would like to also get some clarification on where we are standing on "tests vs 'real code'" stanza. Does making igt tests mandatory replace the need for 'real code' or does it add to the list of requirements? If the later, then I think the bar rises in terms of showing igts' usefulness / benefits. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? I'm a bit reluctant to make it so by fiat. I think that showing the usefulness of having igts tests to newcomers (by adding with this patch some more information about why IGT is a good place to add your testing to) and getting more mature drivers to get tested under IGT on a regular basis would make adoption of IGT for testing a community standard. > > - If yes, are we there yet, or is there something crucially missing > still? I'm just getting back into IGT by refreshing the writeback patches, but by looking at the commit log I get the impression that there aren't that many patches that target drivers other than Intel's. Are all the non-Intel patches so generic that one doesn't need to specify a target driver for those changes (in which case great, but then why is Intel's so different?), or are the others not bothered with IGT support? At the moment I'm a bit on the fence on this. Not having spent too much time with IGT in the last 6 months, I'm probably closer to a newcomer in my attitude towards IGT and at the moment I'm not clear on how to answer the "Why?" and "What is in it for me?" questions. Best regards, Liviu > > And of course there's a bunch of details to figure out. Like we > probably want to also recommend the selftests/unit-tests in > drivers/gpu/drm/selftest, since fairly often that's a much more > effective approach to checking the details than from userspace. > > Feedback and thoughts very much appreciated. > > Cheers, Daniel > --- > Documentation/gpu/drm-uapi.rst | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > index 4b4bf2c5eac5..91cf6e4b6303 100644 > --- a/Documentation/gpu/drm-uapi.rst > +++ b/Documentation/gpu/drm-uapi.rst > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > Testing and validation > ====================== > > +Testing Requirements for userspace API > +-------------------------------------- > + > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > +properties, new files in sysfs or anything else that constitutes an API change > +need to have driver-agnostic testcases in IGT for that feature. > + > Validating changes with IGT > --------------------------- > > -- > 2.19.1 > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel -- /`\ / : | _.._ | '/ /` \ | / | .-._ '-"` ( |_/ / o o\ | == () == \ -- / ______________________________________ / ---<_ ________| |_______ | \\ \ | I would like to fix the world but | / | | \\__ \ | no one gives me the source code. | / / ; |.__) / |______________________________________| \ (_/.-. ; /__________) (_________\ { `| \_/ '-\ / | | / | / \ '-. \__|-----' _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-25 9:58 ` Liviu Dudau 0 siblings, 0 replies; 27+ messages in thread From: Liviu Dudau @ 2018-10-25 9:58 UTC (permalink / raw) To: Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx, DRI Development On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > Hi all, Hi, (Replying from my personal address as the work email seems to have let this one go to /dev/null) > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. I would like to also get some clarification on where we are standing on "tests vs 'real code'" stanza. Does making igt tests mandatory replace the need for 'real code' or does it add to the list of requirements? If the later, then I think the bar rises in terms of showing igts' usefulness / benefits. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? I'm a bit reluctant to make it so by fiat. I think that showing the usefulness of having igts tests to newcomers (by adding with this patch some more information about why IGT is a good place to add your testing to) and getting more mature drivers to get tested under IGT on a regular basis would make adoption of IGT for testing a community standard. > > - If yes, are we there yet, or is there something crucially missing > still? I'm just getting back into IGT by refreshing the writeback patches, but by looking at the commit log I get the impression that there aren't that many patches that target drivers other than Intel's. Are all the non-Intel patches so generic that one doesn't need to specify a target driver for those changes (in which case great, but then why is Intel's so different?), or are the others not bothered with IGT support? At the moment I'm a bit on the fence on this. Not having spent too much time with IGT in the last 6 months, I'm probably closer to a newcomer in my attitude towards IGT and at the moment I'm not clear on how to answer the "Why?" and "What is in it for me?" questions. Best regards, Liviu > > And of course there's a bunch of details to figure out. Like we > probably want to also recommend the selftests/unit-tests in > drivers/gpu/drm/selftest, since fairly often that's a much more > effective approach to checking the details than from userspace. > > Feedback and thoughts very much appreciated. > > Cheers, Daniel > --- > Documentation/gpu/drm-uapi.rst | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > index 4b4bf2c5eac5..91cf6e4b6303 100644 > --- a/Documentation/gpu/drm-uapi.rst > +++ b/Documentation/gpu/drm-uapi.rst > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > Testing and validation > ====================== > > +Testing Requirements for userspace API > +-------------------------------------- > + > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > +properties, new files in sysfs or anything else that constitutes an API change > +need to have driver-agnostic testcases in IGT for that feature. > + > Validating changes with IGT > --------------------------- > > -- > 2.19.1 > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel -- /`\ / : | _.._ | '/ /` \ | / | .-._ '-"` ( |_/ / o o\ | == () == \ -- / ______________________________________ / ---<_ ________| |_______ | \\ \ | I would like to fix the world but | / | | \\__ \ | no one gives me the source code. | / / ; |.__) / |______________________________________| \ (_/.-. ; /__________) (_________\ { `| \_/ '-\ / | | / | / \ '-. \__|-----' _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <20181025095804.GB29083-hOhETlTuV5niMG9XS5x8Mg@public.gmane.org>]
* Re: [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-25 9:58 ` [igt-dev] " Liviu Dudau @ 2018-10-25 10:31 ` Daniel Vetter -1 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-25 10:31 UTC (permalink / raw) To: Liviu Dudau; +Cc: IGT development, intel-gfx, amd-gfx list, dri-devel On Thu, Oct 25, 2018 at 11:58 AM Liviu Dudau <liviu@dudau.co.uk> wrote: > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > Hi all, > > Hi, > > (Replying from my personal address as the work email seems to have let > this one go to /dev/null) > > > > > This is just to collect feedback on this idea, and see whether the > > overall dri-devel community stands on all this. I think the past few > > cross-vendor uapi extensions all came with igts attached, and > > personally I think there's lots of value in having them: A > > cross-vendor interface isn't useful if every driver implements it > > slightly differently. > > I would like to also get some clarification on where we are standing on > "tests vs 'real code'" stanza. Does making igt tests mandatory replace > the need for 'real code' or does it add to the list of requirements? If > the later, then I think the bar rises in terms of showing igts' > usefulness / benefits. It would be in addition. The "real code" userspace is for validating the overall design. Igt (and unit tests) are more for checking all the details, error handling, and having a regression test suite going forward. > > I think there's 2 questions here: > > > > - Do we want to make such testcases mandatory? > > I'm a bit reluctant to make it so by fiat. I think that showing the > usefulness of having igts tests to newcomers (by adding with this patch > some more information about why IGT is a good place to add your testing to) > and getting more mature drivers to get tested under IGT on a regular basis > would make adoption of IGT for testing a community standard. There's 2 motivations here for me: - Most of the generic features in the past 1-2 did come with igts, but sometimes those igts have been treated as 2nd class and forgotten. I think it's at least the right time to discuss whether we should make them an integral part of uapi development. - The other bit is that our intel CI is catching a lot of regressions, and people have started to cc: intel-gfx to get their non-intel patches validated too. So all these igts we have do seem to provide real value in keeping things working. > > - If yes, are we there yet, or is there something crucially missing > > still? > > I'm just getting back into IGT by refreshing the writeback patches, but > by looking at the commit log I get the impression that there aren't that > many patches that target drivers