It has been supersede by intel_error_decode for the user-facing role of
obtaining debug info after a crash.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The maximum filename is limited so we can simply use a large buffer
rather than fail to check the error return from asprintf.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Stops the compiler warning about not checking the potential error return
from asprintf, which was a false positive anyway.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Include a simple series of blits that exhaust the aperture but have the
maximum grace time between reuse.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Include a simple series of blits that exhaust the aperture but have the
maximum grace time between reuse.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This is vital in a multi-GPU system so that we only test the Intel card
and not the discrete GPUs.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
After full-gtt, gem_tiled_blits doesn't allocate enough to force
eviction. So query the total aperture and accommodate.
Also introduce a similar test that utilizes fences rather than
use the BLT to perform the tiling and detiling.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Using for diagnosing some mysterious slowdowns. Should include a variant
for basic benchmarking...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Some of our device class defines were incorrect, and we were missing
several recent ones, so fix those up and dump more child device data if
we have it.
This is something I sometimes want to do in testing, to see if a
mystery bug (say, 29172) is due to broken clock gating. Sadly, in
this case it isn't. Note that it isn't supported on non-ILK chipsets
yet.
Useful for feeding in compressed files from bugzilla:
$ bzcat /tmp/i915_error_state.bz | intel_error_decode | less
Next step would be to use gzfopen or bzfopen to automagically handle
compressed files...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Actually, fix to work on _any_ machine where the ROM file doesn't match
the GPU. Just extract the device ID from the ROM itself rather than
look at the running system.
Signed-off-by: Adam Jackson <ajax@redhat.com>
After bf79cb914dbfe848add8bb76cbb8ff89110d29ff, drm uses ENOENT to
report unknown handles buffer objects, update the tests accordingly.
Fixes:
Bug 29794 - some intel_gpu_tools cases fail
https://bugs.freedesktop.org/show_bug.cgi?id=29794
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
