)]}' { "log": [ { "commit": "7e8475abedbde368b0eebf30871bf02307ec80d0", "tree": "88784cd68e59755393779c434dc381386c8e4a71", "parents": [ "e60344418093002c18384c1f4e2ce8e4fb612268" ], "author": { "name": "seannli", "email": "seannli@google.com", "time": "Wed Jul 29 13:18:40 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 13:22:51 2026" }, "message": "camera: Snapshot Camera3JpegBlob header before validation\n\nThe JPEG blob buffer received from the ChromeOS camera HAL is a shared\nmapping that the HAL may still be writing while SubmitCapturedJpegBuffer\nreads from it. Copy the trailer struct to a local once so that the\nbounds check and the value passed to the blobify callback always observe\nthe same `jpeg_size`.\n\nBug: 517001763\nTest: Build successfully\nChange-Id: I34c198ac76d528c605b770b6d759e06d59d98b28\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8155859\nCommit-Queue: Sean Li \u003cseannli@google.com\u003e\nReviewed-by: Chu-Hsuan Yang \u003cchuhsuan@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1670182}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ce93ab79ad71afdd04349ec9e514eb8825b20e4b\n" }, { "commit": "e60344418093002c18384c1f4e2ce8e4fb612268", "tree": "0468ae4ec280e790173be27ff9b2b0af3fc8b6cd", "parents": [ "659dcfaf385318376d8209b5d4848b7b18df6abe" ], "author": { "name": "seannli", "email": "seannli@google.com", "time": "Wed Jul 29 13:18:18 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 13:22:13 2026" }, "message": "camera: Skip AWB modes without a color-temperature mapping\n\nThe AWB-mode to color-temperature map only covers the seven AWB preset\nmodes (INCANDESCENT…SHADE). However, the HAL\u0027s\nANDROID_CONTROL_AWB_AVAILABLE_MODES list always contains AUTO and may\ncontain OFF, neither of which has an associated temperature.\n\nCameraDeviceDelegate::SetPhotoOptions iterated all available modes and\ndereferenced the lookup result without an end() check, and\nDoGetPhotoState used map.at() to read back the current temperature. Skip\nunmapped modes when picking the nearest preset, and tolerate an unmapped\ncurrent mode when reporting state.\n\nBug: 512995893\nTest: Build successfully\nChange-Id: Ic9b43aaaaacd6569996facc72709c338abe9fa73\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8156579\nReviewed-by: Chu-Hsuan Yang \u003cchuhsuan@chromium.org\u003e\nCommit-Queue: Sean Li \u003cseannli@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1670181}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: d75e9bd6d06bd093dc9e6017879c9758e9e9fa8b\n" }, { "commit": "659dcfaf385318376d8209b5d4848b7b18df6abe", "tree": "50c723b0524b22395ef981ee3f907fba594db9c5", "parents": [ "e8f27ca84a8bccb3f502891d3184af42a51db9c9" ], "author": { "name": "seannli", "email": "seannli@google.com", "time": "Wed Jul 29 13:14:51 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 13:18:11 2026" }, "message": "camera: Validate HAL-returned stream dimensions\n\nStreamBufferManager keeps the client\u0027s requested frame size in\nStreamContext::capture_format while taking the buffer dimension from the\nCamera3Stream returned by ConfigureStreams. The HAL3 contract does not\nallow the HAL to change width/height during stream configuration, so\nreject preview/recording streams whose dimensions do not match what was\nrequested. This keeps the buffer dimension and capture format consistent\nfor the rotation path in AcquireBufferForClientById.\n\nBug: 502760197\nTest: Build successfully\nChange-Id: I9fb41519875a1839e4aecc10ea400ac2cb231711\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8156179\nCommit-Queue: Sean Li \u003cseannli@google.com\u003e\nReviewed-by: Chu-Hsuan Yang \u003cchuhsuan@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1670178}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: f0931c2540f2fc14516de37f8feba9f9555b6050\n" }, { "commit": "e8f27ca84a8bccb3f502891d3184af42a51db9c9", "tree": "c7a2be9b22d69db5baf3fbd0194081a227326d75", "parents": [ "9323938b3be53e10d67ab1c836e14bd71f8a0230" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 29 09:47:22 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 09:50:38 2026" }, "message": "Reland \"Remove non-8-bitprofiles from D3D11/D3D12 decoder GUID mapping.\"\n\nThis is a reland of commit 5fb07ab13dd9c20d3f9cb156d427f6582a1d5671 with\nthe test updated for cases where H.264 support is not built-in.\n\nOriginal change\u0027s description:\n\u003e Remove non-8-bitprofiles from D3D11/D3D12 decoder GUID mapping.\n\u003e\n\u003e This fixes a bug where H.264 10-bit, 4:2:2, 4:4:4, and\n\u003e scalable/stereo/multiview profiles were falling through to 8-bit 4:2:0\n\u003e H.264 GUIDs in Windows D3D11 and D3D12 hardware decoders.\n\u003e\n\u003e TAG\u003dagy\n\u003e CONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\u003e\n\u003e Fixed: 536415966\n\u003e Change-Id: I631be73e2e56486720f816d59359903ac8ab0033\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127926\n\u003e Auto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Reviewed-by: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\n\u003e Commit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1666751}\n\nChange-Id: I57737ada3ff587fcbd3ef440bc8c6cd371d3d07c\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8139025\nReviewed-by: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCr-Commit-Position: refs/heads/main@{#1670089}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 21adbd76c796a7a28bad518f897fac2c0a98b25d\n" }, { "commit": "9323938b3be53e10d67ab1c836e14bd71f8a0230", "tree": "208b40753ae61d409f25e0323579a3af7ac1ce68", "parents": [ "c2ede8f19b36fb8af323841f172fbd73c9aeab75" ], "author": { "name": "Zhibo Wang", "email": "zhibo1.wang@intel.com", "time": "Wed Jul 29 06:51:49 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 06:54:54 2026" }, "message": "Reject non-keyframe with empty reference buffers\n\nWhen manual reference buffer control is enabled, a non-keyframe\nencode request with no reference buffers should be rejected early\nwith kBadReferenceBuffer error.\n\nBug: 536417240\nChange-Id: I0010bf0d05244db700a1a752e1a5a86290ae6522\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8135230\nCommit-Queue: Wang, Zhibo1 \u003czhibo1.wang@intel.com\u003e\nAuto-Submit: Wang, Zhibo1 \u003czhibo1.wang@intel.com\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1670019}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1197129c51eeb228b56e878282c203d094fb5ac7\n" }, { "commit": "c2ede8f19b36fb8af323841f172fbd73c9aeab75", "tree": "2141d59af7f1328c025e1cb7a774c9ee31470341", "parents": [ "e9e440c6dfb27fc49c6a1e3b743c672c4af3f0a3" ], "author": { "name": "Evan Liu", "email": "evliu@google.com", "time": "Wed Jul 29 05:34:45 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 05:38:16 2026" }, "message": "Enable OnDeviceWebSpeechSmallExpertModelMultiLanguage by default\n\nEnables OnDeviceWebSpeechSmallExpertModelMultiLanguage by default and\nbakes the full list of 17 supported languages into the default\nFeatureParam value.