1865921 - Resolve the WebGPU client device.lost promise using the `device_lost_closure` in wgpu

Assignee

Description

•

11 months ago

Now that wgpu allows a closure to be specified which is triggered on "lose the device", WebGPUParent should supply that closure and inform the client when it is triggered. Since there is already a WebGPUParent::LoseDevice method, there's no need to define a new message between the parent and the child. Ideally, the connection just needs to be made between the parent and the wgpu bindings.

Brad Werth [:bradwerth]

Assignee

Updated

•

11 months ago

Assignee: nobody → bwerth

Brad Werth [:bradwerth]

Assignee

Comment 1

•

11 months ago

Attached file Bug 1865921 Part 1: Make WebGPUParent supply a device lost callback. — Details

Brad Werth [:bradwerth]

Assignee

Comment 2

•

11 months ago

Attached file Bug 1865921 Part 2: Add a test that awaits the device.lost promise. — Details

Depends on D194310

Brad Werth [:bradwerth]

Assignee

Comment 3

•

11 months ago

This is dependent on wgpu upstream fixes in https://github.com/gfx-rs/wgpu/pull/4744.

Brad Werth [:bradwerth]

Assignee

Comment 4

•

11 months ago

(In reply to Brad Werth [:bradwerth] from comment #3)

This is dependent on wgpu upstream fixes in https://github.com/gfx-rs/wgpu/pull/4744.

This has been merged, so now this is blocked on the next vendoring of wgpu. I'll mark the blocking Bug when it is filed.

Erich Gubler [:ErichDonGubler]

Updated

•

10 months ago

Blocks: webgpu-v1

Severity: -- → S3

Priority: -- → P1

Brad Werth [:bradwerth]

Assignee

Updated

•

10 months ago

Depends on: 1868143

Pulsebot

Comment 5

•

10 months ago

Pushed by bwerth@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/95b9cef93a5c Part 1: Make WebGPUParent supply a device lost callback. r=ErichDonGubler https://hg.mozilla.org/integration/autoland/rev/e378a6744fb2 Part 2: Add a test that awaits the device.lost promise. r=ErichDonGubler

Narcis Beleuzu [:NarcisB]

Comment 6

•

10 months ago

Backed out for causing webgpu leaks

Backout link: https://hg.mozilla.org/integration/autoland/rev/f778448a1da5f0622c4361d37810461a035f4627
Log link: https://treeherder.mozilla.org/logviewer?job_id=438938413&repo=autoland&lineNumber=3300

Please also check this TV failure

Flags: needinfo?(bwerth)

Brad Werth [:bradwerth]

Assignee

Comment 7

•

10 months ago

Seems like the problem is that the callback is not guaranteed to be called under all conditions, and the callback holds a RefPtr to the WebGPUParent, preventing it from being deallocated until the callback is invoked. One such case would be the very common case where a device is created, but never explicitly destroyed. When GC happens, that will also need to invoke the lost callback. This will require a change in wgpu.

Flags: needinfo?(bwerth)

Brad Werth [:bradwerth]

Assignee

Comment 8

•

10 months ago

(In reply to Brad Werth [:bradwerth] from comment #7)

Seems like the problem is that the callback is not guaranteed to be called under all conditions, and the callback holds a RefPtr to the WebGPUParent, preventing it from being deallocated until the callback is invoked. One such case would be the very common case where a device is created, but never explicitly destroyed. When GC happens, that will also need to invoke the lost callback. This will require a change in wgpu.

Tracked in https://github.com/gfx-rs/wgpu/pull/4835

Brad Werth [:bradwerth]

Assignee

Comment 9

•

10 months ago

(In reply to Brad Werth [:bradwerth] from comment #8)

Tracked in https://github.com/gfx-rs/wgpu/pull/4835

I removed this because it's out of compliance with the WebGPU spec. I have a Firefox-only fix in Part 1 that I'm testing now.