other than Intel's. Are all the non-Intel > patches so generic that one doesn't need to specify a target driver for > those changes (in which case great, but then why is Intel's so different?), > or are the others not bothered with IGT support? So this discussion isn't about igt overall, but just about cross-driver features. The tests we have in igt for intel are a lot more: - We also validate all the intel-specific bits with igts. - We validate all the interactions between generic stuff and intel-specific uapi with igts. - We validate hw features with igt. And finally we've made igts mandatory for all things intel more than 5 years ago. So there's considerable head start. The other side is that I'm only suggesting that we make igts mandatory for cross-driver interfaces. Naturally the tests for those aren't aimed at a specific driver, but use the DRIVER_ANY flag. And the work needed to make that work on a specific driver is usually rather small - e.g. vmwgfx (which doesn't support GEM at all) required very small changes to get things going. Now we do have some tests for other drivers (vc4 and a few amdgpu), but not anything else. So I'd say the lack of non-intel targetted patches is a good thing. > At the moment I'm a bit on the fence on this. Not having spent too much > time with IGT in the last 6 months, I'm probably closer to a newcomer in > my attitude towards IGT and at the moment I'm not clear on how to answer the > "Why?" and "What is in it for me?" questions. Well all the usual reasons for testing: - Only way to make sure different implementations are working correct. - Only real way to make sure regressions are caught before everything is on fire. Of course this gets a lot better if there's also CI running them (which we do for intel, and are open to anyone submitting stuff to it), but just having the tests is already a big step. Cheers, Daniel > Best regards, > Liviu > > > > > And of course there's a bunch of details to figure out. Like we > > probably want to also recommend the selftests/unit-tests in > > drivers/gpu/drm/selftest, since fairly often that's a much more > > effective approach to checking the details than from userspace. > > > > Feedback and thoughts very much appreciated. > > > > Cheers, Daniel > > --- > > Documentation/gpu/drm-uapi.rst | 7 +++++++ > > 1 file changed, 7 insertions(+) > > > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > > index 4b4bf2c5eac5..91cf6e4b6303 100644 > > --- a/Documentation/gpu/drm-uapi.rst > > +++ b/Documentation/gpu/drm-uapi.rst > > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > > Testing and validation > > ====================== > > > > +Testing Requirements for userspace API > > +-------------------------------------- > > + > > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > > +properties, new files in sysfs or anything else that constitutes an API change > > +need to have driver-agnostic testcases in IGT for that feature. > > + > > Validating changes with IGT > > --------------------------- > > > > -- > > 2.19.1 > > > > _______________________________________________ > > dri-devel mailing list > > dri-devel@lists.freedesktop.org > > https://lists.freedesktop.org/mailman/listinfo/dri-devel > > -- > /`\ > / : | > _.._ | '/ > /` \ | / > | .-._ '-"` ( > |_/ / o o\ > | == () == > \ -- / ______________________________________ > / ---<_ ________| |_______ > | \\ \ | I would like to fix the world but | / > | | \\__ \ | no one gives me the source code. | / > / ; |.__) / |______________________________________| \ > (_/.-. ; /__________) (_________\ > { `| \_/ > '-\ / | > | / | > / \ '-. > \__|-----' -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [Intel-gfx] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-25 10:31 ` Daniel Vetter 0 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-25 10:31 UTC (permalink / raw) To: Liviu Dudau; +Cc: IGT development, intel-gfx, amd-gfx list, dri-devel On Thu, Oct 25, 2018 at 11:58 AM Liviu Dudau <liviu@dudau.co.uk> wrote: > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > Hi all, > > Hi, > > (Replying from my personal address as the work email seems to have let > this one go to /dev/null) > > > > > This is just to collect feedback on this idea, and see whether the > > overall dri-devel community stands on all this. I think the past few > > cross-vendor uapi extensions all came with igts attached, and > > personally I think there's lots of value in having them: A > > cross-vendor interface isn't useful if every driver implements it > > slightly differently. > > I would like to also get some clarification on where we are standing on > "tests vs 'real code'" stanza. Does making igt tests mandatory replace > the need for 'real code' or does it add to the list of requirements? If > the later, then I think the bar rises in terms of showing igts' > usefulness / benefits. It would be in addition. The "real code" userspace is for validating the overall design. Igt (and unit tests) are more for checking all the details, error handling, and having a regression test suite going forward. > > I think there's 2 questions here: > > > > - Do we want to make such testcases mandatory? > > I'm a bit reluctant to make it so by fiat. I think that showing the > usefulness of having igts tests to newcomers (by adding with this patch > some more information about why IGT is a good place to add your testing to) > and getting more mature drivers to get tested under IGT on a regular basis > would make adoption of IGT for testing a community standard. There's 2 motivations here for me: - Most of the generic features in the past 1-2 did come with igts, but sometimes those igts have been treated as 2nd class and forgotten. I think it's at least the right time to discuss whether we should make them an integral part of uapi development. - The other bit is that our intel CI is catching a lot of regressions, and people have started to cc: intel-gfx to get their non-intel patches validated too. So all these igts we have do seem to provide real value in keeping things working. > > - If yes, are we there yet, or is there something crucially missing > > still? > > I'm just getting back into IGT by refreshing the writeback patches, but > by looking at the commit log I get the impression that there aren't that > many patches that target drivers other than Intel's. Are all the non-Intel > patches so generic that one doesn't need to specify a target driver for > those changes (in which case great, but then why is Intel's so different?), > or are the others not bothered with IGT support? So this discussion isn't about igt overall, but just about cross-driver features. The tests we have in igt for intel are a lot more: - We also validate all the intel-specific bits with igts. - We validate all the interactions between generic stuff and intel-specific uapi with igts. - We validate hw features with igt. And finally we've made igts mandatory for all things intel more than 5 years ago. So there's considerable head start. The other side is that I'm only suggesting that we make igts mandatory for cross-driver interfaces. Naturally the tests for those aren't aimed at a specific driver, but use the DRIVER_ANY flag. And the work needed to make that work on a specific driver is usually rather small - e.g. vmwgfx (which doesn't support GEM at all) required very small changes to get things going. Now we do have some tests for other drivers (vc4 and a few amdgpu), but not anything else. So I'd say the lack of non-intel targetted patches is a good thing. > At the moment I'm a bit on the fence on this. Not having spent too much > time with IGT in the last 6 months, I'm probably closer to a newcomer in > my attitude towards IGT and at the moment I'm not clear on how to answer the > "Why?" and "What is in it for me?" questions. Well all the usual reasons for testing: - Only way to make sure different implementations are working correct. - Only real way to make sure regressions are caught before everything is on fire. Of course this gets a lot better if there's also CI running them (which we do for intel, and are open to anyone submitting stuff to it), but just having the tests is already a big step. Cheers, Daniel > Best regards, > Liviu > > > > > And of course there's a bunch of details to figure out. Like we > > probably want to also recommend the selftests/unit-tests in > > drivers/gpu/drm/selftest, since fairly often that's a much more > > effective approach to checking the details than from userspace. > > > > Feedback and thoughts very much appreciated. > > > > Cheers, Daniel > > --- > > Documentation/gpu/drm-uapi.rst | 7 +++++++ > > 1 file changed, 7 insertions(+) > > > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > > index 4b4bf2c5eac5..91cf6e4b6303 100644 > > --- a/Documentation/gpu/drm-uapi.rst > > +++ b/Documentation/gpu/drm-uapi.rst > > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > > Testing and validation > > ====================== > > > > +Testing Requirements for userspace API > > +-------------------------------------- > > + > > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > > +properties, new files in sysfs or anything else that constitutes an API change > > +need to have driver-agnostic testcases in IGT for that feature. > > + > > Validating changes with IGT > > --------------------------- > > > > -- > > 2.19.1 > > > > _______________________________________________ > > dri-devel mailing list > > dri-devel@lists.freedesktop.org > > https://lists.freedesktop.org/mailman/listinfo/dri-devel > > -- > /`\ > / : | > _.._ | '/ > /` \ | / > | .-._ '-"` ( > |_/ / o o\ > | == () == > \ -- / ______________________________________ > / ---<_ ________| |_______ > | \\ \ | I would like to fix the world but | / > | | \\__ \ | no one gives me the source code. | / > / ; |.__) / |______________________________________| \ > (_/.-. ; /__________) (_________\ > { `| \_/ > '-\ / | > | / | > / \ '-. > \__|-----' -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter @ 2018-10-26 15:28 ` Alex Deucher -1 siblings, 0 replies; 27+ messages in thread From: Alex Deucher @ 2018-10-26 15:28 UTC (permalink / raw) To: Daniel Vetter Cc: igt-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, Intel Graphics Development, amd-gfx list, Maling list - DRI developers On Fri, Oct 19, 2018 at 4:51 AM Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? > > - If yes, are we there yet, or is there something crucially missing > still? > > And of course there's a bunch of details to figure out. Like we > probably want to also recommend the selftests/unit-tests in > drivers/gpu/drm/selftest, since fairly often that's a much more > effective approach to checking the details than from userspace. > > Feedback and thoughts very much appreciated. As long as we can fix up any cross platform issues, this seems reasonable to me. Alex > > Cheers, Daniel > --- > Documentation/gpu/drm-uapi.rst | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > index 4b4bf2c5eac5..91cf6e4b6303 100644 > --- a/Documentation/gpu/drm-uapi.rst > +++ b/Documentation/gpu/drm-uapi.rst > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > Testing and validation > ====================== > > +Testing Requirements for userspace API > +-------------------------------------- > + > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > +properties, new files in sysfs or anything else that constitutes an API change > +need to have driver-agnostic testcases in IGT for that feature. > + > Validating changes with IGT > --------------------------- > > -- > 2.19.1 > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-26 15:28 ` Alex Deucher 0 siblings, 0 replies; 27+ messages in thread From: Alex Deucher @ 2018-10-26 15:28 UTC (permalink / raw) To: Daniel Vetter Cc: igt-dev, Intel Graphics Development, amd-gfx list, Maling list - DRI developers On Fri, Oct 19, 2018 at 4:51 AM Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? > > - If yes, are we there yet, or is there something crucially missing > still? > > And of course there's a bunch of details to figure out. Like we > probably want to also recommend the selftests/unit-tests in > drivers/gpu/drm/selftest, since fairly often that's a much more > effective approach to checking the details than from userspace. > > Feedback and thoughts very much appreciated. As long as we can fix up any cross platform issues, this seems reasonable to me. Alex > > Cheers, Daniel > --- > Documentation/gpu/drm-uapi.rst | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst > index 4b4bf2c5eac5..91cf6e4b6303 100644 > --- a/Documentation/gpu/drm-uapi.rst > +++ b/Documentation/gpu/drm-uapi.rst > @@ -238,6 +238,13 @@ DRM specific patterns. Note that ENOTTY has the slightly unintuitive meaning of > Testing and validation > ====================== > > +Testing Requirements for userspace API > +-------------------------------------- > + > +New cross-driver userspace interface extensions, like new IOCTL, new KMS > +properties, new files in sysfs or anything else that constitutes an API change > +need to have driver-agnostic testcases in IGT for that feature. > + > Validating changes with IGT > --------------------------- > > -- > 2.19.1 > > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [Intel-gfx] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter @ 2018-10-30 2:17 ` Dave Airlie -1 siblings, 0 replies; 27+ messages in thread From: Dave Airlie @ 2018-10-30 2:17 UTC (permalink / raw) To: Daniel Vetter Cc: igt-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, intel-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, amd-gfx mailing list, dri-devel On Fri, 19 Oct 2018 at 18:51, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? Yes I think if at all practical it probably makes sense to have some mandatory test cases for all cross-vendor features, or features that might become cross vendor in the future. > > - If yes, are we there yet, or is there something crucially missing > still? I think the does igt build in all the places needed is the main one, I've no idea what a baseline IGT test run looks like on non-intel hw, how useful is it? Acked-by: Dave Airlie <airlied@redhat.com> _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [Intel-gfx] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-30 2:17 ` Dave Airlie 0 siblings, 0 replies; 27+ messages in thread From: Dave Airlie @ 2018-10-30 2:17 UTC (permalink / raw) To: Daniel Vetter; +Cc: igt-dev, intel-gfx, amd-gfx mailing list, dri-devel On Fri, 19 Oct 2018 at 18:51, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? Yes I think if at all practical it probably makes sense to have some mandatory test cases for all cross-vendor features, or features that might become cross vendor in the future. > > - If yes, are we there yet, or is there something crucially missing > still? I think the does igt build in all the places needed is the main one, I've no idea what a baseline IGT test run looks like on non-intel hw, how useful is it? Acked-by: Dave Airlie <airlied@redhat.com> _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <CAPM=9tykF1U0Kah42ej7GR=6HyKe+WeV-LRe7_xB0VEUkikdPQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>]
* Re: [Intel-gfx] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-30 2:17 ` [igt-dev] " Dave Airlie @ 2018-10-30 8:54 ` Daniel Vetter -1 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-30 8:54 UTC (permalink / raw) To: Dave Airlie Cc: igt-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, Daniel Vetter, intel-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, amd-gfx mailing list, dri-devel On Tue, Oct 30, 2018 at 12:17:30PM +1000, Dave Airlie wrote: > On Fri, 19 Oct 2018 at 18:51, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > > > Hi all, > > > > This is just to collect feedback on this idea, and see whether the > > overall dri-devel community stands on all this. I think the past few > > cross-vendor uapi extensions all came with igts attached, and > > personally I think there's lots of value in having them: A > > cross-vendor interface isn't useful if every driver implements it > > slightly differently. > > > > I think there's 2 questions here: > > > > - Do we want to make such testcases mandatory? > > Yes I think if at all practical it probably makes sense to have some > mandatory test cases for all cross-vendor features, or features that > might become cross vendor in the future. I've created a few patches to test that in gitlab CI. I think the only thing left now is CI'ing sysroot builds, but I don't know how to do that myself. > > - If yes, are we there yet, or is there something crucially missing > > still? > > I think the does igt build in all the places needed is the main one, > I've no idea what a baseline IGT test run looks like on non-intel hw, > how useful is it? We're in the process of moving i915 tests into a tests/i915/ subfolder. I think after that we could try to them on some hardware (my long term plan is to use vkms for that and put it into gitlab CI with qemu). We have accidentally run igts on amdgpu instead of i915 on KBL-G (and our CI found at least one bug in one of my refactor series), so stuff works :-) Coverage is a bit a mixed bag I think, but that's always the case when you retrofit a testsuite. -Daniel > > Acked-by: Dave Airlie <airlied@redhat.com> -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [Intel-gfx] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-30 8:54 ` Daniel Vetter 0 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-30 8:54 UTC (permalink / raw) To: Dave Airlie Cc: igt-dev, Daniel Vetter, intel-gfx, amd-gfx mailing list, dri-devel On Tue, Oct 30, 2018 at 12:17:30PM +1000, Dave Airlie wrote: > On Fri, 19 Oct 2018 at 18:51, Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > > > Hi all, > > > > This is just to collect feedback on this idea, and see whether the > > overall dri-devel community stands on all this. I think the past few > > cross-vendor uapi extensions all came with igts attached, and > > personally I think there's lots of value in having them: A > > cross-vendor interface isn't useful if every driver implements it > > slightly differently. > > > > I think there's 2 questions here: > > > > - Do we want to make such testcases mandatory? > > Yes I