\n\nAlso removes the redundant entry from fieldtrial_testing_config.json now\nthat the feature is enabled by default on ToT.\n\nBug: 536109559\nChange-Id: I6e795695459800cea8c9a0adcad9dcfd8903f56f\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8157760\nCommit-Queue: Evan Liu \u003cevliu@google.com\u003e\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1670001}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 08dd260cfec6ea685ca2f03c5fdead081e3fa84c\n" }, { "commit": "e9e440c6dfb27fc49c6a1e3b743c672c4af3f0a3", "tree": "22bf1fcacc2f0c8ab690f46f250ece54c6abb279", "parents": [ "b3b543fc3668ad1ff646e37038dee94ee6f9e549" ], "author": { "name": "Xiaohan Wang", "email": "xhwang@chromium.org", "time": "Wed Jul 29 05:28:44 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 05:31:53 2026" }, "message": "cast: Fix audio sender samples count underflow crash\n\nDuring casting, the audio sender tracks the number of samples in the\nencoder using `samples_in_encoder_` on the main thread.\n- Increment: Incremented by `InsertAudio()` when raw audio is inserted.\n- Decrement: Decremented by `OnEncodedAudioFrame()` when the encoder thread posts back completed/skipped frames.\n\nDuring network underruns or CPU spikes, the encoder thread detects that\nit has missed frames, drops its partially-filled buffers, and increments\n`samples_skipped`. Because the raw audio was never inserted during the\ngap, `samples_in_encoder_` was not incremented for those missed frames.\nWhen the callback eventually runs on the main thread, the subtraction of\nthe skipped samples causes `samples_in_encoder_` to underflow below 0,\ntriggering a fatal DCHECK in developer/dcheck builds.\n\nSince `samples_in_encoder_` is only a telemetry estimate used for\ncongestion control (to drop frames if the backlog is too large), we\nshould clamp the value to 0 on underflow and log a warning instead of\ntriggering a fatal crash.\n\nBug: 537805359\nTest: Tab casting works and doesn\u0027t crash on underflows.\nChange-Id: I7750f2ca5075179f4dfff41f43b702b8b7933adb\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8135721\nCommit-Queue: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nReviewed-by: Jordan Bayles \u003cjophba@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669997}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 4084e55ddd98333e0dbfc6d1b47fab81d45fbd1c\n" }, { "commit": "b3b543fc3668ad1ff646e37038dee94ee6f9e549", "tree": "4095f1155157be2f1539bd2a8480c0cbd396add5", "parents": [ "13200e668868259557c58fdc5db47d584aec3851" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Wed Jul 29 01:14:50 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 01:19:21 2026" }, "message": "Remove AudioBus deadcode\n\nNow that AudioBus is fully spanified, AudioBus::CheckOverflow is unused.\n\nChange-Id: I43a318eab94be04067d95fb54dddf7081d7224ed\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8165440\nAuto-Submit: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669920}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: f2c544a284e96dd80cef2e20579a33f43e49944d\n" }, { "commit": "13200e668868259557c58fdc5db47d584aec3851", "tree": "a30eeb6c0faf4ab92199267cc8c0e01f37651bc8", "parents": [ "20cf967deb47b36f2aa11ceb5361f062d2aed51d" ], "author": { "name": "Jordan Bayles", "email": "jophba@chromium.org", "time": "Wed Jul 29 01:12:18 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 01:16:10 2026" }, "message": "[Symphonia] Launch OpusAudioDecoder on Desktop\n\nThis patch launches the DirectOpusAudioDecoding experiment on desktop\nconfigurations (Windows, macOS, Linux, ChromeOS). Android and Fuchsia\nare expected to launch sometime in the next couple of weeks, similar to other codecs.\n\nGooglers may view more info at go/symphonia-audio-decoding-journal\n\nBug: 476203189\nChange-Id: Ib9ee1aeedd1cccef632ea96133d2bc5b06db2a9d\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8093947\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Jordan Bayles \u003cjophba@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669918}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ccc44e22b86e3e0caf1f49f9230b6c8b0096020b\n" }, { "commit": "20cf967deb47b36f2aa11ceb5361f062d2aed51d", "tree": "26ff0b0003b5e2b83324e18bf0708fe11bb3002f", "parents": [ "30cb6e3f4fec1e688f2c77891aabf30e65ed74a7" ], "author": { "name": "Christopher Cameron", "email": "ccameron@chromium.org", "time": "Wed Jul 29 00:49:20 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 29 00:52:56 2026" }, "message": "CodecPicture: Don\u0027t drop metadata in duplicate\n\nWhen calling AV1Picture::Duplicate, forgot to replicate the HDR\nmetadata. When calling VP9Picture::Duplicate, we didn\u0027t forget.\n\nMove the the functionality to copy common fields to the base class to\navoid this sort of thing.\n\nBug: 522100236\nChange-Id: I72a5c69aafa0fea31f0dca16ac0e7e559522dd15\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8159539\nCommit-Queue: ccameron chromium \u003cccameron@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669911}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e80ac019dfcc256102286789b1ae77c1a382ac6d\n" }, { "commit": "30cb6e3f4fec1e688f2c77891aabf30e65ed74a7", "tree": "7bfa183912a3e0130c641f40aa8522b00d6d3dc3", "parents": [ "43b1ecdeac53d41ba94f721dd5adbbadb2d847c2" ], "author": { "name": "Syed AbuTalib", "email": "lowkey@google.com", "time": "Tue Jul 28 22:31:48 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 22:35:45 2026" }, "message": "Enable IAMF build flag\n\nDefault enable the IAMF build flag. However, it is still guarded by the\n`kIamfAudioDecoding` feature flag.\n\nBug: 535279329\nBinary-Size: Enabled a repo that was already approved by ATLs.\nFuchsia-Binary-Size: Enabled a repo that was already approved by ATLs.\nChange-Id: I8db27afbea9eb8e06ec6e762ba07ea3d92487cde\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8162149\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669820}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 43d2c2c571a31fc7a8ff89d182207ce93d827571\n" }, { "commit": "43b1ecdeac53d41ba94f721dd5adbbadb2d847c2", "tree": "3fd30d03b49c1a398a15ff921b629cb14fb34c70", "parents": [ "7d3112ac831b23ceff39ab6e7845c6b37ea1b5d7" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Tue Jul 28 21:03:15 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 21:09:13 2026" }, "message": "media: Remove UNSAFE_TODO for converting UUIDs to ByteArray\n\nToJavaByteArray implicitly converts vectors to span.\nUUID is actually a vector.