Re: [Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-03-23 Thread Andi Shyti
Hi, On Mon, Mar 20, 2023 at 08:21:17PM +0100, Andi Shyti wrote: > From: Jonathan Cavitt > > The gt_tlb live selftest has the same code coverage as the > igt_cs_tlb subtest of gtt, except it is better at detecting > TLB bugs. Furthermore, while igt_cs_tlb is hitting some > unforeseen issues, the

[Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-03-20 Thread Andi Shyti
From: Jonathan Cavitt The gt_tlb live selftest has the same code coverage as the igt_cs_tlb subtest of gtt, except it is better at detecting TLB bugs. Furthermore, while igt_cs_tlb is hitting some unforeseen issues, these issues are either false positives due to the test being poorly formatted,

Re: [Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-03-20 Thread Das, Nirmoy
On 2/17/2023 11:33 PM, Jonathan Cavitt wrote: The gt_tlb live selftest has the same code coverage as the igt_cs_tlb subtest of gtt, except it is better at detecting TLB bugs. Furthermore, while igt_cs_tlb is hitting some unforeseen issues, these issues are either false positives due to the tes

Re: [Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-02-22 Thread Andrzej Hajda
On 17.02.2023 23:33, Jonathan Cavitt wrote: The gt_tlb live selftest has the same code coverage as the igt_cs_tlb subtest of gtt, except it is better at detecting TLB bugs. Furthermore, while igt_cs_tlb is hitting some unforeseen issues, these issues are either false positives due to the test be

[Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-02-17 Thread Jonathan Cavitt
The gt_tlb live selftest has the same code coverage as the igt_cs_tlb subtest of gtt, except it is better at detecting TLB bugs. Furthermore, while igt_cs_tlb is hitting some unforeseen issues, these issues are either false positives due to the test being poorly formatted, or are true positives th

Re: [Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-02-17 Thread Chris Wilson
Quoting Jonathan Cavitt (2023-02-17 17:20:19) > The gt_tlb live selftest has the same code coverage as the > igt_cs_tlb subtest of gtt. True, the intent of the code is the same, but gt_tlb has had a much high success rate at hitting TLB bugs, so is my preferred test. > However, igt_cs_tlb is hitt

[Intel-gfx] [PATCH] drm/i915/selftests: Drop igt_cs_tlb

2023-02-17 Thread Jonathan Cavitt
The gt_tlb live selftest has the same code coverage as the igt_cs_tlb subtest of gtt. However, igt_cs_tlb is hitting an issue in that we are updating PTE in gt0->vm but executing on the media tile without updating gt1->vm. This issue is corrected in gt_tlb, and thus igt_cs_tlb is obsolete and shou