think if at all practical it probably makes sense to have some > mandatory test cases for all cross-vendor features, or features that > might become cross vendor in the future. I've created a few patches to test that in gitlab CI. I think the only thing left now is CI'ing sysroot builds, but I don't know how to do that myself. > > - If yes, are we there yet, or is there something crucially missing > > still? > > I think the does igt build in all the places needed is the main one, > I've no idea what a baseline IGT test run looks like on non-intel hw, > how useful is it? We're in the process of moving i915 tests into a tests/i915/ subfolder. I think after that we could try to them on some hardware (my long term plan is to use vkms for that and put it into gitlab CI with qemu). We have accidentally run igts on amdgpu instead of i915 on KBL-G (and our CI found at least one bug in one of my refactor series), so stuff works :-) Coverage is a bit a mixed bag I think, but that's always the case when you retrofit a testsuite. -Daniel > > Acked-by: Dave Airlie <airlied@redhat.com> -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-19 8:50 ` [igt-dev] " Daniel Vetter @ 2018-10-25 12:51 ` Sean Paul -1 siblings, 0 replies; 27+ messages in thread From: Sean Paul @ 2018-10-25 12:51 UTC (permalink / raw) To: Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx, DRI Development On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? > Yes, more testing == better code. > - If yes, are we there yet, or is there something crucially missing > still? In my experience, no. Last week while trying to replicate an intel-gfx CI failure, I tried compiling igt for one of my (intel) chromebooks. It seems like cross-compilation (or, in my case, just specifying prefix/ld_library_path/sbin_path) is broken on igt. If we want to impose restrictions across the entire subsystem, we need to make sure that everyone can build and deploy igt easily. I managed to hack around everything and get it working, but I still haven't tried switching out the toolchain. Once we have some GitLab CI to validate cross-compilation, then we can consider making IGT mandatory. It's possible that I'm just a meson n00b and didn't use the right incantation, so maybe it already works, but then we need better documentation. I've pasted my horrible hacks below, I also didn't have libunwind, so removed its usage. Sean /snip From ab8c7d274c32559295b38d6ceeaabded14b207d4 Mon Sep 17 00:00:00 2001 From: Sean Paul <seanpaul@chromium.org> Date: Thu, 25 Oct 2018 08:40:28 -0400 Subject: [PATCH] igt: Hacks to compile in CrOS chroot Signed-off-by: Sean Paul <seanpaul@chromium.org> --- lib/igt_core.c | 78 --------------------------------------- lib/meson.build | 1 - meson.build | 4 +- tests/gem_userptr_blits.c | 2 + tools/meson.build | 7 ---- 5 files changed, 5 insertions(+), 87 deletions(-) diff --git a/lib/igt_core.c b/lib/igt_core.c index 23bb858f..ca65d7cc 100644 --- a/lib/igt_core.c +++ b/lib/igt_core.c @@ -71,8 +71,6 @@ #include "igt_sysrq.h" #include "igt_rc.h" -#define UNW_LOCAL_ONLY -#include <libunwind.h> #include <elfutils/libdwfl.h> #ifdef HAVE_LIBGEN_H @@ -1209,63 +1207,6 @@ static void write_stderr(const char *str) static void print_backtrace(void) { - unw_cursor_t cursor; - unw_context_t uc; - int stack_num = 0; - - Dwfl_Callbacks cbs = { - .find_elf = dwfl_linux_proc_find_elf, - .find_debuginfo = dwfl_standard_find_debuginfo, - }; - - Dwfl *dwfl = dwfl_begin(&cbs); - - if (dwfl_linux_proc_report(dwfl, getpid())) { - dwfl_end(dwfl); - dwfl = NULL; - } else - dwfl_report_end(dwfl, NULL, NULL); - - igt_info("Stack trace:\n"); - - unw_getcontext(&uc); - unw_init_local(&cursor, &uc); - while (unw_step(&cursor) > 0) { - char name[255]; - unw_word_t off, ip; - Dwfl_Module *mod = NULL; - - unw_get_reg(&cursor, UNW_REG_IP, &ip); - - if (dwfl) - mod = dwfl_addrmodule(dwfl, ip); - - if (mod) { - const char *src, *dwfl_name; - Dwfl_Line *line; - int lineno; - GElf_Sym sym; - - line = dwfl_module_getsrc(mod, ip); - dwfl_name = dwfl_module_addrsym(mod, ip, &sym, NULL); - - if (line && dwfl_name) { - src = dwfl_lineinfo(line, NULL, &lineno, NULL, NULL, NULL); - igt_info(" #%d %s:%d %s()\n", stack_num++, src, lineno, dwfl_name); - continue; - } - } - - if (unw_get_proc_name(&cursor, name, 255, &off) < 0) - igt_info(" #%d [<unknown>+0x%x]\n", stack_num++, - (unsigned int) ip); - else - igt_info(" #%d [%s+0x%x]\n", stack_num++, name, - (unsigned int) off); - } - - if (dwfl) - dwfl_end(dwfl); } static const char hex[] = "0123456789abcdef"; @@ -1420,25 +1361,6 @@ xprintf(const char *fmt, ...) static void print_backtrace_sig_safe(void) { - unw_cursor_t cursor; - unw_context_t uc; - int stack_num = 0; - - write_stderr("Stack trace: \n"); - - unw_getcontext(&uc); - unw_init_local(&cursor, &uc); - while (unw_step(&cursor) > 0) { - char name[255]; - unw_word_t off; - - if (unw_get_proc_name(&cursor, name, 255, &off) < 0) - xstrlcpy(name, "<unknown>", 10); - - xprintf(" #%d [%s+0x%x]\n", stack_num++, name, - (unsigned int) off); - - } } void __igt_fail_assert(const char *domain, const char *file, const int line, diff --git a/lib/meson.build b/lib/meson.build index 7e2c9b7a..26cef0c6 100644 --- a/lib/meson.build +++ b/lib/meson.build @@ -59,7 +59,6 @@ lib_deps = [ libkmod, libprocps, libudev, - libunwind, libdw, pciaccess, pthreads, diff --git a/meson.build b/meson.build index eff35585..86ad6602 100644 --- a/meson.build +++ b/meson.build @@ -102,7 +102,6 @@ build_info += 'With libdrm: ' + ','.join(libdrm_info) pciaccess = dependency('pciaccess', version : '>=0.10') libkmod = dependency('libkmod') libprocps = dependency('libprocps', required : true) -libunwind = dependency('libunwind', required : true) libdw = dependency('libdw', required : true) ssl = dependency('openssl', required : true) pixman = dependency('pixman-1', required : true) @@ -217,12 +216,15 @@ prefix = get_option('prefix') bindir = get_option('bindir') datadir = join_paths(get_option('datadir'), 'igt-gpu-tools') includedir = get_option('includedir') +fullincdir = join_paths(prefix, includedir) libdir = get_option('libdir') libexecdir = join_paths(get_option('libexecdir'), 'igt-gpu-tools') amdgpudir = join_paths(libexecdir, 'amdgpu') mandir = get_option('mandir') pkgconfigdir = join_paths(libdir, 'pkgconfig') +inc = [ include_directories(fullincdir, join_paths( fullincdir, 'libdrm')), inc ] + if get_option('use_rpath') # Set up runpath for the test executables towards libigt.so. # The path should be relative to $ORIGIN so the library is diff --git a/tests/gem_userptr_blits.c b/tests/gem_userptr_blits.c index 909dd19d..92d5a6e1 100644 --- a/tests/gem_userptr_blits.c +++ b/tests/gem_userptr_blits.c @@ -1069,6 +1069,7 @@ static void store_dword_rand(int i915, unsigned int engine, static void test_readonly(int i915) { +#if 0 unsigned char orig[SHA_DIGEST_LENGTH]; uint64_t aperture_size; uint32_t whandle, rhandle; @@ -1178,6 +1179,7 @@ static void test_readonly(int i915) munmap(space, total); munmap(pages, sz); +#endif } static jmp_buf sigjmp; diff --git a/tools/meson.build b/tools/meson.build index 79f36aa9..5e990cd7 100644 --- a/tools/meson.build +++ b/tools/meson.build @@ -90,11 +90,6 @@ install_subdir('registers', install_dir : datadir, 'Makefile', 'Makefile.in', 'Makefile.am', ]) -shared_library('intel_aubdump', 'aubdump.c', - dependencies : [ lib_igt_chipset, dlsym ], - name_prefix : '', - install : true) - executable('intel_gpu_top', 'intel_gpu_top.c', install : true, install_rpath : bindir_rpathdir, @@ -104,7 +99,5 @@ conf_data = configuration_data() conf_data.set('prefix', prefix) conf_data.set('exec_prefix', '${prefix}') conf_data.set('libdir', join_paths('${prefix}', libdir)) -configure_file(input : 'intel_aubdump.in', output : 'intel_aubdump', - configuration : conf_data, install_dir : bindir) subdir('null_state_gen') -- Sean Paul, Software Engineer, Google / Chromium OS _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply related [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-25 12:51 ` Sean Paul 0 siblings, 0 replies; 27+ messages in thread From: Sean Paul @ 2018-10-25 12:51 UTC (permalink / raw) To: Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx, DRI Development On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > Hi all, > > This is just to collect feedback on this idea, and see whether the > overall dri-devel community stands on all this. I think the past few > cross-vendor uapi extensions all came with igts attached, and > personally I think there's lots of value in having them: A > cross-vendor interface isn't useful if every driver implements it > slightly differently. > > I think there's 2 questions here: > > - Do we want to make such testcases mandatory? > Yes, more testing == better code. > - If yes, are we there yet, or is there something crucially missing > still? In my experience, no. Last week while trying to replicate an intel-gfx CI failure, I tried compiling igt for one of my (intel) chromebooks. It seems like cross-compilation (or, in my case, just specifying prefix/ld_library_path/sbin_path) is broken on igt. If we want to impose restrictions across the entire