\n\nBug: 40284755\nChange-Id: I27c2bcf5af1c168786766b614469bc3e2f87caf1\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8162569\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669747}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 4382616be71abe5d83111e179ea04ba75b1dfa46\n" }, { "commit": "7d3112ac831b23ceff39ab6e7845c6b37ea1b5d7", "tree": "7c180729f0e48e71916bdfe15d837a0f0a2d9f67", "parents": [ "8e8681a7bc0b135327557f016cbd873b52f4af9e" ], "author": { "name": "Wooseok Sung", "email": "tjddntjr1@gmail.com", "time": "Tue Jul 28 21:01:58 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 21:05:49 2026" }, "message": "media: Add MojoLPM fuzzer for the VideoEncodeAccelerator service\n\nAdds a libFuzzer target driving the renderer-\u003eGPU media.mojom.\nVideoEncodeAccelerator interface (Initialize / Encode /\nUseOutputBitstreamBuffer / RequestEncodingParametersChange*) through\nMojoVideoEncodeAcceleratorService, backed by a FakeVideoEncodeAccelerator so\nno GPU bring-up is required. The primary deserialize target is the VideoFrame\nStructTraits::Read plane-layout reconstruction (offsets[]/strides[] in\nvideo_frame_mojom_traits.cc). No existing fuzzer drives this service.\n\nBug: 526778515\nTest: built with is_asan\u003dtrue use_libfuzzer\u003dtrue enable_mojom_fuzzer\u003dtrue; the\nseed corpus drives Initialize+Encode and smoke-runs with no crashes.\n\nChange-Id: I591ea02227170e07bfd1ae4aa4c94a80fdae3aeb\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7976699\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669745}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 8210f6a2f5f19166b510467aa84f4b59c7ff1bf7\n" }, { "commit": "8e8681a7bc0b135327557f016cbd873b52f4af9e", "tree": "fd2b87cf4fa34ea648a35c6add4f2dfb6eadb2c1", "parents": [ "b67d81e24adcf05334e372aa531ddbc7703b0cf1" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 28 20:26:22 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 20:29:57 2026" }, "message": "[media] Clean up overlay usage for Android decoder with SharedImage\n\nVideoFrameMetadata allow_overlay is going away and Android\u0027s\nVideoFrameFactoryImpl can be simpler by just checking for Scanout usage\non SharedImage. Viz display compositor can do the right thing based on\nother metadata on TransferableResource like wants_promotion_hint.\n\nBug: 425634684\nChange-Id: I01eef4fdb3a35221ac1d4dbc8e1751efdfe434c2\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8162023\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669705}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 9a4243c35fc9b97c07deeed94da83709b076b11f\n" }, { "commit": "b67d81e24adcf05334e372aa531ddbc7703b0cf1", "tree": "d57c0c688a081b517d1a7c01864d13782f0e95ea", "parents": [ "6b67aedd4b213ee9dfef071973ea5c730d7b50c5" ], "author": { "name": "Qiu Jianlin", "email": "jianlin.qiu@intel.com", "time": "Tue Jul 28 20:07:26 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 20:11:59 2026" }, "message": "[media] Add H.265 prefix SEI builder for HDR static metadata.\n\nAdd BuildPackedH265SEI() to media/gpu/h265_builder, packing the HDR10\nstatic metadata SEI messages into an H.265 prefix SEI NALU: mastering\ndisplay colour volume (payload type 137) and content light level\ninformation (payload type 144). Only the messages that are present are\nemitted. Also allow SEI NALU types in H26xAnnexBBitstreamBuilder so the\nprefix SEI NALU can be started.\n\nThis is the codec-level packing used to plumb HDR metadata when encoding\nHEVC main10; the D3D12 VEA wiring lives in the child CL.\n\nBug: 537818862\nChange-Id: I09129ec3eae75c49c02e3cdada666313b5f5789a\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8161680\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669688}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: cae7231bc1ffed27e0ed4babd8187f569b5f4456\n" }, { "commit": "6b67aedd4b213ee9dfef071973ea5c730d7b50c5", "tree": "6503206f985ae6b4372341675601380b767232fa", "parents": [ "4e7e3f4601ddceb700f27d2237a7124bc8b59268" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Tue Jul 28 19:04:36 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 19:10:02 2026" }, "message": "Fix AAudio input timestamps alignment\n\nThe media pipeline expects capture timestamps to point to the first\n(oldest) sample in a given input buffer. However, since AAudio updates\nits frame counters before running data callbacks, we\u0027ve been\naccidentally calculating timestamps that point to the last (newest)\nsample.\n\nWhen `kAAudioVariableSizedCallbacks` is enabled, this end-of-buffer\nalignment injects the variable callback duration directly into the\ntimestamp, which causes massive fluctuations in the\n`TimestampDiscontinuitiesPer10s` metric.\n\nThis CL fixes the issue by taking into account the size of incoming\ncallbacks when calculating input timestamps. To mitigate the risk of\nregressions, the `kAAudioFixCaptureTimestamp` flag is introduced to\nrevert to the legacy behavior in case of emergency.\n\nThis CL also adds a V2 version of the `InputTimestampDiscontinuitySize`\nmetric to differentiate the behavior. The legacy metric will be marked\nas obsolete in a future cleanup CL, once we confirm there are no\nregressions.\n\nBug: 538642984\nChange-Id: I55a8a1e0269b832abb634ac8525f69994e15a75c\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8144299\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nAuto-Submit: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Evan Liu \u003cevliu@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669641}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 70de8d121cf6dda871a2d746ecc1e4b1fffce6a2\n" }, { "commit": "4e7e3f4601ddceb700f27d2237a7124bc8b59268", "tree": "1d375d82d9c6698a233d5dc14b7d493a9c25ebbb", "parents": [ "ca756368f0c69c6d58fe2368c5ba2cc165ed7e8f" ], "author": { "name": "Piet Schouten", "email": "Piet.Schouten@microsoft.com", "time": "Tue Jul 28 18:07:58 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 18:11:43 2026" }, "message": "media: Remove Vorbis, FLAC and Opus from PlayReady queried audio codecs\n\nMediaFoundationService queries hardware secure CDM capabilities for the\nPlayReady CDM, which does not support Vorbis, FLAC or Opus. Querying\nthem adds IsTypeSupportedEx calls that can never report supported, and\nthese calls are expensive.\n\nRemove the three codecs from kAllAudioCodecs along with their now\nunreachable FourCC mappings.\n\nBug: 40218626\nChange-Id: I87a6a272ebf6785b540d53c6b4ddb1015bff4a93\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8157979\nReviewed-by: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nCommit-Queue: Piet Schouten \u003cPiet.Schouten@microsoft.com\u003e\nReviewed-by: Sangbaek Park \u003csangbaekpark@chromium.org\u003e\nReviewed-by: Daoyuan Li \u003cdaoyuanli@microsoft.