kernel/linux.git/drivers/gpu/drm/v3d, branch v5.1.16

drm/v3d: Handle errors from IRQ setup.

2019-05-31T13:43:58+00:00

[ Upstream commit fc22771547e7e8a63679f0218e943d72b107de65 ] Noted in review by Dave Emett for V3D 4.2 support. Signed-off-by: Eric Anholt Link: https://patchwork.freedesktop.org/patch/msgid/20190308174336.7866-1-eric@anholt.net Reviewed-by: Dave Emett Signed-off-by: Sasha Levin

drm/sched: Refactor ring mirror list handling.

2019-01-25T21:15:36+00:00

Decauple sched threads stop and start and ring mirror list handling from the policy of what to do about the guilty jobs. When stoppping the sched thread and detaching sched fences from non signaled HW fenes wait for all signaled HW fences to complete before rerunning the jobs. v2: Fix resubmission of guilty job into HW after refactoring. v4: Full restart for all the jobs, not only from guilty ring. Extract karma increase into standalone function. v5: Rework waiting for signaled jobs without relying on the job struct itself as those might already be freed for non 'guilty' job's schedulers. Expose karma increase to drivers. v6: Use list_for_each_entry_safe_continue and drm_sched_process_job in case fence already signaled. Call drm_sched_increase_karma only once for amdgpu and add documentation. v7: Wait only for the latest job's fence. Suggested-by: Christian Koenig Signed-off-by: Andrey Grodzovsky Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/v3d: Invalidate the caches from the outside in.

2018-12-07T18:56:51+00:00

This would be a fairly obscure race, but let's make sure we don't ever lose it. Signed-off-by: Eric Anholt Link: https://patchwork.freedesktop.org/patch/msgid/20181203222438.25417-6-eric@anholt.net Reviewed-by: Dave Emett

drm/v3d: Stop trying to flush L2C on V3D 3.3+

2018-12-07T18:56:36+00:00

This cache was replaced with the slice accessing the L2T in the newer generations. Noted by Dave during review. Signed-off-by: Eric Anholt Link: https://patchwork.freedesktop.org/patch/msgid/20181203222438.25417-5-eric@anholt.net Reviewed-by: Dave Emett

drm/v3d: Drop the wait for L2T flush to complete.

2018-12-07T18:56:25+00:00

According to Dave, once you've started an L2T flush, all L2T accesses will be blocked until the flush completes. This fixes a consistent 3-4ms stall between the ioctl and running the job, and 3DMMES Taiji goes from 27fps to 110fps. v2: Leave a note about why we don't need to wait for completion. Signed-off-by: Eric Anholt Fixes: 57692c94dcbe ("drm/v3d: Introduce a new DRM driver for Broadcom V3D V3.x+") Reviewed-by: Dave Emett Link: https://patchwork.freedesktop.org/patch/msgid/20181203222438.25417-4-eric@anholt.net

drm/v3d: Don't bother flushing L1TD at job start.

2018-12-07T18:56:01+00:00

This is the write combiner for TMU writes. You're supposed to flush that at job end if you had dirtied any cachelines. Flushing it at job start then doesn't make any sense. Signed-off-by: Eric Anholt Fixes: 57692c94dcbe ("drm/v3d: Introduce a new DRM driver for Broadcom V3D V3.x+") Reviewed-by: Dave Emett Link: https://patchwork.freedesktop.org/patch/msgid/20181203222438.25417-3-eric@anholt.net

drm/v3d: Drop unused v3d_flush_caches().

2018-12-07T18:55:57+00:00

Now that I've specified how the end-of-pipeline flushing should work, we're never going to use this function. Signed-off-by: Eric Anholt Reviewed-by: Dave Emett Link: https://patchwork.freedesktop.org/patch/msgid/20181203222438.25417-2-eric@anholt.net

drm/v3d: fix broken build

2018-12-06T09:29:32+00:00

I missed one case during the recent revert of the replace_fence interface change. Fixes: 0b258ed1a219 drm: revert "expand replace_fence to support timeline point v2" Signed-off-by: Christian König Acked-by: Alex Deucher Link: https://patchwork.freedesktop.org/patch/266134/

drm: revert "expand replace_fence to support timeline point v2"

2018-12-05T10:01:11+00:00

This reverts commit 9a09a42369a4a37a959c051d8e1a1f948c1529a4. The whole interface isn't thought through. Since this function can't fail we actually can't allocate an object to store the sync point. Sorry, I should have taken the lead on this from the very beginning and reviewed it more thoughtfully. Going to propose a new interface as a follow up change. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Link: https://patchwork.freedesktop.org/patch/265580/

drm/v3d: Add more tracepoints for V3D GPU rendering.

2018-12-03T19:26:23+00:00

The core scheduler tells us when the job is pushed to the scheduler's queue, and I had the job_run functions saying when they actually queue the job to the hardware. By adding tracepoints for the very top of the ioctls and the IRQs signaling job completion, "perf record -a -e v3d:.\* -e gpu_scheduler:.\* ; perf script" gets you a pretty decent timeline. Signed-off-by: Eric Anholt Link: https://patchwork.freedesktop.org/patch/msgid/20181201005759.28093-5-eric@anholt.net Reviewed-by: Dave Emett