Re: [Intel-gfx] [-next PATCH 3/4] treewide: Use DEVICE_ATTR_RO

2017-12-21 Thread Sakari Ailus
Hi Joe, On Tue, Dec 19, 2017 at 10:15:08AM -0800, Joe Perches wrote: > diff --git a/drivers/staging/media/atomisp/pci/atomisp2/hmm/hmm.c > b/drivers/staging/media/atomisp/pci/atomisp2/hmm/hmm.c > index a1c81c12718c..4338b8a1309f 100644 > --- a/drivers/staging/media/atomisp/pci/atomisp2/hmm/hmm.c

Re: [Intel-gfx] [PATCH i-g-t] scripts/trace.pl: Optimize event parsing and processing

2017-12-21 Thread Tvrtko Ursulin
On 20/12/2017 23:50, John Harrison wrote: On 12/20/2017 1:54 AM, Tvrtko Ursulin wrote: What was the effect of all this on your big traces? I am only testing with a smaller one which goes from ~3.3s to ~2.2s. On a larger trace it might be non-linear gains due to double sort avoidance, unless t

[Intel-gfx] [PATCH v2] drm/syncobj: Stop reusing the same struct file for all syncobj -> fd

2017-12-21 Thread Chris Wilson
The vk cts test: dEQP-VK.api.external.semaphore.opaque_fd.export_multiple_times_temporary triggers a lot of VFS: Close: file count is 0 Dave pointed out that clearing the syncobj->file from drm_syncobj_file_release() was sufficient to silence the test, but that opens a can of worm since we assume

[Intel-gfx] [PATCH igt] igt/syncobj: Stress handle->fd/close

2017-12-21 Thread Chris Wilson
Dave found a bug where creating and closing an fd twice from the syncobj would lead to a crash. In addition to explicitly testing for a previous bug, include a stress test that covers a few more permutations of the create/close pattern to see if can uncover a race in future. Signed-off-by: Chris W

Re: [Intel-gfx] [RFC 3/4] drm/i915/perf: Extract raw GPU timestamps from OA reports

2017-12-21 Thread Sagar Arun Kamble
On 12/7/2017 1:25 AM, Lionel Landwerlin wrote: On 15/11/17 12:13, Sagar Arun Kamble wrote: From: Sourab Gupta The OA reports contain the least significant 32 bits of the gpu timestamp. This patch enables retrieval of the timestamp field from OA reports, to forward as 64 bit raw gpu timesta

Re: [Intel-gfx] [RFC 2/4] drm/i915/perf: Add support for collecting 64 bit timestamps with OA reports

2017-12-21 Thread Sagar Arun Kamble
On 12/6/2017 9:31 PM, Lionel Landwerlin wrote: On 15/11/17 12:13, Sagar Arun Kamble wrote: --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1447,6 +1447,12 @@ enum drm_i915_perf_property_id {   DRM_I915_PERF_PROP_SAMPLE_OA,     /** + * The value of this prope

Re: [Intel-gfx] [PATCH i-g-t] syncobj: add test for handle->fd + close twice.

2017-12-21 Thread Chris Wilson
Quoting Dave Airlie (2017-12-21 02:59:37) > From: Dave Airlie > > This demos a bug found in the vulkan CTS that causes > VFS: Close: file count is 0 > in dmesg, and ls /proc/self/fd to oops. > > Signed-off-by: Dave Airlie Aye catches the issue you found. Reviewed-by: Chris Wilson -Chris _

[Intel-gfx] ✓ Fi.CI.IGT: success for drm/dp: Power cycle display if LINK_ADDRESS fails.

2017-12-21 Thread Patchwork
== Series Details == Series: drm/dp: Power cycle display if LINK_ADDRESS fails. URL : https://patchwork.freedesktop.org/series/35659/ State : success == Summary == Test kms_cursor_crc: Subgroup cursor-64x64-suspend: skip -> PASS (shard-snb) fdo#102365 Test g

<    1   2