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669611}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5702b766647980b66eb8a661d7820b813ac2694f\n" }, { "commit": "ca756368f0c69c6d58fe2368c5ba2cc165ed7e8f", "tree": "2c98e96dca52f3075e1c4d78c1dec878254f19a9", "parents": [ "cdb86e83337eb18960435b87b836b8f4bf9e1fea" ], "author": { "name": "Eugene Zemtsov", "email": "eugene@chromium.org", "time": "Tue Jul 28 17:01:55 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 17:05:04 2026" }, "message": "media: Reject HEVC non-first slice segment when prior slice is missing\n\nDisallow non-first slice segments (first_slice_segment_in_pic_flag \u003d\u003d 0)\nwhen no prior slice header exists (e.g. after mid-stream SPS/PPS\nupdates). Enforce this in H265Parser::ParseSliceHeader() when\nvalidate_extended_bitstream_ is set, and in\nH265Decoder::PreprocessCurrentSlice() when curr_pic_ is null.\n\nBug: 536470854\nChange-Id: Ie7f159fbb9cdb229f516f0e46699a22aedefa184\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8160680\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669582}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5dc7ba4f4e4770e36b567afbf457740e227d9edb\n" }, { "commit": "cdb86e83337eb18960435b87b836b8f4bf9e1fea", "tree": "13f4ba316178cfad6e651d63c7165bcede0af855", "parents": [ "8ba80815aa31f922d13aeb30a4684df2d8366992" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Tue Jul 28 16:28:54 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 16:32:31 2026" }, "message": "media: Add P010LE output support to VideoFrameConverter\n\nAdds PIXEL_FORMAT_P010LE output support to VideoFrameConverter.\n\nKey changes:\n- Adds WrapBiplanarFrameInTriplanarFrame() to map biplanar formats\n (NV12, NV12A, P010LE) to triplanar equivalents (I420, I420A,\n YUV420P10), utilizing pooled scratch frames from frame_pool_ for\n efficient U/V plane separation and merging.\n- Implements internals::Convert8To16Plane() and I4xxxPxxToP010() to\n convert 8-bit, 10-bit, and 12-bit planar/biplanar frames to P010LE\n using libyuv primitives (Convert8To16Plane, ConvertToMSBPlane_16,\n MergeUVPlane_16) and 12-to-10 bit LSB masking.\n- Expands VideoFrameConverterTest to cover all conversion and scaling\n permutations targeting P010LE. Adds test names for readability.\n\nBug: 537818862\nChange-Id: I6b5cd57c0428638fc225ac3f6863e06ced7c9db6\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8160320\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Sida Zhu \u003czhusida@bytedance.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669565}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 28e7aac8dfd7b14b85ee609f4f3703d5b2a1215c\n" }, { "commit": "8ba80815aa31f922d13aeb30a4684df2d8366992", "tree": "3700dd0ec5ab2c549e660d0bf6a549c4268aef22", "parents": [ "c5f92f73bfe971d66590fcf907bc2178b2dbe918" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Tue Jul 28 16:28:36 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 16:31:56 2026" }, "message": "[VideoFrame] Fix error in default initialization for NV12A frames\n\nI think this is only used in tests since NV12A is basically a\nhardware only output format.\n\nChange-Id: I9640b566f7e1b81724f23b056fba004b67084519\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8160600\nReviewed-by: Alexander Cooper \u003calcooper@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Sida Zhu \u003czhusida@bytedance.com\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669564}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: d2adf622f0c6c4f541ec92e0e63bac40f96d9bbe\n" }, { "commit": "c5f92f73bfe971d66590fcf907bc2178b2dbe918", "tree": "ffdbb0ecf58bff069a7089eff21656c237220ca2", "parents": [ "0fa3451ce632f0c6c699904b486ed35787406174" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 28 13:26:57 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 13:32:51 2026" }, "message": "[media] Update D3D11Decoder and DCompWrapper to set scanout properly\n\nUpdate D3D11TextureWrapper and DCompWrapper to set Scanout SharedImage\nusages properly based on the data used for allow_overlay by the client.\nThis will help remove allow_overlay eventually as all clients should\nset correct scanout usage when creating the SharedImage.\n\nFor DCompTextureWrapper, it is called from\nMediaFoundationRendererClient which sets and checks usage properly.\nFor D3D11VideoDecoder, allow_overlay can be set for NV12/P010 formats.\n\nBug: 425634684\nChange-Id: I3ae81b73b4224dcb95e54521b50eb25a3db83d3b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8153302\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669415}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ac3c46afff884b8373e1d1e4c4d5bb4369407313\n" }, { "commit": "0fa3451ce632f0c6c699904b486ed35787406174", "tree": "def611bd0448c3c615cec4c4a622dce91d155a80", "parents": [ "6b40c46263c9e5dda23c7292becefecf8db47e36" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 28 13:25:47 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 13:29:26 2026" }, "message": "[media] Update scanout usage for Mac/Fuchsia decoder, MappablePool\n\nAs part of removing VideoFrameMetadata::allow_overlay in favor of\nSharedImageUsageScanout, this change updates callsites in Mac/Fuchsia\nvideo decoders and MappableSIVFPool to set allow_overlay based on the\nscanout usage which will make it easier to remove the bool.\n- FuchsiaVideoDecoder now passes `use_overlays_for_video_` for setting\nSharedImage usage, making it consistent.\n- VideoToolboxVideoDecoder remove scanout usage for invalid color space\nand makes usages consistent.\n- MappableSIVFPool plumbs `allow_overlay` from input frame to\nSharedImage creation, and follows logic similar to earlier based on\noutput formats. Also add tests for YV12 case which is bit complex\nbut gets simplified as allow_overlay goes away.\n\nBug: 425634684\nChange-Id: Id69f8eb2c6e206d2693cdf1cc803ef98a244509a\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8144939\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669413}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 8cd3622c449d89e1280a5e38919cb3bb4b936d73\n" }, { "commit": "6b40c46263c9e5dda23c7292becefecf8db47e36", "tree": "b29bac8bcc2875ffd6dd7274121590d898f99b3e", "parents": [ "9fd8505d8b63821d3583014932643d9b2e2468e2" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Tue Jul 28 06:41:54 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 06:45:17 2026" }, "message": "Avoid calling media::AudioBus::Create from the realtime thread\n\nInstead of calling media::AudioBus::Create on a realtime thread this\nchange creates just the minimum necessary audio buses and stores them\nin ConvertingAudioFifo using the use_input_bus_pool option. New audio\nbuses are only created on the initial calls to ProcessAudio.