ReferrerSanitizerTest.SanitizesPolicyForNonEmptyReferrers flaked on Fuchsia in content_unittests |
||
Issue descriptionLooks like it might have crashed on a gpu thing rather than the expected crash? I'm not sure. It's only happened on the dbg fyi for now so I'll let it go for the time being to see if it re-occurs. https://logs.chromium.org/v/?s=chromium%2Fbuildbucket%2Fcr-buildbucket.appspot.com%2F8935398978524198176%2F%2B%2Fsteps%2Fcontent_unittests%2F0%2Fstdout [3108/5603] FrameSwapMessageQueueTest.TestNextSwapMessageSentWithCurrentFrame (0 ms) [3109/5603] FrameSwapMessageQueueTest.TestDrainsVisualStateMessagesForCorrespondingFrames (5 ms) [3110/5603] FrameSwapMessageQueueTest.TestQueueNextSwapMessagePreservesFifoOrdering (0 ms) [00543.309] pkgsvr: 2018/09/14 15:42:01 pkgfs:unsupported(/packages/content_unittests/0): dir unlink "content_unittests__exec.log" [00543.417] dlsvc: could not open 'libGLES_CM.so' [00543.422] dlsvc: could not open 'libGLES_CM.so' [00543.428] dlsvc: could not open 'libGLES_CM.so' [00543.433] dlsvc: could not open 'libGLES_CM.so' [00543.438] dlsvc: could not open 'libGLES_CM.so' [00543.442] dlsvc: could not open 'libGLES_CM.so' [00543.448] netstack: ioctlNetcGetNodename: error accessing device settings: ErrPeerClosed: zx.Channel.Call [00543.448] netstack: ioctlNetcGetNodename: falling back to default nodename. [00543.448] netstack: ioctlNetcGetNodename: device settings error: [00543.467] netstack: ioctlNetcGetNodename: error accessing device settings: ErrPeerClosed: zx.Channel.Call [00543.474] netstack: ioctlNetcGetNodename: falling back to default nodename. [00543.474] netstack: ioctlNetcGetNodename: device settings error: [ RUN ] ReferrerSanitizerTest.SanitizesPolicyForNonEmptyReferrers ../../content/public/test/referrer_unittest.cc:23: Failure Death test: ignore_result(Referrer::SanitizeForRequest( GURL("https://a"), Referrer(GURL("http://b"), static_cast<blink::WebReferrerPolicy>(200)))) Result: died but not with expected error. Expected: Check failed Actual msg: [ DEATH ] [1355971:1388343703:0914/154158.958321:540425052:WARNING:test_suite.cc(258)] Test launcher output path /tmp/.org.chromium.Chromium.kibelK/test_results.xml exists. Not adding test launcher result printer. [ DEATH ] [1355971:1388343703:0914/154159.016740:540483469:ERROR:gpu_info_collector_fuchsia.cc(22)] Not implemented reached in bool gpu::CollectBasicGraphicsInfo(gpu::GPUInfo *) [ DEATH ] Stack trace: #00: pc 0x1491b70a2549 (libwebgpu.so,0x1186e3efe549) #01: pc 0x1491b70c1367 (libwebgpu.so,0x1186e3f1d367) #02: pc 0x1491b70c0707 (libwebgpu.so,0x1186e3f1c707) #03: pc 0x1491b668aa63 (libwebgpu.so,0x1186e34e6a63) #04: pc 0x1491b5badb0e (libwebgpu.so,0x1186e2a09b0e) [ FAILED ] ReferrerSanitizerTest.SanitizesPolicyForNonEmptyReferrers (1877 ms) [3111/5603] ReferrerSanitizerTest.SanitizesPolicyForNonEmptyReferrers (1877 ms)
,
Sep 17
fyi +mblsha +jochen as author/recent toucher nearby.
,
Sep 17
looks like an unrelated check triggering in gpu_info_collector_fuchsia.cc ?
,
Sep 17
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/ba649d6e4cdc403ae6ea3ad1abb369c386f24ce9 commit ba649d6e4cdc403ae6ea3ad1abb369c386f24ce9 Author: Scott Graham <scottmg@chromium.org> Date: Mon Sep 17 17:36:54 2018 fuchsia: Disable ReferrerSanitizerTest.SanitizesPolicyForNonEmptyReferrers Bug: 884250 Change-Id: Iec6f3bb34e0c26a7602ddf08ff79b26104206ee4 Reviewed-on: https://chromium-review.googlesource.com/1228499 Commit-Queue: Fabrice de Gans-Riberi <fdegans@chromium.org> Reviewed-by: Fabrice de Gans-Riberi <fdegans@chromium.org> Cr-Commit-Position: refs/heads/master@{#591724} [modify] https://crrev.com/ba649d6e4cdc403ae6ea3ad1abb369c386f24ce9/testing/buildbot/filters/fuchsia.content_unittests.filter
,
Sep 17
I thought that too at first, but I think it's https://cs.chromium.org/chromium/src/gpu/config/gpu_info_collector_fuchsia.cc?rcl=591031b8ef75c36cb929d1e3e1ef401fcaaf799a&l=22 just spam https://cs.chromium.org/chromium/src/base/logging.h?rcl=591031b8ef75c36cb929d1e3e1ef401fcaaf799a&l=1161 not the actual cause. I could be mistaken though. |
||
►
Sign in to add a comment |
||
Comment 1 by scottmg@chromium.org
, Sep 17