Brad Werth [:bradwerth]

Assignee

Comment 10

•

9 months ago

There's some tricky stuff we need to do to get this right. Firstly, wgpu will need another update to ensure that its asserts for DeviceLostCallbackC::Drop are not triggered after a closure is passed to wgpu_server_set_device_lost_callback due to the callback being dropped before it is called -- since the data is copied out of it and the original is dropped.

Secondly, the behavior of WebGPUChild::UnregisterDevice needs a re-think. That function calls SendDeviceDrop and then immediately forgets about the device. When the parent does the drop, which triggers the device lost callback, then informs the child, the child doesn't know about the device anymore. In such a case, anything waiting on the device.lost promise will wait forever.

Solvable problems, but they'll need to be solved before this can be landed. I'm on it.

Brad Werth [:bradwerth]

Assignee

Comment 11

•

9 months ago

The necessary wgpu parts are in https://github.com/gfx-rs/wgpu/pull/5032.

Brad Werth [:bradwerth]

Assignee

Updated

•

9 months ago

Depends on: 1875543
No longer depends on: 1868143

Pulsebot

Comment 12

•

8 months ago

Pushed by bwerth@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/b0f89dfa3291 Part 1: Make WebGPUParent supply a device lost callback. r=ErichDonGubler,webgpu-reviewers,nical https://hg.mozilla.org/integration/autoland/rev/39cf0feca383 Part 2: Add a test that awaits the device.lost promise. r=ErichDonGubler

Natalia Csoregi [:nataliaCs]

Comment 13

•

8 months ago

Backed out for causing failures on test_device_lost.html .

[task 2024-01-24T01:13:27.810Z] 01:13:27     INFO - TEST-PASS | dom/webgpu/mochitest/test_device_lost.html | Pref should be enabled. 
[task 2024-01-24T01:13:27.810Z] 01:13:27     INFO - Buffered messages finished
[task 2024-01-24T01:13:27.812Z] 01:13:27     INFO - TEST-UNEXPECTED-PASS | dom/webgpu/mochitest/test_device_lost.html | fail-if condition in manifest - We expected at least one failure
[task 2024-01-24T01:13:27.812Z] 01:13:27     INFO - TEST-INFO | expected FAIL
[task 2024-01-24T01:13:27.825Z] 01:13:27     INFO - GECKO(1654) | MEMORY STAT | vsize 2583MB | residentFast 139MB | heapAllocated 7MB
[task 2024-01-24T01:13:27.832Z] 01:13:27     INFO - GECKO(1654) | Validation error without device target: No suitable adapter found
[task 2024-01-24T01:13:27.834Z] 01:13:27     INFO - GECKO(1654) | Validation error without device target: No suitable adapter found
[task 2024-01-24T01:13:27.846Z] 01:13:27     INFO - TEST-PASS | dom/webgpu/mochitest/test_device_lost.html | Unhandled exception TypeError: can't access property "limits", adapter is null

Flags: needinfo?(bwerth)

Pulsebot

Comment 14

•

8 months ago

Pushed by bwerth@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/dbb08bccb5e3 Part 1: Make WebGPUParent supply a device lost callback. r=ErichDonGubler,webgpu-reviewers,nical https://hg.mozilla.org/integration/autoland/rev/46b5a7e46018 Part 2: Add a test that awaits the device.lost promise. r=ErichDonGubler

Treeherder Bug Filer

Updated

•

8 months ago

Regressions: 1876655

Treeherder Bug Filer

Updated

•

8 months ago

Regressions: 1876663

Iulian Moraru

Comment 15

•

8 months ago

bugherder

https://hg.mozilla.org/mozilla-central/rev/dbb08bccb5e3
https://hg.mozilla.org/mozilla-central/rev/46b5a7e46018

Status: NEW → RESOLVED

Closed: 8 months ago

status-firefox124: --- → fixed

Resolution: --- → FIXED

Target Milestone: --- → 124 Branch

Iulian Moraru

Comment 16

•

8 months ago

Backed out for causing Bug 1876663.

