Resend of [1] with the order rearranged to put a6xx gpu
state at the end and attach an add-on patch to better
manage a6xx state memory.
I think everything here is good to go for 4.21 (maybe
even some 4.20 fixes) but I can understand if we want
to let the gpu state soak for another cycle. Sharat
has been doing an excellent job of finding the bugs
in it.
Jordan Crouse (10):
drm/msm: Update generated headers
drm/msm/gpu: Allocate the correct size for the GPU memptrs
drm/msm: Gracefully handle failure in _msm_gem_kernel_new
drm/msm/gpu: Add per-submission statistics
drm/msm/gpu: Add trace events for tracking GPU submissions
drm/msm/gpu: Only store local command buffers in the GPU state
drm/msm/gpu: Move gpu_poll_timeout() to adreno_gpu.h
drm/msm/adreno: Don't capture register values if target doesn't define
them
drm/msm/a6xx: Add a6xx gpu state
drm/msm/a6xx: Track and manage a6xx state memory
drivers/gpu/drm/msm/Makefile | 4 +-
drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 5 -
drivers/gpu/drm/msm/adreno/a6xx.xml.h | 54 +-
drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 45 +-
drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 3 +
drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 80 +-
drivers/gpu/drm/msm/adreno/a6xx_gpu.h | 8 +
drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c | 1171 +++++++++++++++++++
drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h | 430 +++++++
drivers/gpu/drm/msm/adreno/adreno_gpu.c | 19 +-
drivers/gpu/drm/msm/adreno/adreno_gpu.h | 6 +
drivers/gpu/drm/msm/msm_gem.c | 18 +-
drivers/gpu/drm/msm/msm_gem.h | 1 +
drivers/gpu/drm/msm/msm_gem_submit.c | 15 +-
drivers/gpu/drm/msm/msm_gpu.c | 49 +-
drivers/gpu/drm/msm/msm_gpu_trace.h | 90 ++
drivers/gpu/drm/msm/msm_gpu_tracepoints.c | 6 +
drivers/gpu/drm/msm/msm_ringbuffer.h | 16 +
18 files changed, 1921 insertions(+), 99 deletions(-)
create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
create mode 100644 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h
create mode 100644 drivers/gpu/drm/msm/msm_gpu_trace.h
create mode 100644 drivers/gpu/drm/msm/msm_gpu_tracepoints.c
--
2.18.0
