Looking at the intersection of CLs in the two runs, this CL could
be at fault:
    https://chromium-review.googlesource.com/#/c/560629/

If not, we've got a new but intermittent bug affecting reef.

This reef-paladin run also saw a similar failure:
    https://luci-milo.appspot.com/buildbot/chromeos/reef-paladin/2943

Test logs here:
    http://cautotest.corp.google.com/afe/#tab_id=view_job&object_id=128286469

The same CL was present in that failure as well.

Passing to the CL owner.

gurchetansingh@ - If the CL is causing the failures, you can close this
bug, since there's nothing wrong with the tree.

If the CL isn't causing the failure (meaning, most likely, that
you can demonstrate this failure without the CL), you can pass
it to some appropriate graphics expert for further debug.

Thanks for filing the bug jrbarnette@.  This is definitely caused by the Mesa uprev.  The failing tests are:

07/21 01:37:56.430 INFO |     graphics_dEQP:0400| Result: Fail
07/21 01:37:56.431 INFO |     graphics_dEQP:0317| [32/43] TestCase: dEQP-VK.api.smoke.asm_triangle
07/21 01:37:56.432 DEBUG|     graphics_dEQP:0343| Running single: /usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.asm_triangle --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.asm_triangle.log
07/21 01:37:56.433 DEBUG|             utils:0212| Running '/usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.asm_triangle --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.asm_triangle.log'
07/21 01:38:06.387 INFO |     graphics_dEQP:0400| Result: Fail
07/21 01:38:06.388 INFO |     graphics_dEQP:0317| [33/43] TestCase: dEQP-VK.api.smoke.asm_triangle_no_opname
07/21 01:38:06.389 DEBUG|     graphics_dEQP:0343| Running single: /usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.asm_triangle_no_opname --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.asm_triangle_no_opname.log
07/21 01:38:06.390 DEBUG|             utils:0212| Running '/usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.asm_triangle_no_opname --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.asm_triangle_no_opname.log'
07/21 01:38:16.337 INFO |     graphics_dEQP:0400| Result: Fail
07/21 01:38:16.339 INFO |     graphics_dEQP:0317| [34/43] TestCase: dEQP-VK.api.smoke.unused_resolve_attachment
07/21 01:38:16.339 DEBUG|     graphics_dEQP:0343| Running single: /usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.unused_resolve_attachment --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.unused_resolve_attachment.log
07/21 01:38:16.340 DEBUG|             utils:0212| Running '/usr/local/deqp/external/vulkancts/modules/vulkan/deqp-vk --deqp-case=dEQP-VK.api.smoke.unused_resolve_attachment --deqp-surface-type=pbuffer --deqp-gl-config-name=rgba8888d24s8ms0 --deqp-log-images=disable --deqp-watchdog=enable --deqp-surface-width=256 --deqp-surface-height=256 --deqp-log-filename=/tmp/bvt.txt-logs/dEQP-VK.api.smoke.unused_resolve_attachment.log'
07/21 01:38:24.395 INFO |     graphics_dEQP:0400| Result: Fail

We could just blacklist these specific tests temporarily, so we can land the much delayed Mesa uprev.  Perhaps our resident Vulkan expert chadversary@ has an easy fix [if you do Chad :-), you can lock and test on a reef here:

http://cautotest

For example, 

ssh root@chromeos2-row7-rack6-host17.cros (password is the same as usual)]

Also, does Intel run these Vulkan tests continuously on ToT Mesa -- it would be good if they did.

> ssh root@chromeos2-row7-rack6-host17.cros (password is the same as usual)]

Please don't use that specific host.  It's reserved for CQ testing.
Also, it's preferable not to lock devices at all; use `test_that :lab:`
(see test_that --help).

If it's necessary to borrow a DUT by locking it:
  * Borrow devices in state "Ready", with the label "pool:suites".
  * Don't forget to unlock when you're done.
  * Don't leave the device locked overnight.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/6a03236cb77acde486ecfe3391ccbd289e3736a6

commit 6a03236cb77acde486ecfe3391ccbd289e3736a6
Author: Gurchetan Singh <gurchetansingh@chromium.org>
Date: Mon Jul 24 21:36:09 2017

graphics_dEQP: remove failing Vulkan tests for now

The following graphics_dEQP Vulkan tests:

dEQP-VK.api.smoke.triangle
dEQP-VK.api.smoke.asm_triangle
dEQP-VK.api.smoke.asm_triangle_no_opname
dEQP-VK.api.smoke.unused_resolve_attachment

fail on reef (Broxton). Rather than bisect, let's disable them
for now so we can land the Mesa uprev [Vulkan is still in the
experimental stages, after all].

BUG=chromium:747628
TEST=none

Change-Id: I174cdc4f28afe84c13619ee838f7b057b37c4372
Reviewed-on: https://chromium-review.googlesource.com/583932
Reviewed-by: Ilja H. Friedel <ihf@chromium.org>
Tested-by: Ilja H. Friedel <ihf@chromium.org>

[modify] https://crrev.com/6a03236cb77acde486ecfe3391ccbd289e3736a6/client/site_tests/graphics_dEQP/master/bvt.txt

Chad is looking into fixing Vulkan dEQP in general, and this looks a subset of that task.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/bc5fa6dbf4862c90beba1a6a84f9c3be286b9ee7

commit bc5fa6dbf4862c90beba1a6a84f9c3be286b9ee7
Author: Deepak Sharma <Deepak.Sharma@amd.com>
Date: Sun Oct 29 06:04:02 2017

graphics_dEQP: remove failing Vulkan test

dEQP-VK.api.info.device.extensions fails on intel platform.
disabling it to land Mesa uprev.

BUG=b:68165024,chromium:747628
Test=None

Change-Id: Ic386d4fdb92693c0ae5b37eead8ea6d3bd1af999
Signed-off-by: Deepak Sharma <Deepak.Sharma@amd.com>
Reviewed-on: https://chromium-review.googlesource.com/742121
Commit-Ready: Deepak Sharma <deepak.sharma@amd.com>
Tested-by: Deepak Sharma <deepak.sharma@amd.com>
Reviewed-by: Ilja H. Friedel <ihf@chromium.org>
Reviewed-by: Chad Versace <chadversary@chromium.org>

[modify] https://crrev.com/bc5fa6dbf4862c90beba1a6a84f9c3be286b9ee7/client/site_tests/graphics_dEQP/master/bvt.txt