[task 2024-01-26T01:21:38.090Z] 01:21:38     INFO - TEST-START | dom/webgpu/mochitest/test_basic_canvas.worker.html
[task 2024-01-26T01:21:38.404Z] 01:21:38     INFO - GECKO(10148) | ### XPCOM_MEM_BLOAT_LOG defined -- logging bloat/leaks to C:\Users\task_170622997413941\AppData\Local\Temp\tmpzh4dcing.mozrunner\runtests_leaks_tab_pid5736.log
[task 2024-01-26T01:21:40.140Z] 01:21:40     INFO - GECKO(10148) | [WARN  wgpu_core::instance] Missing downlevel flags: DownlevelFlags(VERTEX_AND_INSTANCE_INDEX_RESPECTS_RESPECTIVE_FIRST_VALUE_IN_INDIRECT_DRAW)
[task 2024-01-26T01:21:40.141Z] 01:21:40     INFO - GECKO(10148) |     The underlying API or device in use does not support enough features to be a fully compliant implementation of WebGPU. A subset of the features can still be used. If you are running this program on native and not in a browser and wish to limit the features you use to the supported subset, call Adapter::downlevel_properties or Device::downlevel_properties to get a listing of the features the current platform supports.
[task 2024-01-26T01:21:40.142Z] 01:21:40     INFO - GECKO(10148) | [WARN  wgpu_core::instance] DownlevelCapabilities {
[task 2024-01-26T01:21:40.143Z] 01:21:40     INFO - GECKO(10148) |         flags: DownlevelFlags(
[task 2024-01-26T01:21:40.144Z] 01:21:40     INFO - GECKO(10148) |             COMPUTE_SHADERS | FRAGMENT_WRITABLE_STORAGE | INDIRECT_EXECUTION | BASE_VERTEX | READ_ONLY_DEPTH_STENCIL | NON_POWER_OF_TWO_MIPMAPPED_TEXTURES | CUBE_ARRAY_TEXTURES | COMPARISON_SAMPLERS | INDEPENDENT_BLEND | VERTEX_STORAGE | ANISOTROPIC_FILTERING | FRAGMENT_STORAGE | MULTISAMPLED_SHADING | DEPTH_TEXTURE_AND_BUFFER_COPIES | WEBGPU_TEXTURE_FORMAT_SUPPORT | BUFFER_BINDINGS_NOT_16_BYTE_ALIGNED | UNRESTRICTED_INDEX_BUFFER | FULL_DRAW_INDEX_UINT32 | DEPTH_BIAS_CLAMP | VIEW_FORMATS | UNRESTRICTED_EXTERNAL_TEXTURE_COPIES | SURFACE_VIEW_FORMATS | NONBLOCKING_QUERY_RESOLVE,
[task 2024-01-26T01:21:40.144Z] 01:21:40     INFO - GECKO(10148) |         ),
[task 2024-01-26T01:21:40.145Z] 01:21:40     INFO - GECKO(10148) |         limits: DownlevelLimits,
[task 2024-01-26T01:21:40.145Z] 01:21:40     INFO - GECKO(10148) |         shader_model: Sm5,
[task 2024-01-26T01:21:40.145Z] 01:21:40     INFO - GECKO(10148) |     }
[task 2024-01-26T01:21:40.210Z] 01:21:40     INFO - GECKO(10148) | MEMORY STAT | vsize 574MB | vsizeMaxContiguous 1806MB | residentFast 71MB | heapAllocated 7MB
[task 2024-01-26T01:21:40.287Z] 01:21:40     INFO - GECKO(10148) | Assertion failure: mErrorScopeStackByDevice.empty() (All Devices should have been dropped.), at /builds/worker/checkouts/gecko/dom/webgpu/ipc/WebGPUParent.cpp:328
[task 2024-01-26T01:21:40.306Z] 01:21:40     INFO - TEST-OK | dom/webgpu/mochitest/test_basic_canvas.worker.html | took 2216ms
[task 2024-01-26T01:21:40.383Z] 01:21:40     INFO - TEST-START | dom/webgpu/mochitest/test_buffer_mapping.html
[task 2024-01-26T01:21:40.527Z] 01:21:40     INFO -  Initializing stack-fixing for the first stack frame, this may take a while...
