8dbd1fb9d4
lib/drmtest: igt_skip when drm_get_card fails
...
Also add printf support to igt_skip to make sure there's always a
reason why we fail a testcase.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 11:09:25 +02:00
9ad062d0b3
lib/drmtest: extract gem_create_context
...
libdrm requirements are now up to it.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 10:55:28 +02:00
23eee39489
tests: roll out igt_require some more
...
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 10:47:53 +02:00
5951ffb6a0
lib/drmtest: rip out drm_open_any_master
...
It's unused. Also most of our tests failed to ask for the right type
of drm fd anyway. So it's imo better to just let them fall over when
they don't get master but want it, like they already do today.
This also allows us to garbage-collect the master parameter to
drm_get_card and associated code.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 10:34:34 +02:00
4ba1f2e104
lib/drmtest: Properly skip if no intel gfx supported
...
drm/* kernel test nirvana!
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 08:09:25 +02:00
47a32a3aec
lib/drmtest: make igt_fail work outside of subtests
...
With the abortable fixtures we can now correctly fail all subsequent
sutbests if we hit upon an igt_assert outside of a subtest.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 08:09:25 +02:00
78865847f9
lib/drmtest: skip fixtures after an igt_skip
...
This way we can just enclose all igt_skip/igt_require calls into
fixtures even when we have subtests, and still output correct SKIP
message for all of them. Wohoo, magic!
The only thing which doesn't work yet is enumerating failed subtests,
but I think that should work out on top of this.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-19 08:09:25 +02:00
3df8300e23
overlay: Tweak label locations to include baseline offset.
...
Still approximately, too lazy to query the metrics.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-19 08:20:15 +01:00
34e4780c8e
overlay: Add a X11 window backend
...
Useful for remote hosts.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 21:10:26 +01:00
1c1a279806
overlay: Tweak a couple of the fill colours to be slightly less horrid
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 19:21:02 +01:00
1391ae08eb
overlay: Don't smooth gpu freq
...
This is supposed to be discrete jumps, so use straight lines to emphasis
this nature.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 19:21:02 +01:00
90ef68859c
test/gem_dummy_reloc_loop: fix igt_require conversion
...
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68170
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-18 19:58:28 +02:00
5d41b13824
overlay: Remove the double ':' from object comm names
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:55:12 +01:00
6a38200b87
overlay: Use a more compact 2x2 layout for the graphs
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:51:22 +01:00
74032f4b45
overlay: Rearrange GUI state
...
Planning a new layout...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:40:03 +01:00
7d950fa6f2
overlay: Print min/max GPU frequencies.
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:27:47 +01:00
3f5129471f
overlay: Convert gem objects to megabytes
...
Less confusing, I hope.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:24:43 +01:00
e1ed5606c6
overlay: Parse gem objects
...
Condense the information and begin graphing it. Remaining todo for
memory is to measure bind/evict flux, and perhaps clflush.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 18:17:05 +01:00
de2c97b27f
overlay: Add GPU frequency
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 16:53:18 +01:00
6a64ee938b
overlay: Include CPU usage in the overview chart
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 15:57:47 +01:00
98572f0446
overlay: Count flips per plane
...
Stop the misleading double-accounting of flips when we have multiple
displays active.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-18 11:22:52 +01:00
b79a28355d
overlay: Improve the horizontal fit of the charts
...
Notably fix the fill boundary to not extend past the right hand side of
the chart.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 23:50:21 +01:00
f185f46114
overlay: Correct layout of ring/seqno in raw sample
...
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 >
2013-08-17 23:21:34 +01:00
67f533f836
overlay: Accumulate busy times
...
Still a little too course as we add multiple overlapping seqnos and
waits.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 22:33:35 +01:00
6233cac9c9
overlay: Couple wait begin/end events together to fix accounting
...
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 >
2013-08-17 22:22:21 +01:00
e1d8d774f3
overlay: Perf output redirection must be done after mmap()
...