\n\nBug: 512016773\nChange-Id: I9349346418b7c5c22af50aec885a1dfabba7959c\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8131747\nReviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1669243}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 7b202bb1750000776029a9ef52077cc9bd724619\n" }, { "commit": "9fd8505d8b63821d3583014932643d9b2e2468e2", "tree": "45f6a45b9c2cedfa1b5076c7a87fc533eb5882b7", "parents": [ "811c9980e0ff7f4c876d5ea8d7d240d521a83e55" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Tue Jul 28 01:09:53 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 01:14:58 2026" }, "message": "D3D11H265Accelerator: Recompute range extension flag from SPS\n\nPicParamsFromSPS() only ever set is_rext_ to true, so once a range\nextension SPS was seen the flag remained set even when the active SPS\nwas no longer a range extension profile. This caused SubmitSlice() to\nwrite a range-extension-sized picture parameter buffer to the driver for\nnon-range-extension content.\n\nAssign is_rext_ unconditionally from the active SPS so it always tracks\nthe current profile.\n\nTAG\u003dagy\nCONV\u003df85f2b20-8967-4437-8b89-ebb1d9b16b7f\n\nFixed: 536417208\nChange-Id: Iab1fef4f75eb3d8a61eb8e8067d74be14740dc04\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8142185\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669131}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: c24a439719b907b37c7254db6882f42ae85adfe6\n" }, { "commit": "811c9980e0ff7f4c876d5ea8d7d240d521a83e55", "tree": "c6f5d0de8b1c3c13a309f7f44f2043f116755839", "parents": [ "fed35dda19468cf5635e9363f8732d66646f37d5" ], "author": { "name": "Alan Ding", "email": "alanding@chromium.org", "time": "Tue Jul 28 00:29:51 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 28 00:34:08 2026" }, "message": "Web Speech API: Add audio start and end timestamps to SpeechRecognitionResult.\n\nThis CL introduces audioStartTime and audioEndTime attributes to SpeechRecognitionResult in the Web Speech API. These attributes represent the start and end times of the source audio corresponding to the recognition result. To mitigate hardware fingerprinting risks, the timestamps are quantized to 2ms precision. The new attributes are gated behind the experimental Web Speech Timestamps runtime-enabled feature.\n\nPlease see the following for more details:\n- I2P: https://groups.google.com/a/chromium.org/g/blink-dev/c/r78IyGSnclg\n- Explainer: github.com/WebAudio/web-speech-api/blob/main/explainers/speech-recognition-result-timestamps.md\n\nBug: crbug.com/528037568\nChange-Id: I66574097051bcf610052201465a3148523b531a0\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8129198\nReviewed-by: Will Harris \u003cwfh@chromium.org\u003e\nReviewed-by: Evan Liu \u003cevliu@google.com\u003e\nReviewed-by: Reilly Grant \u003creillyg@chromium.org\u003e\nReviewed-by: Mike Wasserman \u003cmsw@chromium.org\u003e\nCommit-Queue: Alan Ding \u003calanding@chromium.org\u003e\nReviewed-by: Stefan Zager \u003cszager@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1669111}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 60e7e25285380ae43d1db53677e9bbc665f3c330\n" }, { "commit": "fed35dda19468cf5635e9363f8732d66646f37d5", "tree": "0a38985dfd854cf91381ab47da1f8c59b9c06fee", "parents": [ "378886a2303f4d329db70320686e45a7914d133d" ], "author": { "name": "Gazal", "email": "agazal@google.com", "time": "Mon Jul 27 21:19:16 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 27 21:23:05 2026" }, "message": "[Code Health] Optimize logging in media/base/android\n\nConvert some log statements in the media/base/android directory to use\nformat specifiers. This avoids runtime string concatenation and\nStringBuilder overhead in non-production log statements.\n\nBug: 529082276\nChange-Id: I34b0f8d40db14bb1f816985f7f85b4023362cf12\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8146028\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nAuto-Submit: Gazal Agarwal \u003cagazal@google.com\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668979}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 942797da8487c9b3249dd6e18a7e8b5904e3bc74\n" }, { "commit": "378886a2303f4d329db70320686e45a7914d133d", "tree": "634976c2ac2a1f0d2e7b5b8c059805ddf7e71664", "parents": [ "931e13eaea41c9ec102b74a0c7bd885800c0a4b5" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Mon Jul 27 19:53:50 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 27 19:57:33 2026" }, "message": "h264_decoder: Prevent signed integer overflow in frame_num_offset\n\nCheck frame_num_offset and resulting pic_order_cnt calculations using\nbase::CheckedNumeric\u003cint\u003e in H264Decoder::CalculatePicOrderCounts to\nprevent signed integer overflow on invalid bitstream inputs.\n\nFixed: 525520948\nChange-Id: If1179a06c1f4ace38c92936170867a8478c84fd4\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8138663\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668902}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: a1a8e5fbb4611b99cab2bed23097040907027d84\n" }, { "commit": "931e13eaea41c9ec102b74a0c7bd885800c0a4b5", "tree": "6cd1cb9ce05e170cdaf7f9cf17ea4eb173dd813a", "parents": [ "1ffa6927ec54c9fdb022c2683f45295d5eb9cf29" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Mon Jul 27 17:22:21 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 27 17:26:19 2026" }, "message": "Remove epb bit count from h264 parser parameters\n\nFixed: 536434345\nChange-Id: Ia3fec6e2bf93c8cb98ef6e10b55ac673408b1d49\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8142642\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCommit-Queue: Frank Liberato \u003cliberato@chromium.org\u003e\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668770}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 0ea76a3fb40b82246dcca440ac45c04ea288948b\n" }, { "commit": "1ffa6927ec54c9fdb022c2683f45295d5eb9cf29", "tree": "57c4e7b88014cc02200800009e415899648169b8", "parents": [ "475e55c9323cb11d9e42ba6722dc40923bcc445a" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Mon Jul 27 16:56:32 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 27 17:00:27 2026" }, "message": "Use ValidateAndGetPlaneInfo for external buffers\n\nA previous patch introduced this helper and made use of it in the\nFillVADRMPRIMESurfaceDescriptor function, but we should also be using it\nin the FillVASurfaceAttribExternalBuffers for similar purposes.