[task 2024-01-26T01:22:00.420Z] 01:22:00     INFO - GECKO(10148) | #01: mozilla::webgpu::WebGPUParent::~WebGPUParent() [dom/webgpu/ipc/WebGPUParent.cpp:327]
[task 2024-01-26T01:22:00.420Z] 01:22:00     INFO - GECKO(10148) | #02: mozilla::webgpu::WebGPUParent::Release() [dom/webgpu/ipc/WebGPUParent.h:45]
[task 2024-01-26T01:22:00.421Z] 01:22:00     INFO - GECKO(10148) | #03: mozilla::webgpu::PWebGPUParent::ActorDealloc() [s3:gecko-generated-sources:aeef733bfca21d1ebba4267342ba726d6492564e508931b412ddf8e12d05ac1090ca2704aee69688f137c0c12eb4b5525d56d369b2294cb4aa325430c0f2a0d7/ipc/ipdl/PWebGPUParent.cpp::51]
[task 2024-01-26T01:22:00.422Z] 01:22:00     INFO - GECKO(10148) | #04: mozilla::ipc::ActorLifecycleProxy::~ActorLifecycleProxy() [ipc/glue/ProtocolUtils.cpp:300]
[task 2024-01-26T01:22:00.423Z] 01:22:00     INFO - GECKO(10148) | #05: mozilla::ipc::ActorLifecycleProxy::Release() [ipc/glue/ProtocolUtils.h:652]
[task 2024-01-26T01:22:00.424Z] 01:22:00     INFO - GECKO(10148) | #06: mozilla::gfx::PCanvasManagerParent::ClearSubtree() [s3:gecko-generated-sources:f0f92794d59ce211f13900af1347d7644d03d6aefb2c3506c31d5cf7d17cdaf4d0ed23d4efbb1ac569448c252819bfd6ac27d880567c4125b46fe97a40f86e97/ipc/ipdl/PCanvasManagerParent.cpp::652]
[task 2024-01-26T01:22:00.425Z] 01:22:00     INFO - GECKO(10148) | #07: mozilla::gfx::PCanvasManagerParent::OnChannelClose() [s3:gecko-generated-sources:f0f92794d59ce211f13900af1347d7644d03d6aefb2c3506c31d5cf7d17cdaf4d0ed23d4efbb1ac569448c252819bfd6ac27d880567c4125b46fe97a40f86e97/ipc/ipdl/PCanvasManagerParent.cpp::608]
[task 2024-01-26T01:22:00.426Z] 01:22:00     INFO - GECKO(10148) | #08: mozilla::ipc::MessageChannel::NotifyChannelClosed(mozilla::ReleasableMonitorAutoLock&) [ipc/glue/MessageChannel.cpp:2215]
[task 2024-01-26T01:22:00.426Z] 01:22:00     INFO - GECKO(10148) | #09: mozilla::ipc::MessageChannel::NotifyMaybeChannelError(mozilla::ReleasableMonitorAutoLock&) [ipc/glue/MessageChannel.cpp:2064]
[task 2024-01-26T01:22:00.427Z] 01:22:00     INFO - GECKO(10148) | #10: mozilla::ipc::MessageChannel::OnNotifyMaybeChannelError() [ipc/glue/MessageChannel.cpp:2090]
[task 2024-01-26T01:22:00.427Z] 01:22:00     INFO - GECKO(10148) | #11: mozilla::detail::RunnableMethodImpl<mozilla::ipc::MessageChannel *,void (mozilla::ipc::MessageChannel::*)() __attribute__((thiscall)),0,1>::Run() [xpcom/threads/nsThreadUtils.h:1216]
[task 2024-01-26T01:22:00.428Z] 01:22:00     INFO - GECKO(10148) | #12: nsThread::ProcessNextEvent(bool, bool*) [xpcom/threads/nsThread.cpp:1194]
[task 2024-01-26T01:22:00.428Z] 01:22:00     INFO - GECKO(10148) | #13: NS_ProcessNextEvent(nsIThread*, bool) [xpcom/threads/nsThreadUtils.cpp:480]
[task 2024-01-26T01:22:00.429Z] 01:22:00     INFO - GECKO(10148) | #14: mozilla::ipc::MessagePumpForNonMainThreads::Run(base::MessagePump::Delegate*) [ipc/glue/MessagePump.cpp:330]
[task 2024-01-26T01:22:00.429Z] 01:22:00     INFO - GECKO(10148) | #15: MessageLoop::RunInternal() [ipc/chromium/src/base/message_loop.cc:370]
[task 2024-01-26T01:22:00.430Z] 01:22:00     INFO - GECKO(10148) | #16: MessageLoop::Run() [ipc/chromium/src/base/message_loop.cc:346]