subsystem, we need to make sure that everyone can build and deploy igt easily. I managed to hack around everything and get it working, but I still haven't tried switching out the toolchain. Once we have some GitLab CI to validate cross-compilation, then we can consider making IGT mandatory. It's possible that I'm just a meson n00b and didn't use the right incantation, so maybe it already works, but then we need better documentation. I've pasted my horrible hacks below, I also didn't have libunwind, so removed its usage. Sean /snip From ab8c7d274c32559295b38d6ceeaabded14b207d4 Mon Sep 17 00:00:00 2001 From: Sean Paul <seanpaul@chromium.org> Date: Thu, 25 Oct 2018 08:40:28 -0400 Subject: [PATCH] igt: Hacks to compile in CrOS chroot Signed-off-by: Sean Paul <seanpaul@chromium.org> --- lib/igt_core.c | 78 --------------------------------------- lib/meson.build | 1 - meson.build | 4 +- tests/gem_userptr_blits.c | 2 + tools/meson.build | 7 ---- 5 files changed, 5 insertions(+), 87 deletions(-) diff --git a/lib/igt_core.c b/lib/igt_core.c index 23bb858f..ca65d7cc 100644 --- a/lib/igt_core.c +++ b/lib/igt_core.c @@ -71,8 +71,6 @@ #include "igt_sysrq.h" #include "igt_rc.h" -#define UNW_LOCAL_ONLY -#include <libunwind.h> #include <elfutils/libdwfl.h> #ifdef HAVE_LIBGEN_H @@ -1209,63 +1207,6 @@ static void write_stderr(const char *str) static void print_backtrace(void) { - unw_cursor_t cursor; - unw_context_t uc; - int stack_num = 0; - - Dwfl_Callbacks cbs = { - .find_elf = dwfl_linux_proc_find_elf, - .find_debuginfo = dwfl_standard_find_debuginfo, - }; - - Dwfl *dwfl = dwfl_begin(&cbs); - - if (dwfl_linux_proc_report(dwfl, getpid())) { - dwfl_end(dwfl); - dwfl = NULL; - } else - dwfl_report_end(dwfl, NULL, NULL); - - igt_info("Stack trace:\n"); - - unw_getcontext(&uc); - unw_init_local(&cursor, &uc); - while (unw_step(&cursor) > 0) { - char name[255]; - unw_word_t off, ip; - Dwfl_Module *mod = NULL; - - unw_get_reg(&cursor, UNW_REG_IP, &ip); - - if (dwfl) - mod = dwfl_addrmodule(dwfl, ip); - - if (mod) { - const char *src, *dwfl_name; - Dwfl_Line *line; - int lineno; - GElf_Sym sym; - - line = dwfl_module_getsrc(mod, ip); - dwfl_name = dwfl_module_addrsym(mod, ip, &sym, NULL); - - if (line && dwfl_name) { - src = dwfl_lineinfo(line, NULL, &lineno, NULL, NULL, NULL); - igt_info(" #%d %s:%d %s()\n", stack_num++, src, lineno, dwfl_name); - continue; - } - } - - if (unw_get_proc_name(&cursor, name, 255, &off) < 0) - igt_info(" #%d [<unknown>+0x%x]\n", stack_num++, - (unsigned int) ip); - else - igt_info(" #%d [%s+0x%x]\n", stack_num++, name, - (unsigned int) off); - } - - if (dwfl) - dwfl_end(dwfl); } static const char hex[] = "0123456789abcdef"; @@ -1420,25 +1361,6 @@ xprintf(const char *fmt, ...) static void print_backtrace_sig_safe(void) { - unw_cursor_t cursor; - unw_context_t uc; - int stack_num = 0; - - write_stderr("Stack trace: \n"); - - unw_getcontext(&uc); - unw_init_local(&cursor, &uc); - while (unw_step(&cursor) > 0) { - char name[255]; - unw_word_t off; - - if (unw_get_proc_name(&cursor, name, 255, &off) < 0) - xstrlcpy(name, "<unknown>", 10); - - xprintf(" #%d [%s+0x%x]\n", stack_num++, name, - (unsigned int) off); - - } } void __igt_fail_assert(const char *domain, const char *file, const int line, diff --git a/lib/meson.build b/lib/meson.build index 7e2c9b7a..26cef0c6 100644 --- a/lib/meson.build +++ b/lib/meson.build @@ -59,7 +59,6 @@ lib_deps = [ libkmod, libprocps, libudev, - libunwind, libdw, pciaccess, pthreads, diff --git a/meson.build b/meson.build index eff35585..86ad6602 100644 --- a/meson.build +++ b/meson.build @@ -102,7 +102,6 @@ build_info += 'With libdrm: ' + ','.join(libdrm_info) pciaccess = dependency('pciaccess', version : '>=0.10') libkmod = dependency('libkmod') libprocps = dependency('libprocps', required : true) -libunwind = dependency('libunwind', required : true) libdw = dependency('libdw', required : true) ssl = dependency('openssl', required : true) pixman = dependency('pixman-1', required : true) @@ -217,12 +216,15 @@ prefix = get_option('prefix') bindir = get_option('bindir') datadir = join_paths(get_option('datadir'), 'igt-gpu-tools') includedir = get_option('includedir') +fullincdir = join_paths(prefix, includedir) libdir = get_option('libdir') libexecdir = join_paths(get_option('libexecdir'), 'igt-gpu-tools') amdgpudir = join_paths(libexecdir, 'amdgpu') mandir = get_option('mandir') pkgconfigdir = join_paths(libdir, 'pkgconfig') +inc = [ include_directories(fullincdir, join_paths( fullincdir, 'libdrm')), inc ] + if get_option('use_rpath') # Set up runpath for the test executables towards libigt.so. # The path should be relative to $ORIGIN so the library is diff --git a/tests/gem_userptr_blits.c b/tests/gem_userptr_blits.c index 909dd19d..92d5a6e1 100644 --- a/tests/gem_userptr_blits.c +++ b/tests/gem_userptr_blits.c @@ -1069,6 +1069,7 @@ static void store_dword_rand(int i915, unsigned int engine, static void test_readonly(int i915) { +#if 0 unsigned char orig[SHA_DIGEST_LENGTH]; uint64_t aperture_size; uint32_t whandle, rhandle; @@ -1178,6 +1179,7 @@ static void test_readonly(int i915) munmap(space, total); munmap(pages, sz); +#endif } static jmp_buf sigjmp; diff --git a/tools/meson.build b/tools/meson.build index 79f36aa9..5e990cd7 100644 --- a/tools/meson.build +++ b/tools/meson.build @@ -90,11 +90,6 @@ install_subdir('registers', install_dir : datadir, 'Makefile', 'Makefile.in', 'Makefile.am', ]) -shared_library('intel_aubdump', 'aubdump.c', - dependencies : [ lib_igt_chipset, dlsym ], - name_prefix : '', - install : true) - executable('intel_gpu_top', 'intel_gpu_top.c', install : true, install_rpath : bindir_rpathdir, @@ -104,7 +99,5 @@ conf_data = configuration_data() conf_data.set('prefix', prefix) conf_data.set('exec_prefix', '${prefix}') conf_data.set('libdir', join_paths('${prefix}', libdir)) -configure_file(input : 'intel_aubdump.in', output : 'intel_aubdump', - configuration : conf_data, install_dir : bindir) subdir('null_state_gen') -- Sean Paul, Software Engineer, Google / Chromium OS _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply related [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-25 12:51 ` [igt-dev] " Sean Paul @ 2018-10-25 16:35 ` Eric Anholt -1 siblings, 0 replies; 27+ messages in thread From: Eric Anholt @ 2018-10-25 16:35 UTC (permalink / raw) To: Sean Paul, Daniel Vetter Cc: IGT development, Intel Graphics Development, DRI Development, amd-gfx [-- Attachment #1.1: Type: text/plain, Size: 1706 bytes --] Sean Paul <sean@poorly.run> writes: > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: >> Hi all, >> >> This is just to collect feedback on this idea, and see whether the >> overall dri-devel community stands on all this. I think the past few >> cross-vendor uapi extensions all came with igts attached, and >> personally I think there's lots of value in having them: A >> cross-vendor interface isn't useful if every driver implements it >> slightly differently. >> >> I think there's 2 questions here: >> >> - Do we want to make such testcases mandatory? >> > > Yes, more testing == better code. > > >> - If yes, are we there yet, or is there something crucially missing >> still? > > In my experience, no. Last week while trying to replicate an intel-gfx CI > failure, I tried compiling igt for one of my (intel) chromebooks. It seems like > cross-compilation (or, in my case, just specifying > prefix/ld_library_path/sbin_path) is broken on igt. If we want to impose > restrictions across the entire subsystem, we need to make sure that everyone can > build and deploy igt easily. > > I managed to hack around everything and get it working, but I still haven't > tried switching out the toolchain. Once we have some GitLab CI to validate > cross-compilation, then we can consider making IGT mandatory. > > It's possible that I'm just a meson n00b and didn't use the right incantation, > so maybe it already works, but then we need better documentation. > > I've pasted my horrible hacks below, I also didn't have libunwind, so removed > its usage. I've also had to cut out libunwind for cross-compiling on many occasions. Worst library. [-- Attachment #1.2: signature.asc --] [-- Type: application/pgp-signature, Size: 832 bytes --] [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [Intel-gfx] [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-25 16:35 ` Eric Anholt 0 siblings, 0 replies; 27+ messages in thread From: Eric Anholt @ 2018-10-25 16:35 UTC (permalink / raw) To: Sean Paul, Daniel Vetter Cc: IGT development, Intel Graphics Development, DRI Development, amd-gfx [-- Attachment #1.1: Type: text/plain, Size: 1706 bytes --] Sean Paul <sean@poorly.run> writes: > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: >> Hi all, >> >> This is just to collect feedback on this idea, and see whether the >> overall dri-devel community stands on all this. I think the past few >> cross-vendor uapi extensions all came with igts attached, and >> personally I think there's lots of value in having them: A >> cross-vendor interface isn't useful if every driver implements it >> slightly differently. >> >> I think there's 2 questions here: >> >> - Do