\n\nFixed: 536428988\nChange-Id: I1bcc16a56c1170ab99c5eed37c9119a51d5ec4db\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8144537\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Frank Liberato \u003cliberato@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668750}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 4f2bb3f790c6898f6a4065e0e6d65c437826775d\n" }, { "commit": "475e55c9323cb11d9e42ba6722dc40923bcc445a", "tree": "21ea34209150415cde5b5b07dc97d663382a6fa3", "parents": [ "d3ed74bef73097720ebad973073102cd1c095c95" ], "author": { "name": "Evan Liu", "email": "evliu@google.com", "time": "Fri Jul 24 21:57:00 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 22:02:58 2026" }, "message": "Add multi-language support to TinyGemma in Web Speech API\n\nThis CL adds multi-language support to TinyGemma via a parameterized\nfeature. These languages will be defined by a server-side Finch config.\n\nFixed: 536109559\nChange-Id: I935b3a84afdfea7d24ce12ac4674ac2034d9f2e2\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8117354\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCommit-Queue: Evan Liu \u003cevliu@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1668187}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: edbdb20855ea9a3e8e7884b8e3cbbd4c13f72ea7\n" }, { "commit": "d3ed74bef73097720ebad973073102cd1c095c95", "tree": "538bb32b6dae782abd62b95200849f3fd216b554", "parents": [ "e554d29dee34d46890ae35128089425c0901d4a9" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Fri Jul 24 21:15:53 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 21:19:26 2026" }, "message": "media: Safely query secure decoder capabilities without fallback creation\n\nPreviously, MediaCodecUtil created an insecure decoder in order to query\nif adaptive playback was supported, and then blindly appended `.secure`\nto the insecure decoder\u0027s name to instantiate the secure codec. This was\na legacy workaround for b:15587335 on older versions of Android.\n\nOn modern Android platforms, `MediaCodecList` properly exposes secure\ncodecs and their capabilities directly via `FEATURE_SecurePlayback`.\nThis CL removes the legacy `.secure` string hack and the redundant\ninsecure codec instantiation, instead directly querying the\n`MediaCodecList` to safely discover and instantiate secure codecs.\n\nManual Trace Info on my Pixel 8a:\n\n1. Opened Chrome Stable to integration.widevine.com/player.\n\n2. Set up chrome://tracing via chrome://inspect -\u003e trace.\n\n3. Started recording.\n\n4. Played the video, and after 5 seconds of playback, stopped recording.\n\n5. I built Chrome with standard gn args.\n\n6. Set up tracing again.\n\n7. Played the video, and after 5 seconds of playback, stopped recording.\n\nThe results showed that the CodecAllocator::CreateMediaCodec was ~180ms\nfor stable chrome, and ~42ms for the built chrome. This makes sense\nbecause we are eliminating the redundant hardware allocation and IPC\nrequired to create and destroy the redundant insecure codec.\n\nWe will see an improvement in `Media.TimeToFirstFrame.EME` for Android.\n\nBug: b:15587335\nChange-Id: Iee631349c50a704b8bf0581bf90434adfc13e227\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8134210\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668165}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: a1bd72788c66d64bd95508be8c2fe5d534ef88fd\n" }, { "commit": "e554d29dee34d46890ae35128089425c0901d4a9", "tree": "9032cdd2bfe36384c6faffe81d24bd4571cad1b6", "parents": [ "b486f9c7af5cef775cbc3a5c0273e5b609bdb636" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Fri Jul 24 19:37:33 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 19:40:48 2026" }, "message": "Reland \"Enable kAAudioPerStreamDeviceSelection by default\"\n\nThis reverts commit de92b900b2385efee8e9b2cba2bcdaa595487c4b, relanding\nthe CL which turns on kAAudioPerStreamDeviceSelection by default.\n\nThe original CL caused the `setSinkId()` API to unexpectedly launch on\nAndroid as well. This API is tied to the `EnableAudioOutputDevices`\nruntime flag, which is conditionally set by\nkAAudioPerStreamDeviceSelection on Android. This CL now ensures the API\nis only set on desktop.\n\nOriginal change\u0027s description:\n\u003e Revert \"Enable kAAudioPerStreamDeviceSelection by default\"\n\u003e\n\u003e This reverts commit b725e6c6c12d58a0c801d2537170a30f1d8464dc.\n\u003e\n\u003e Reason for revert: Causing test failure in external/wpt/audio-output/idlharness.https.window.html\n\u003e\n\u003e Failure Link: see b/537839417\n\u003e\n\u003e Original change\u0027s description:\n\u003e \u003e Enable kAAudioPerStreamDeviceSelection by default\n\u003e \u003e\n\u003e \u003e This CL enables the kAAudioPerStreamDeviceSelection flag by default, but\n\u003e \u003e gates the feature for desktop versions of android.\n\u003e \u003e\n\u003e \u003e This matches the current behavior of the Finch experiments, currently\n\u003e \u003e rolled out to 100% on desktop Android devices.\n\u003e \u003e\n\u003e \u003e Bug: 425608358\n\u003e \u003e Change-Id: Id8f929dfd19a64c683941b04ab5a71d83cdcc109\n\u003e \u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8111797\n\u003e \u003e Commit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\n\u003e \u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e \u003e Cr-Commit-Position: refs/heads/main@{#1665782}\n\u003e\n\u003e Bug: 425608358\n\u003e No-Presubmit: true\n\u003e No-Tree-Checks: true\n\u003e No-Try: true\n\u003e Change-Id: I092beb5095736cec86775ad1a32fadf23d3000aa\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8135260\n\u003e Bot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\n\u003e Commit-Queue: Hailey Wang \u003chaileywang@google.com\u003e\n\u003e Owners-Override: Hailey Wang \u003chaileywang@google.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1666603}\n\nBug: 425608358\nChange-Id: I38a501ffbe2470a61baae8f5d670423587af5fe0\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8131033\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nReviewed-by: Joey Arhar \u003cjarhar@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1668092}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: fc5fdf03f263cd72d85914f6480c9128d95a2c14\n" }, { "commit": "b486f9c7af5cef775cbc3a5c0273e5b609bdb636", "tree": "1e9e59781ec7aba6ab8b45d378f2a75765aa0815", "parents": [ "8dbd39c821c2c9d51c98371d7dab73a54b0b6ae0" ], "author": { "name": "Michael Tang", "email": "tangm@microsoft.com", "time": "Fri Jul 24 15:34:48 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 15:38:38 2026" }, "message": "Include dxheaders d3d12.h before SDK headers when using d3dx12_core\n\nThis change prepares Chromium for rolling the DirectX-Headers repo. All\nusers of d3dx12_core.h now must ensure that the dxheaders version of\nd3d12.h is included before the SDK version.\n\nThe DirectX-Headers repo added D3D12_RESOURCE_FLAG_USE_TIGHT_ALIGNMENT\nto d3d12.h, which is now used by d3dx12_core.h. This means that the\ndxheaders version of d3d12.h now must be included before d3dx12_core.h\nis included, otherwise there will be a build failure when d3dx12_core.h\ntries to use the new enum value.