On Thu, Jun 3, 2021 at 5:11 PM Daniel Vetter daniel@ffwll.ch wrote:
On Thu, Jun 03, 2021 at 10:22:37AM -0400, Andrey Grodzovsky wrote:
Ping
Andrey
On 2021-06-02 10:20 a.m., Andrey Grodzovsky wrote:
On 2021-06-02 3:59 a.m., Daniel Vetter wrote:
On Tue, Jun 1, 2021 at 10:17 PM Andrey Grodzovsky andrey.grodzovsky@amd.com wrote:
Adding some tests to acompany the recently added hot-unplug feature. For now the test suite is disabled until the feature propagates from drm-misc-next to drm-next.
Andrey Grodzovsky (7): tests/amdgpu: Fix valgrind warning xf86drm: Add function to retrieve char device path test/amdgpu: Add helper functions for hot unplug test/amdgpu/hotunplug: Add test suite for GPU unplug test/amdgpu/hotunplug: Add basic test tests/amdgpu/hotunplug: Add unplug with cs test. tests/amdgpu/hotunplug: Add hotunplug with exported bo test
Given how nasty hotunplug is I really think collaborating on igt tests on this would be best for everyone ... do we have to keep doing parallel tests here for amdgpu? -Daniel
AFAIK as far as AMD goes a lot of developers use libdrm for regression testing while developing their features and also QA as i can see from some internal ticket specifically opened for failing to pass libdrm tests. From my bitter experience with GPU reset - features which are not part of a common use case such as device loading, mode setting or commands submissions tend to very quickly break as people develop features but never test them in those uncommon use cases - this is why I feel it will be very helpful to include those tests in libdrm.
Also given that this is libdrm amdgpu code it fits naturally into libdrm.
Regarding IGT - as you may remember I have them there too - https://gitlab.freedesktop.org/agrodzov/igt-gpu-tools/-/commits/master I hit some compile breakage on debian platform there which i need to resolve before i will submit for review there too.
Why can't amd run the igt tests? Afaiui on the display side this is happening already, at least sometimes.
And yes regression tests matter, it just feels silly that we need to have them 2x for amdgpu. For old stuff the old repo is all fine, but for new feature it looks a bit silly.
Different teams, different CI infrastructures... It ends up being easier to port tests between frameworks than to switch frameworks in some cases.
Alex
-Daniel
Andrey
tests/amdgpu/amdgpu_test.c | 42 +++- tests/amdgpu/amdgpu_test.h | 26 +++ tests/amdgpu/basic_tests.c | 5 +- tests/amdgpu/hotunplug_tests.c | 357 +++++++++++++++++++++++++++++++++ tests/amdgpu/meson.build | 1 + xf86drm.c | 23 +++ xf86drm.h | 1 + 7 files changed, 450 insertions(+), 5 deletions(-) create mode 100644 tests/amdgpu/hotunplug_tests.c
-- 2.25.1
-- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch