Something that I've been using locally for a while, and found quite
helpful for debugging gallium driver issues:
1) Better messages about what process triggered a crash, which is
in particular particularly useful for piglit
2) Mechanism to dump just submits which triggered GPU hangs.
Eventually I want to extend #2 to also dump submits that trigger
faults, but this is going to require spiffing out IOMMU framework
somewhat.. in particular we can't dump from IRQ context, so a
workqueue is needed, but we want to keep the IOMMU stalled until
after we dump the submit. Maybe we'll get something useable out
of the currently discussed SVM patches.
Rob Clark (6):
drm/msm: show task cmdline in gpu recovery messages
drm/msm: add special _get_vaddr_active() for cmdstream dumps
drm/msm: split rd debugfs file
drm/msm/rd: allow adding addition msg to top of dump
drm/msm: preserve IOVAs in submit's bo table
drm/msm: dump submits which triggered gpu hang
drivers/gpu/drm/msm/msm_drv.h | 7 ++-
drivers/gpu/drm/msm/msm_gem.c | 22 ++++++-
drivers/gpu/drm/msm/msm_gem_submit.c | 11 ++--
drivers/gpu/drm/msm/msm_gpu.c | 90 +++++++++++++++++++--------
drivers/gpu/drm/msm/msm_rd.c | 116 ++++++++++++++++++++++++++---------
5 files changed, 183 insertions(+), 63 deletions(-)
--
2.13.6
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