[task 2024-01-26T01:22:00.431Z] 01:22:00     INFO - GECKO(10148) | #17: nsThread::ThreadFunc(void*) [xpcom/threads/nsThread.cpp:372]
[task 2024-01-26T01:22:00.647Z] 01:22:00     INFO - GECKO(10148) | #18: _PR_NativeRunThread(void*) [nsprpub/pr/src/threads/combined/pruthr.c:399]
[task 2024-01-26T01:22:00.648Z] 01:22:00     INFO - GECKO(10148) | #19: pr_root(void*) [nsprpub/pr/src/md/windows/w95thred.c:139]
[task 2024-01-26T01:22:00.651Z] 01:22:00     INFO -  fix-stacks: error: failed to read debug info file `ucrtbase.pdb` for `C:\Windows\System32\ucrtbase.dll`
[task 2024-01-26T01:22:00.652Z] 01:22:00     INFO -  fix-stacks: note: this is expected and harmless for all PDB files on opt automation runs
[task 2024-01-26T01:22:00.652Z] 01:22:00     INFO -  fix-stacks: The system cannot find the file specified. (os error 2)
[task 2024-01-26T01:22:00.653Z] 01:22:00     INFO - GECKO(10148) | #20: time32 [C:\Windows\System32\ucrtbase.dll + 0x459d3]
[task 2024-01-26T01:22:00.656Z] 01:22:00     INFO -  fix-stacks: error: failed to read debug info file `wkernel32.pdb` for `C:\Windows\System32\KERNEL32.DLL`
[task 2024-01-26T01:22:00.656Z] 01:22:00     INFO -  fix-stacks: note: this is expected and harmless for all PDB files on opt automation runs
[task 2024-01-26T01:22:00.657Z] 01:22:00     INFO -  fix-stacks: The system cannot find the file specified. (os error 2)
[task 2024-01-26T01:22:00.657Z] 01:22:00     INFO - GECKO(10148) | #21: BaseThreadInitThunk [C:\Windows\System32\KERNEL32.DLL + 0x17ba9]
[task 2024-01-26T01:22:00.831Z] 01:22:00     INFO - GECKO(10148) | #22: Z23DllBlocklist_Initializej [Z:\task_170622997413941\build\application\firefox\mozglue.dll + 0x8222c]
[task 2024-01-26T01:22:00.835Z] 01:22:00     INFO -  fix-stacks: error: failed to read debug info file `wntdll.pdb` for `C:\Windows\SYSTEM32\ntdll.dll`
[task 2024-01-26T01:22:00.835Z] 01:22:00     INFO -  fix-stacks: note: this is expected and harmless for all PDB files on opt automation runs
[task 2024-01-26T01:22:00.835Z] 01:22:00     INFO -  fix-stacks: The system cannot find the file specified. (os error 2)
[task 2024-01-26T01:22:00.836Z] 01:22:00     INFO - GECKO(10148) | #23: RtlInitializeExceptionChain [C:\Windows\SYSTEM32\ntdll.dll + 0x6bd3b]
[task 2024-01-26T01:22:00.837Z] 01:22:00     INFO - GECKO(10148) | #24: RtlClearBits [C:\Windows\SYSTEM32\ntdll.dll + 0x6bcbf]
[task 2024-01-26T01:22:00.837Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.838Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.839Z] 01:22:00     INFO - GECKO(10148) | [GFX1-]: CompositorBridgeChild receives IPC close with reason=Ab[Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.840Z] 01:22:00     INFO - GECKO(10148) | normalShutdown
[task 2024-01-26T01:22:00.840Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, Main Thread] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.841Z] 01:22:00     INFO - GECKO(10148) | [GFX1-]: CompositorBridgeChild receives IPC close with reason=AbnormalShutdown