\n\nChange-Id: I068e7dbb56510c5324799f50e98bc3866a6a6964\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8139027\nReviewed-by: Rafael Cintron \u003crafael.cintron@microsoft.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCommit-Queue: Michael Tang \u003ctangm@microsoft.com\u003e\nCr-Commit-Position: refs/heads/main@{#1667876}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 7e3e2d40e24eb6e647fdd66d17b3b30ec9b80b8e\n" }, { "commit": "8dbd39c821c2c9d51c98371d7dab73a54b0b6ae0", "tree": "a71564822ad315e5f692984427af0c8e9a8eb7ba", "parents": [ "9e65a085910ed43fd6e95ad4bcba411853d95e0c" ], "author": { "name": "Xiaohan Wang", "email": "xhwang@chromium.org", "time": "Fri Jul 24 08:48:37 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 08:51:40 2026" }, "message": "media: Add Save Video Frame feature in Global Media Controls\n\nThis CL implements the \"Save Video Frame\" feature in Global Media\nControls (GMC), allowing users to download the current video frame as an\nimage directly from the media hub.\n\nKey features and implementation details:\n- Added the \"Save Video Frame\" button to the Zenith GMC UI, positioned\n to the left of the Cast and Picture-in-Picture buttons.\n- Implemented state-based visibility: the button is only shown when a\n saveable video frame is available (unprotected content, single\n player).\n- Extended the MediaSession Mojo interface to include a new action\n (kSaveVideoFrame) and an availability field (is_video_frame_available)\n in MediaSessionInfo.\n- Updated the Blink media stack to report video frame availability to\n the browser, including checks for encryption and frame readability.\n- Added the kGlobalMediaControlsSaveVideoFrame feature flag (disabled by\n default) to control the feature\u0027s visibility in the UI.\n- Optimized availability reporting in Blink to minimize redundant IPC\n communication between the renderer and browser processes.\n\nManual Test:\n- Verified manually based on the test plan in go/save-video-frame-in-gmc.\n- Screenshot of the UI: screen/5vMX6h9rucJKQd4\n\nTAG\u003dagy\nCONV\u003d62abc0b1-2764-423c-9637-c845802f9966\n\nBug: 419840781\nChange-Id: Id9c788ff8c35cb3e9f314d0ccf58783bbfcdaf52\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7770777\nReviewed-by: Evan Liu \u003cevliu@google.com\u003e\nReviewed-by: Basia Zimirska \u003cbasiaz@google.com\u003e\nReviewed-by: Tommy Steimel \u003csteimel@chromium.org\u003e\nCommit-Queue: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nReviewed-by: Arthur Sonzogni \u003carthursonzogni@chromium.org\u003e\nAuto-Submit: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nReviewed-by: David Yeung \u003cdayeung@chromium.org\u003e\nReviewed-by: Zijie He \u003czijiehe@google.com\u003e\nReviewed-by: Will Harris \u003cwfh@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Michael Wilson \u003cmjwilson@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667720}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 9490ac84ed9cab4982b8081c95bb19b916b78544\n" }, { "commit": "9e65a085910ed43fd6e95ad4bcba411853d95e0c", "tree": "dbf4171fac0089a2cc53f9e60a52c7563cbb0318", "parents": [ "7812ca0b1bd4661674ce21bc2a861eda2f48d297" ], "author": { "name": "henrika", "email": "henrika@google.com", "time": "Fri Jul 24 08:40:40 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 08:43:37 2026" }, "message": "Audio: Add UMA for CoreAudioUtil::SharedModeInitialize\n\nThe Media.AudioOutputController.ProcessDeviceChangeTime metric currently\nshows massive outliers on Windows. Recent Chrometto traces indicate that\nthe root cause is the OS-level call to initialize the shared-mode audio\nendpoint.\n\nThis CL adds a new UMA,\nMedia.Audio.CoreAudioUtil.SharedModeInitializeTime, to isolate and\nmeasure the exact execution time of this Windows API call. This will\nprovide population-wide proof that waiting for the Windows WASAPI driver\nis the true origin of the extreme delays.\n\nBug: 516482033\nChange-Id: I6362bee8c5262a6d3ee98fc30c65b5dd7e1665ec\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8137724\nReviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\nReviewed-by: Takashi Toyoshima \u003ctoyoshim@chromium.org\u003e\nCommit-Queue: Henrik Andreasson \u003chenrika@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667713}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 91920dbf1d48bafad61fb5ee89253d7c22af1823\n" }, { "commit": "7812ca0b1bd4661674ce21bc2a861eda2f48d297", "tree": "7f5c5d0651d56696e2ff119627e16002087e3b55", "parents": [ "8eceac355f3d51afcc22c901110227323ae93654" ], "author": { "name": "Kelsen Liu", "email": "kelsenliu@chromium.org", "time": "Fri Jul 24 02:14:29 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 02:18:31 2026" }, "message": "media: Spanify WebMListParser::Parse() buf parameter\n\nReplace raw pointer + size pair with base::span\u003cconst uint8_t\u003e in\nWebMListParser::Parse(), WebMParseElementHeader() and the\nTracks/Info/Cluster/StreamParser parse paths.\n\nBug: 40284755\nChange-Id: I55fe883866e6f7784452b7549773f28ba8d580f5\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8117967\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Kelsen Liu \u003ckelsenliu@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667597}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: d9684da13e13e5a7bcfd816e14b1fcf84dd5a849\n" }, { "commit": "8eceac355f3d51afcc22c901110227323ae93654", "tree": "616cc69bfb0ec118b584ee75899f9e0995673824", "parents": [ "4d5c3ceb8f55a3dc5f8aae6353d2fc743c7e2c96" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Fri Jul 24 01:59:45 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 02:03:56 2026" }, "message": "Use checked numerics for h264 pic params\n\nFixed: 536426855\nChange-Id: I3d7320ae137bde2cdfc3b6790e8f11cb3ac3c44b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8140465\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nReviewed-by: Phil Yan \u003cphilyan@chromium.org\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667592}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 0ed2eb16254e987a91750f99f6a6009357f399db\n" }, { "commit": "4d5c3ceb8f55a3dc5f8aae6353d2fc743c7e2c96", "tree": "e265c47b6cd3995820d85d5e40bed4449f419484", "parents": [ "cfd87ce0f2ba95a84cdcb97d6a5c93fd83ffd6c2" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Fri Jul 24 01:41:41 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 01:44:58 2026" }, "message": "Reset VA buffer when context is destroyed\n\nFixed: 536416576\nChange-Id: I76a414ca7b28ff17686e7176171f14cfef1ee21e\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8142739\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nReviewed-by: Phil Yan \u003cphilyan@chromium.org\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667584}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 38fbb0d4e163e92e40888430f11ed8b32c790d6a\n" }, { "commit": "cfd87ce0f2ba95a84cdcb97d6a5c93fd83ffd6c2", "tree": "13011e798762ff80f6c818d1411d7e8f2aa6cc1f", "parents": [ "2bb65e018cc027ee6411c7bbee6fee2d758de3d2" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Fri Jul 24 01:37:17 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 01:41:15 2026" }, "message": "Fix bounds check in D3D shared decoder utilities\n\nIf the start code\u0027s size matches the bytes available, The condition here\nfails, the start code gets written to the current bitstream buffer,\nwhich then gets written again on line 85, including the weird trailing\nstart code that should have been bundled in the _next_ bitstream buffer.