we want to make such testcases mandatory? >> > > Yes, more testing == better code. > > >> - If yes, are we there yet, or is there something crucially missing >> still? > > In my experience, no. Last week while trying to replicate an intel-gfx CI > failure, I tried compiling igt for one of my (intel) chromebooks. It seems like > cross-compilation (or, in my case, just specifying > prefix/ld_library_path/sbin_path) is broken on igt. If we want to impose > restrictions across the entire subsystem, we need to make sure that everyone can > build and deploy igt easily. > > I managed to hack around everything and get it working, but I still haven't > tried switching out the toolchain. Once we have some GitLab CI to validate > cross-compilation, then we can consider making IGT mandatory. > > It's possible that I'm just a meson n00b and didn't use the right incantation, > so maybe it already works, but then we need better documentation. > > I've pasted my horrible hacks below, I also didn't have libunwind, so removed > its usage. I've also had to cut out libunwind for cross-compiling on many occasions. Worst library. [-- Attachment #1.2: signature.asc --] [-- Type: application/pgp-signature, Size: 832 bytes --] [-- Attachment #2: Type: text/plain, Size: 160 bytes --] _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <87efcernkp.fsf-WhKQ6XTQaPysTnJN9+BGXg@public.gmane.org>]
* RE: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-25 16:35 ` [Intel-gfx] " Eric Anholt @ 2018-10-26 3:49 ` Zhou, David(ChunMing) -1 siblings, 0 replies; 27+ messages in thread From: Zhou, David(ChunMing) @ 2018-10-26 3:49 UTC (permalink / raw) To: Eric Anholt, Sean Paul, Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org, DRI Development Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. 😊 Regards, David > -----Original Message----- > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > Anholt > Sent: Friday, October 26, 2018 12:36 AM > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > mandatory? > > Sean Paul <sean@poorly.run> writes: > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > >> Hi all, > >> > >> This is just to collect feedback on this idea, and see whether the > >> overall dri-devel community stands on all this. I think the past few > >> cross-vendor uapi extensions all came with igts attached, and > >> personally I think there's lots of value in having them: A > >> cross-vendor interface isn't useful if every driver implements it > >> slightly differently. > >> > >> I think there's 2 questions here: > >> > >> - Do we want to make such testcases mandatory? > >> > > > > Yes, more testing == better code. > > > > > >> - If yes, are we there yet, or is there something crucially missing > >> still? > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > It seems like cross-compilation (or, in my case, just specifying > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > impose restrictions across the entire subsystem, we need to make sure > > that everyone can build and deploy igt easily. > > > > I managed to hack around everything and get it working, but I still > > haven't tried switching out the toolchain. Once we have some GitLab CI > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > It's possible that I'm just a meson n00b and didn't use the right > > incantation, so maybe it already works, but then we need better > documentation. > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > removed its usage. > > I've also had to cut out libunwind for cross-compiling on many occasions. > Worst library. _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-26 3:49 ` Zhou, David(ChunMing) 0 siblings, 0 replies; 27+ messages in thread From: Zhou, David(ChunMing) @ 2018-10-26 3:49 UTC (permalink / raw) To: Eric Anholt, Sean Paul, Daniel Vetter Cc: IGT development, Intel Graphics Development, amd-gfx@lists.freedesktop.org, DRI Development Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. 😊 Regards, David > -----Original Message----- > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > Anholt > Sent: Friday, October 26, 2018 12:36 AM > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > mandatory? > > Sean Paul <sean@poorly.run> writes: > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > >> Hi all, > >> > >> This is just to collect feedback on this idea, and see whether the > >> overall dri-devel community stands on all this. I think the past few > >> cross-vendor uapi extensions all came with igts attached, and > >> personally I think there's lots of value in having them: A > >> cross-vendor interface isn't useful if every driver implements it > >> slightly differently. > >> > >> I think there's 2 questions here: > >> > >> - Do we want to make such testcases mandatory? > >> > > > > Yes, more testing == better code. > > > > > >> - If yes, are we there yet, or is there something crucially missing > >> still? > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > It seems like cross-compilation (or, in my case, just specifying > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > impose restrictions across the entire subsystem, we need to make sure > > that everyone can build and deploy igt easily. > > > > I managed to hack around everything and get it working, but I still > > haven't tried switching out the toolchain. Once we have some GitLab CI > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > It's possible that I'm just a meson n00b and didn't use the right > > incantation, so maybe it already works, but then we need better > documentation. > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > removed its usage. > > I've also had to cut out libunwind for cross-compiling on many occasions. > Worst library. _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <BY1PR12MB0502FFCA69A2C76817C22C8DB4F00-PicGAnIBOobrCwm+z9iKNgdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>]
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-26 3:49 ` Zhou, David(ChunMing) @ 2018-10-26 8:32 ` Daniel Vetter -1 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-26 8:32 UTC (permalink / raw) To: Chunming Zhou, Christian König, Alex Deucher Cc: intel-gfx, amd-gfx list, IGT development, Anholt, Eric, dri-devel, Sean Paul On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) <David1.Zhou@amd.com> wrote: > > Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. > You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. We renamed it ot IGT gpu tools, that was even enough for ARM folks. If this is seriously what AMD expects before considering, I'm not sure what to say. Alex, Christian, is this the official AMD stance that you can't touch stuff because of the letter i? -Daniel > Regards, > David > > > -----Original Message----- > > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > > Anholt > > Sent: Friday, October 26, 2018 12:36 AM > > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > > mandatory? > > > > Sean Paul <sean@poorly.run> writes: > > > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > >> Hi all, > > >> > > >> This is just to collect feedback on this idea, and see whether the > > >> overall dri-devel community stands on all this. I think the past few > > >> cross-vendor uapi extensions all came with igts attached, and > > >> personally I think there's lots of value in having them: A > > >> cross-vendor interface isn't useful if every driver implements it > > >> slightly differently. > > >> > > >> I think there's 2 questions here: > > >> > > >> - Do we want to make such testcases mandatory? > > >> > > > > > > Yes, more testing == better code. > > > > > > > > >> - If yes, are we there yet, or is there something crucially missing > > >> still? > > > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > > It seems like cross-compilation (or, in my case, just specifying > > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > > impose restrictions across the entire subsystem, we need to make sure > > > that everyone can build and deploy igt easily. > > > > > > I managed to hack around everything and get it working, but I still > > > haven't tried switching out the toolchain. Once we have some GitLab CI > > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > > > It's possible that I'm just a meson n00b and didn't use the right > > > incantation, so maybe it already works, but then we need better > > documentation. > > > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > > removed its usage. > > > > I've also had to cut out libunwind for cross-compiling on many occasions. > > Worst library. -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-26 8:32 ` Daniel Vetter 0 siblings, 0 replies; 27+ messages in thread From: Daniel Vetter @ 2018-10-26 8:32 UTC (permalink / raw) To: Chunming Zhou, Christian König, Alex Deucher Cc: intel-gfx, amd-gfx list, IGT development, dri-devel, Sean Paul On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) <David1.Zhou@amd.com> wrote: > > Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. > You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. We renamed it ot IGT gpu tools, that was even enough for ARM folks. If this is