summaryrefslogtreecommitdiff
path: root/overlay/gpu-perf.c
AgeCommit message (Collapse)Author
2014-12-21overlay: Hide kworker threads in overviewChris Wilson
The kworker threads are used for flip handling and other non-userspace driver tasks. They are non-blocking and so do not impact upon how userspace performs, but they do obscure that information in the overview. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-29overlay: Monitor per-ring context switch rateChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-27overlay: Read power from perf_eventsChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-26overlay: Count number of semaphores used by each processChris Wilson
This required me to contract the per-process information considerably, hopefully readability is not sacrificed too much. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-23overlay: Search for right debugfs pathChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-22overlay: Fill the blank charts with error messagesChris Wilson
Instead of showing nothing, show a "space left intentionally blank" message. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-22overlay: Remove busy estimationChris Wilson
Too inaccurate to be useful. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-20overlay: Add copyright statementsChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-18overlay: Count flips per planeChris Wilson
Stop the misleading double-accounting of flips when we have multiple displays active. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Correct layout of ring/seqno in raw sampleChris Wilson
For seqno completion, the events are too coarse i.e. one event may signal the completion of a few seqno. We will need to sort the events to properly compute the busy times. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Accumulate busy timesChris Wilson
Still a little too course as we add multiple overlapping seqnos and waits. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Couple wait begin/end events together to fix accountingChris Wilson
Since the events may be processed out of order (due to per-cpu ringbuffers) we need to be careful to associated wait pairs in order to compute the correct elapsed time. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Perf output redirection must be done after mmap()Chris Wilson
Effectively you choose to redirect the output of an event instead of mmapping it directly - but the target must already be mmapped. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Show per-process wait timesChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Track requests per-processChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Drop unused mmap/comm eventsChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Constify perf ringbufferChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Fix wraparound handling of perf ringbufferChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2013-08-17overlay: Primitive integration with perfChris Wilson
We can now record when a pageflip occurs by listening for the flip tracepoint. Merely proof of principle at this point. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-16 04:20:22 +0000