[task 2024-01-26T01:22:00.841Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.842Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.842Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.843Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Ignoring message 'EVENT_MESSAGE' to unknown peer 1C2DFF274C0D9208.4DBB95D1E38B28F1: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:344
[task 2024-01-26T01:22:00.844Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: GetUserData call for port 'FD6D27EE3C80E88A.7C2E4CB430F818B2' failed: file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:406
[task 2024-01-26T01:22:00.846Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from 'DC1B4E7F60D5FBA6.3A7F8B7CBC94BA9D' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.846Z] 01:22:00     INFO - GECKO(10148) | [GFX1-]: Compos[[GFiGXFto1-X]1-: rC]B: oComrmipposoistiotrodBgerrBiCrhiidldgde[Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from 'BCDA589BF1A262B8.1C1FE52422002A58' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.846Z] 01:22:00     INFO - GECKO(10148) | geC hCrhieicellidvd  rers eecIceeiPiCv ecsv eIlsoP sIPCC e  clcwlioostsee h  wwrieitha s[GFoX1-]n[Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from '4D6289E7E1C3B8BB.EC7A6E5A6447E576' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.847Z] 01:22:00     INFO - GECKO(10148) | =rAet:h  absrnCeoon=Aoabsornmoarmposinl=AtmoalbnoSrmSrahlBhSurthdutoiduowwdtnnd
[task 2024-01-26T01:22:00.848Z] 01:22:00     INFO - GECKO(10148) | gowe
[task 2024-01-26T01:22:00.848Z] 01:22:00     INFO - GECKO(10148) | n
[task 2024-01-26T01:22:00.849Z] 01:22:00     INFO - GECKO(10148) | Child receives IPC close with rea[Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from 'BD012D446632F07C.E5C486BAC3260992' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.849Z] 01:22:00     INFO - GECKO(10148) | son=AbnormalShutdown
[task 2024-01-26T01:22:00.850Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from '41BE41D4EE96F9F4.60F6E30A6638339D' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.851Z] 01:22:00     INFO - GECKO(10148) | [Parent 10816, IPC I/O Parent] WARNING: [1.1]: Rejecting introduction request from 'E81D7CC11565F987.70C009AAB62894BA' for unknown peer '1C2DFF274C0D9208.4DBB95D1E38B28F1': file /builds/worker/checkouts/gecko/ipc/glue/NodeController.cpp:676
[task 2024-01-26T01:22:00.851Z] 01:22:00     INFO - TEST-INFO | started process screenshot
[task 2024-01-26T01:22:01.095Z] 01:22:01     INFO - TEST-INFO | screenshot: exit 0
[task 2024-01-26T01:22:01.097Z] 01:22:01     INFO - Buffered messages logged at 01:21:40
[task 2024-01-26T01:22:01.097Z] 01:22:01     INFO - TEST-PASS | dom/webgpu/mochitest/test_buffer_mapping.html | Pref should be enabled. 
[task 2024-01-26T01:22:01.097Z] 01:22:01     INFO - Buffered messages finished
[task 2024-01-26T01:22:01.098Z] 01:22:01     INFO - TEST-UNEXPECTED-FAIL | dom/webgpu/mochitest/test_buffer_mapping.html | Unhandled exception AbortError: Internal communication error!