seriously what AMD expects before considering, I'm not sure what to say. Alex, Christian, is this the official AMD stance that you can't touch stuff because of the letter i? -Daniel > Regards, > David > > > -----Original Message----- > > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > > Anholt > > Sent: Friday, October 26, 2018 12:36 AM > > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > > mandatory? > > > > Sean Paul <sean@poorly.run> writes: > > > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > >> Hi all, > > >> > > >> This is just to collect feedback on this idea, and see whether the > > >> overall dri-devel community stands on all this. I think the past few > > >> cross-vendor uapi extensions all came with igts attached, and > > >> personally I think there's lots of value in having them: A > > >> cross-vendor interface isn't useful if every driver implements it > > >> slightly differently. > > >> > > >> I think there's 2 questions here: > > >> > > >> - Do we want to make such testcases mandatory? > > >> > > > > > > Yes, more testing == better code. > > > > > > > > >> - If yes, are we there yet, or is there something crucially missing > > >> still? > > > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > > It seems like cross-compilation (or, in my case, just specifying > > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > > impose restrictions across the entire subsystem, we need to make sure > > > that everyone can build and deploy igt easily. > > > > > > I managed to hack around everything and get it working, but I still > > > haven't tried switching out the toolchain. Once we have some GitLab CI > > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > > > It's possible that I'm just a meson n00b and didn't use the right > > > incantation, so maybe it already works, but then we need better > > documentation. > > > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > > removed its usage. > > > > I've also had to cut out libunwind for cross-compiling on many occasions. > > Worst library. -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev ^ permalink raw reply [flat|nested] 27+ messages in thread
[parent not found: <CAKMK7uFsET+riNKw6eGt1+Uiuhq8eww1E4OFYVCC83CfQFN5qg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>]
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-26 8:32 ` Daniel Vetter @ 2018-10-26 8:48 ` zhoucm1 -1 siblings, 0 replies; 27+ messages in thread From: zhoucm1 @ 2018-10-26 8:48 UTC (permalink / raw) To: Daniel Vetter, Chunming Zhou, Christian König, Alex Deucher Cc: intel-gfx, amd-gfx list, IGT development, Anholt, Eric, dri-devel, Sean Paul On 2018年10月26日 16:32, Daniel Vetter wrote: > On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) > <David1.Zhou@amd.com> wrote: >> Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. >> You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. > We renamed it ot IGT gpu tools, that was even enough for ARM folks. > If this is seriously what AMD expects before considering, I'm not sure > what to say. > > Alex, Christian, is this the official AMD stance that you can't touch > stuff because of the letter i? Nope, as I said last, this is just my personal thought. And I'm not sure what opinion of others. -David > -Daniel > > >> Regards, >> David >> >>> -----Original Message----- >>> From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric >>> Anholt >>> Sent: Friday, October 26, 2018 12:36 AM >>> To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> >>> Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics >>> Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- >>> devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org >>> Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff >>> mandatory? >>> >>> Sean Paul <sean@poorly.run> writes: >>> >>>> On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: >>>>> Hi all, >>>>> >>>>> This is just to collect feedback on this idea, and see whether the >>>>> overall dri-devel community stands on all this. I think the past few >>>>> cross-vendor uapi extensions all came with igts attached, and >>>>> personally I think there's lots of value in having them: A >>>>> cross-vendor interface isn't useful if every driver implements it >>>>> slightly differently. >>>>> >>>>> I think there's 2 questions here: >>>>> >>>>> - Do we want to make such testcases mandatory? >>>>> >>>> Yes, more testing == better code. >>>> >>>> >>>>> - If yes, are we there yet, or is there something crucially missing >>>>> still? >>>> In my experience, no. Last week while trying to replicate an intel-gfx >>>> CI failure, I tried compiling igt for one of my (intel) chromebooks. >>>> It seems like cross-compilation (or, in my case, just specifying >>>> prefix/ld_library_path/sbin_path) is broken on igt. If we want to >>>> impose restrictions across the entire subsystem, we need to make sure >>>> that everyone can build and deploy igt easily. >>>> >>>> I managed to hack around everything and get it working, but I still >>>> haven't tried switching out the toolchain. Once we have some GitLab CI >>>> to validate cross-compilation, then we can consider making IGT mandatory. >>>> >>>> It's possible that I'm just a meson n00b and didn't use the right >>>> incantation, so maybe it already works, but then we need better >>> documentation. >>>> I've pasted my horrible hacks below, I also didn't have libunwind, so >>>> removed its usage. >>> I've also had to cut out libunwind for cross-compiling on many occasions. >>> Worst library. > > _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-26 8:48 ` zhoucm1 0 siblings, 0 replies; 27+ messages in thread From: zhoucm1 @ 2018-10-26 8:48 UTC (permalink / raw) To: Daniel Vetter, Chunming Zhou, Christian König, Alex Deucher Cc: intel-gfx, amd-gfx list, IGT development, dri-devel, Sean Paul On 2018年10月26日 16:32, Daniel Vetter wrote: > On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) > <David1.Zhou@amd.com> wrote: >> Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. >> You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. > We renamed it ot IGT gpu tools, that was even enough for ARM folks. > If this is seriously what AMD expects before considering, I'm not sure > what to say. > > Alex, Christian, is this the official AMD stance that you can't touch > stuff because of the letter i? Nope, as I said last, this is just my personal thought. And I'm not sure what opinion of others. -David > -Daniel > > >> Regards, >> David >> >>> -----Original Message----- >>> From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric >>> Anholt >>> Sent: Friday, October 26, 2018 12:36 AM >>> To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> >>> Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics >>> Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- >>> devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org >>> Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff >>> mandatory? >>> >>> Sean Paul <sean@poorly.run> writes: >>> >>>> On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: >>>>> Hi all, >>>>> >>>>> This is just to collect feedback on this idea, and see whether the >>>>> overall dri-devel community stands on all this. I think the past few >>>>> cross-vendor uapi extensions all came with igts attached, and >>>>> personally I think there's lots of value in having them: A >>>>> cross-vendor interface isn't useful if every driver implements it >>>>> slightly differently. >>>>> >>>>> I think there's 2 questions here: >>>>> >>>>> - Do we want to make such testcases mandatory? >>>>> >>>> Yes, more testing == better code. >>>> >>>> >>>>> - If yes, are we there yet, or is there something crucially missing >>>>> still? >>>> In my experience, no. Last week while trying to replicate an intel-gfx >>>> CI failure, I tried compiling igt for one of my (intel) chromebooks. >>>> It seems like cross-compilation (or, in my case, just specifying >>>> prefix/ld_library_path/sbin_path) is broken on igt. If we want to >>>> impose restrictions across the entire subsystem, we need to make sure >>>> that everyone can build and deploy igt easily. >>>> >>>> I managed to hack around everything and get it working, but I still >>>> haven't tried switching out the toolchain. Once we have some GitLab CI >>>> to validate cross-compilation, then we can consider making IGT mandatory. >>>> >>>> It's possible that I'm just a meson n00b and didn't use the right >>>> incantation, so maybe it already works, but then we need better >>> documentation. >>>> I've pasted my horrible hacks below, I also didn't have libunwind, so >>>> removed its usage. >>> I've also had to cut out libunwind for cross-compiling on many occasions. >>> Worst library. > > _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? 