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 >
2013-08-17 22:07:49 +01:00
eed59eac47
overlay: Show per-process wait times
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 21:14:10 +01:00
cc9de398cf
overlay: Graph per-process requests over time
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 20:32:58 +01:00
cbbd55af15
overlay: Track requests per-process
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 20:04:11 +01:00
be9937b65c
overlay: Drop unused mmap/comm events
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 18:48:28 +01:00
8cdb5bc5e8
overlay: Constify perf ringbuffer
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 18:24:39 +01:00
3e430a8b01
overlay: Fix wraparound handling of perf ringbuffer
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 18:09:40 +01:00
cc45a9a3db
overlay: Primitive integration with perf
...
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 >
2013-08-17 17:42:14 +01:00
06c0cc0bb4
overlay: Show GPU waits
...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 12:35:50 +01:00
f9a50de3dc
Introduce intel-gpu-overlay
...
A realtime display of GPU activity. Note, this is just at the point of
minimum usability...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
2013-08-17 11:21:51 +01:00
7df9caeea1
quick_dump: add is_haswell()
...
This was missed beforehand.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net >
2013-08-16 14:54:36 -07:00
0d320fdced
gem_concurrent_blit: Purge the child bufmgr's cache
...
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68169
2013-08-16 12:23:17 +01:00
6332994c26
drmtest: Squelch the rude interrupter
...
Stop printing the debugging information of how many times we emitted a
signal.
2013-08-16 12:23:00 +01:00
1b17cb9d04
gem_concurrent_blit: Fix the leak from the children.
...
As the children use the parent's fd, the kernel only has a single filp
for everyone. Therefore we cannot rely on the process termination
reaping the buffers we allocate for the children.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68169
2013-08-16 11:26:00 +01:00
1ca607b458
gem_concurrent_blit: Share total num_buffers between all children
...
Apparently not all machines have more than 4GiB of RAM available.
Spoilsports.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68169
2013-08-16 09:52:43 +01:00
6229832935
assembler: error for the wrong syntax of SEND instruction on GEN6+
...
predicate SEND execsize dst sendleadreg payload directsrcoperand instoptions
predicate SEND execsize dst sendleadreg payload imm32reg instoptions
predicate SEND execsize dst sendleadreg payload sndopr imm32reg instoptions
predicate SEND execsize dst sendleadreg payload exp directsrcoperand instoptions
The above four syntaxes are only used on legacy platforms which support implied move
from payload to dst.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com >
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com >
Signed-off-by: Ben Widawsky <ben@bwidawsk.net >
2013-08-15 14:46:57 -07:00
9d5e4fac20
gem_ctx_basic: Fix misc warnings
...
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com >
2013-08-15 16:46:58 +01:00
92aca3b19e
intel_infoframes: Dump 3D_Ext_Data when using Side-by-side (half)
...
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com >
2013-08-15 16:40:47 +01:00
bc1cf77534
intel_infoframes: Add support for decoding HDMI VICs
...
The HDMI vendor infoframe can contain a HDMI VIC (as of HDMI 1.4, only
used for 4k formats).
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com >
Reviewed-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk >
2013-08-15 16:40:02 +01:00
e58891d0ae
tests/prime: test dma-buf llseek support
...
This exercises a proposed patch from Christopher James Halse Rogers.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-15 14:01:10 +02:00
79abed393f
lib/drmtest: Use igt_assert in the do_ioctl/do_or_die helpers
...
abort results in "crash" test results in the piglit runner, igt gives
us neater output.
The downside is that we can now only use these in igt tests, but
that's already the case (and given their placement in drmtest.h their
intended use-case, too).
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-15 14:01:10 +02:00
610630622b
tests/gem_ctx_basic: Use igt_require
...
... instead of threading the 77 return value through threads. Even
better would be to check for requirements in the single-threaded setup
part of the test, since now it's done by multiple threads in parallel
and will result in interleaved and multiplied output on stdout.
Whatever.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-15 13:47:15 +02:00
176011eea7
Add intel_framebuffer_dump
...
A simple utility to capture the currently active framebuffers and record
them as a png.
2013-08-15 10:58:31 +01:00
f219889608
lib/drmtest: fix subtest output when skipping all tests
...
(null) isn't too nice ...
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-14 22:13:38 +02:00
ec834c995f
tests/prime_nv_api: fixup conversion goof-up
...
Oops.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2013-08-14 22:24:43 +02:00
Daniel Vetter