Status: RESOLVED → REOPENED

status-firefox124: fixed → ---

Flags: needinfo?(bwerth)

Resolution: FIXED → ---

Target Milestone: 124 Branch → ---

Iulian Moraru

Updated

•

8 months ago

Flags: needinfo?(bwerth)

Pulsebot

Comment 17

•

8 months ago

Pushed by bwerth@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/39c97da20626 Part 1: Make WebGPUParent supply a device lost callback. r=ErichDonGubler,webgpu-reviewers,nical https://hg.mozilla.org/integration/autoland/rev/26f908c2a616 Part 2: Add a test that awaits the device.lost promise. r=ErichDonGubler

Narcis Beleuzu [:NarcisB]

Comment 18

•

8 months ago

bugherder

https://hg.mozilla.org/mozilla-central/rev/39c97da20626
https://hg.mozilla.org/mozilla-central/rev/26f908c2a616

Status: REOPENED → RESOLVED

Closed: 8 months ago → 8 months ago

status-firefox124: --- → fixed

Resolution: --- → FIXED

Target Milestone: --- → 124 Branch

Brad Werth [:bradwerth]

Assignee

Updated

•

8 months ago

Flags: needinfo?(bwerth)

Erich Gubler [:ErichDonGubler]

Updated

•

8 months ago

Regressions: 1877046

Alex Finder

Comment 19

•

8 months ago

(In reply to Natalia Csoregi [:nataliaCs] from comment #13)

Backed out for causing failures on test_device_lost.html .

[task 2024-01-24T01:13:27.810Z] 01:13:27     INFO - TEST-PASS | dom/webgpu/mochitest/test_device_lost.html | Pref should be enabled. 
[task 2024-01-24T01:13:27.810Z] 01:13:27     INFO - Buffered messages finished
[task 2024-01-24T01:13:27.812Z] 01:13:27     INFO - TEST-UNEXPECTED-PASS | dom/webgpu/mochitest/test_device_lost.html | fail-if condition in manifest - We expected at least one failure
[task 2024-01-24T01:13:27.812Z] 01:13:27     INFO - TEST-INFO | expected FAIL
[task 2024-01-24T01:13:27.825Z] 01:13:27     INFO - GECKO(1654) | MEMORY STAT | vsize 2583MB | residentFast 139MB | heapAllocated 7MB
[task 2024-01-24T01:13:27.832Z] 01:13:27     INFO - GECKO(1654) | Validation error without device target: No suitable adapter found
[task 2024-01-24T01:13:27.834Z] 01:13:27     INFO - GECKO(1654) | Validation error without device target: No suitable adapter found
[task 2024-01-24T01:13:27.846Z] 01:13:27     INFO - TEST-PASS | dom/webgpu/mochitest/test_device_lost.html | Unhandled exception TypeError: can't access property "limits", adapter is null

== Change summary for alert #41228 (as of Tue, 30 Jan 2024 10:31:50 GMT) ==

Improvements:

Ratio	Test	Platform	Options	Absolute values (old vs new)	Performance Profiles
8%	office loadtime	windows10-64-shippable-qr	cold fission webrender	730.08 -> 669.44	Before/After

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=41228

Keywords: perf-alert

Tyson Smith [:tsmith]

Updated

•

5 months ago

Regressions: 1896195

Lee Salzman [:lsalzman]

Updated

•

5 months ago

No longer regressions: 1896195

Lee Salzman [:lsalzman]

Updated

•

5 months ago

Regressions: 1896195

Bug 1865921 Part 1: Make WebGPUParent supply a device lost callback. 11 months ago Brad Werth [:bradwerth] 48 bytes, text/x-phabricator-request		Details \| Review
Bug 1865921 Part 2: Add a test that awaits the device.lost promise. 11 months ago Brad Werth [:bradwerth] 48 bytes, text/x-phabricator-request		Details \| Review