2018-10-26 8:32 ` Daniel Vetter @ 2018-10-26 14:41 ` Alex Deucher -1 siblings, 0 replies; 27+ messages in thread From: Alex Deucher @ 2018-10-26 14:41 UTC (permalink / raw) To: Daniel Vetter Cc: Chunming Zhou, Christian König, Intel Graphics Development, amd-gfx list, igt-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW, Eric Anholt, Maling list - DRI developers, Deucher, Alexander, sean-p7yTbzM4H96eqtR555YLDQ On Fri, Oct 26, 2018 at 4:32 AM Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) > <David1.Zhou@amd.com> wrote: > > > > Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. > > You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. > > We renamed it ot IGT gpu tools, that was even enough for ARM folks. > If this is seriously what AMD expects before considering, I'm not sure > what to say. > > Alex, Christian, is this the official AMD stance that you can't touch > stuff because of the letter i? We don't have any restrictions. Alex > -Daniel > > > > Regards, > > David > > > > > -----Original Message----- > > > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > > > Anholt > > > Sent: Friday, October 26, 2018 12:36 AM > > > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > > > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > > > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > > > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > > > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > > > mandatory? > > > > > > Sean Paul <sean@poorly.run> writes: > > > > > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > > >> Hi all, > > > >> > > > >> This is just to collect feedback on this idea, and see whether the > > > >> overall dri-devel community stands on all this. I think the past few > > > >> cross-vendor uapi extensions all came with igts attached, and > > > >> personally I think there's lots of value in having them: A > > > >> cross-vendor interface isn't useful if every driver implements it > > > >> slightly differently. > > > >> > > > >> I think there's 2 questions here: > > > >> > > > >> - Do we want to make such testcases mandatory? > > > >> > > > > > > > > Yes, more testing == better code. > > > > > > > > > > > >> - If yes, are we there yet, or is there something crucially missing > > > >> still? > > > > > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > > > It seems like cross-compilation (or, in my case, just specifying > > > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > > > impose restrictions across the entire subsystem, we need to make sure > > > > that everyone can build and deploy igt easily. > > > > > > > > I managed to hack around everything and get it working, but I still > > > > haven't tried switching out the toolchain. Once we have some GitLab CI > > > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > > > > > It's possible that I'm just a meson n00b and didn't use the right > > > > incantation, so maybe it already works, but then we need better > > > documentation. > > > > > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > > > removed its usage. > > > > > > I've also had to cut out libunwind for cross-compiling on many occasions. > > > Worst library. > > > > -- > Daniel Vetter > Software Engineer, Intel Corporation > +41 (0) 79 365 57 48 - http://blog.ffwll.ch > _______________________________________________ > amd-gfx mailing list > amd-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/amd-gfx _______________________________________________ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx ^ permalink raw reply [flat|nested] 27+ messages in thread
* Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff mandatory? @ 2018-10-26 14:41 ` Alex Deucher 0 siblings, 0 replies; 27+ messages in thread From: Alex Deucher @ 2018-10-26 14:41 UTC (permalink / raw) To: Daniel Vetter Cc: Christian König, Intel Graphics Development, amd-gfx list, igt-dev, Maling list - DRI developers, Deucher, Alexander, sean On Fri, Oct 26, 2018 at 4:32 AM Daniel Vetter <daniel.vetter@ffwll.ch> wrote: > > On Fri, Oct 26, 2018 at 5:50 AM Zhou, David(ChunMing) > <David1.Zhou@amd.com> wrote: > > > > Make igt for cross-driver, I think you should rename it first, not an intel specific. NO company wants their employee working on other company stuff. > > You can rename it to DGT(drm graphics test), and published following libdrm, or directly merge to libdrm, then everyone can use it and develop it in same page, which is only my personal opinion. > > We renamed it ot IGT gpu tools, that was even enough for ARM folks. > If this is seriously what AMD expects before considering, I'm not sure > what to say. > > Alex, Christian, is this the official AMD stance that you can't touch > stuff because of the letter i? We don't have any restrictions. Alex > -Daniel > > > > Regards, > > David > > > > > -----Original Message----- > > > From: dri-devel <dri-devel-bounces@lists.freedesktop.org> On Behalf Of Eric > > > Anholt > > > Sent: Friday, October 26, 2018 12:36 AM > > > To: Sean Paul <sean@poorly.run>; Daniel Vetter <daniel.vetter@ffwll.ch> > > > Cc: IGT development <igt-dev@lists.freedesktop.org>; Intel Graphics > > > Development <intel-gfx@lists.freedesktop.org>; DRI Development <dri- > > > devel@lists.freedesktop.org>; amd-gfx@lists.freedesktop.org > > > Subject: Re: [igt-dev] [PATCH] RFC: Make igts for cross-driver stuff > > > mandatory? > > > > > > Sean Paul <sean@poorly.run> writes: > > > > > > > On Fri, Oct 19, 2018 at 10:50:49AM +0200, Daniel Vetter wrote: > > > >> Hi all, > > > >> > > > >> This is just to collect feedback on this idea, and see whether the > > > >> overall dri-devel community stands on all this. I think the past few > > > >> cross-vendor uapi extensions all came with igts attached, and > > > >> personally I think there's lots of value in having them: A > > > >> cross-vendor interface isn't useful if every driver implements it > > > >> slightly differently. > > > >> > > > >> I think there's 2 questions here: > > > >> > > > >> - Do we want to make such testcases mandatory? > > > >> > > > > > > > > Yes, more testing == better code. > > > > > > > > > > > >> - If yes, are we there yet, or is there something crucially missing > > > >> still? > > > > > > > > In my experience, no. Last week while trying to replicate an intel-gfx > > > > CI failure, I tried compiling igt for one of my (intel) chromebooks. > > > > It seems like cross-compilation (or, in my case, just specifying > > > > prefix/ld_library_path/sbin_path) is broken on igt. If we want to > > > > impose restrictions across the entire subsystem, we need to make sure > > > > that everyone can build and deploy igt easily. > > > > > > > > I managed to hack around everything and get it working, but I still > > > > haven't tried switching out the toolchain. Once we have some GitLab CI > > > > to validate cross-compilation, then we can consider making IGT mandatory. > > > > > > > > It's possible that I'm just a meson n00b and didn't use the right > > > > incantation, so maybe it already works, but then we need better > > > documentation. > > > > > > > > I've pasted my horrible hacks below, I also didn't have libunwind, so > > > > removed its usage. > > > > > > I've also had to cut out libunwind for cross-compiling on many occasions. > > > Worst library. > > > > -- > Daniel Vetter > Software Engineer, Intel Corporation > +41 (0) 79 365 57 48 - http://blog.ffwll.ch > _______________________________________________ > amd-gfx mailing list > amd-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/amd-gfx _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel ^ permalink raw reply [flat|nested] 27+ messages in thread
end of thread, other threads:[~2018-10-30 8:54 UTC | newest]
Thread overview: 27+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-10-19 8:50 [PATCH] RFC: Make igts for cross-driver stuff mandatory? Daniel Vetter
2018-10-19 8:50 ` [igt-dev] " Daniel Vetter
2018-10-22 10:27 ` ✗ Fi.CI.CHECKPATCH: warning for " Patchwork
2018-10-22 10:52 ` ✓ Fi.CI.BAT: success " Patchwork
2018-10-22 12:59 ` ✗ Fi.CI.IGT: failure " Patchwork
[not found] ` <20181019085049.25482-1-daniel.vetter-/w4YWyX8dFk@public.gmane.org>
2018-10-25 9:58 ` [PATCH] " Liviu Dudau
2018-10-25 9:58 ` [igt-dev] " Liviu Dudau
[not found] ` <20181025095804.GB29083-hOhETlTuV5niMG9XS5x8Mg@public.gmane.org>
2018-10-25 10:31 ` Daniel Vetter
2018-10-25 10:31 ` [Intel-gfx] " Daniel Vetter
2018-10-26 15:28 ` Alex Deucher
2018-10-26 15:28 ` Alex Deucher
2018-10-30 2:17 ` [Intel-gfx] " Dave Airlie
2018-10-30 2:17 ` [igt-dev] " Dave Airlie
[not found] ` <CAPM=9tykF1U0Kah42ej7GR=6HyKe+WeV-LRe7_xB0VEUkikdPQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2018-10-30 8:54 ` Daniel Vetter
2018-10-30 8:54 ` Daniel Vetter
2018-10-25 12:51 ` Sean Paul
2018-10-25 12:51 ` [igt-dev] " Sean Paul
2018-10-25 16:35 ` Eric Anholt
2018-10-25 16:35 ` [Intel-gfx] " Eric Anholt
[not found] ` <87efcernkp.fsf-WhKQ6XTQaPysTnJN9+BGXg@public.gmane.org>
2018-10-26 3:49 ` Zhou, David(ChunMing)
2018-10-26 3:49 ` Zhou, David(ChunMing)
[not found] ` <BY1PR12MB0502FFCA69A2C76817C22C8DB4F00-PicGAnIBOobrCwm+z9iKNgdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2018-10-26 8:32 ` Daniel Vetter
2018-10-26 8:32 ` Daniel Vetter
[not found] ` <CAKMK7uFsET+riNKw6eGt1+Uiuhq8eww1E4OFYVCC83CfQFN5qg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2018-10-26 8:48 ` zhoucm1
2018-10-26 8:48 ` zhoucm1
2018-10-26 14:41 ` Alex Deucher
2018-10-26 14:41 ` Alex Deucher
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.