\n\nFixed: 536438861\nChange-Id: I9b5d1109eb93f8257757994c6cdc2483743c2b46\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8139406\nReviewed-by: Phil Yan \u003cphilyan@chromium.org\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667581}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: b7122f1590c75624ae4eacad689a055c4101b13c\n" }, { "commit": "2bb65e018cc027ee6411c7bbee6fee2d758de3d2", "tree": "8e0442bab681147c6c569343cf7218a8194bee2b", "parents": [ "91ff24e649b761af3710efd370ce349ba575f054" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Fri Jul 24 01:20:45 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 24 01:24:45 2026" }, "message": "Disallow durations \u003c\u003d 0 for HLS segments.\n\nFixed: 536465757\nChange-Id: I0bbcb25930530cdcd26dccf191470f545080faae\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8139903\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nReviewed-by: Phil Yan \u003cphilyan@chromium.org\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Phil Yan \u003cphilyan@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667570}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 9737cef1c175cf6eeadc89830d20b30353a09f8f\n" }, { "commit": "91ff24e649b761af3710efd370ce349ba575f054", "tree": "0541b5bcb1c813fabc8de9e8ac34bc036efb853e", "parents": [ "9568683cf5ed8c19122e0c3c7368a2ae962bad11" ], "author": { "name": "Syed AbuTalib", "email": "lowkey@google.com", "time": "Thu Jul 23 22:30:27 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 22:34:55 2026" }, "message": "Create IAMF flac test files\n\nCreated the same way as the opus files were, but with flac and 96kHz. To\nbe used as testing files for the IAMF prober.\n\nBug: 41486346\nChange-Id: I1c7e9dfb3de1ec26dbc245d5c01fd039c99e742c\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8137682\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1667461}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: dd6f6a5efe9ea011167118cf6108ef4062393bea\n" }, { "commit": "9568683cf5ed8c19122e0c3c7368a2ae962bad11", "tree": "08a1150a392e92eaadb3671146b22be7390e2903", "parents": [ "64ea33fb142e1f5f497674f432f8b82a7d9599ef" ], "author": { "name": "Palak Agarwal", "email": "agpalak@chromium.org", "time": "Thu Jul 23 20:50:38 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 20:54:09 2026" }, "message": "Fix CoreAudio deadlock in PropertyListenerHelper on Mac\n\nPropertyListenerHelper on macOS was registering its CoreAudio property\nlistener block on dispatch_get_main_queue(). When the helper is\ndestroyed, it calls AudioObjectRemovePropertyListenerBlock synchronously\non the main thread of the audio service (audio.CrUtilityMain).\n\nIf CoreAudio needs to synchronize with the main queue to complete the\nremoval while a notification is in flight, it results in a deadlock.\n\nBug: 538143114\nChange-Id: I8a73ee541682bb438b95cc6721fd1d705b78a906\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8138063\nCommit-Queue: Palak Agarwal \u003cagpalak@chromium.org\u003e\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667382}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 77e4d99de7b707292c9a9064ce6282b1439e94f2\n" }, { "commit": "64ea33fb142e1f5f497674f432f8b82a7d9599ef", "tree": "fdbd400efc748c0e2c8b2a0695fcfa27c1bcb085", "parents": [ "41643ad044737fcf92a6aa52bf023059aea6ab39" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Thu Jul 23 19:29:16 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 19:32:53 2026" }, "message": "[media] Keep ref on MemoryTypeTracker in SharedImageReadLock\n\nAdd MemoryTypeTracker ref on SharedImageReadLock to avoid UAF based\non unexpected destruction orders, as the lock can be passed to\nIMFSample. We will eventually get rid of MemoryTypeTracker in\nSharedImage to instead use MemoryTracker directly which is\nRefCountedThreadSafe and should avoid future issues.\n\nNote that this does not pass CommandBufferHelper as we want to get\nrid of it as well eventually.\n\nBug: 537832446\nChange-Id: I048a2e20527748c0e169d6b285d0cab5d6c22fa9\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8139742\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667317}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 38acc8b39421655426c218609986023ad483c241\n" }, { "commit": "41643ad044737fcf92a6aa52bf023059aea6ab39", "tree": "8df4155693890fd469498bd13ced594e70a8a717", "parents": [ "0175021b0b33458e941a300a9dd0972de6bfda8f" ], "author": { "name": "Sam Zackrisson", "email": "saza@chromium.org", "time": "Thu Jul 23 18:29:06 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 18:32:52 2026" }, "message": "ML residual echo estimation: delay initialization\n\nUse an async factory function for creating and initializing the ML-based echo estimator, to reduce the impact on getUserMedia latency.\n\nGuarded by a feature flag to support testing / evaluation.\n\nWebRTC code for delayed init is added in https://webrtc-review.googlesource.com/489580\n\nTested: Manual test with/without feature and verified ML model loads correctly (via perf trace) with good audio quality.\nBug: webrtc:442444736\nChange-Id: I67486dd621a3b22b82d7a7852b82936103cccc61\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8108258\nCommit-Queue: Sam Zackrisson \u003csaza@chromium.org\u003e\nReviewed-by: Mirko Bonadei \u003cmbonadei@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667275}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 010fa365a336a07cf8abb7130a3760f56e8b6867\n" }, { "commit": "0175021b0b33458e941a300a9dd0972de6bfda8f", "tree": "7cdaed348bbaef14501229a3de594954b17a7624", "parents": [ "c6c03aa26995c973f2a972c4ff6981492d8dc3ca" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Thu Jul 23 18:14:39 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 18:19:55 2026" }, "message": "Reland \"Include model inference in VoiceIsolation\"\n\nThis is a reland of commit edb3d697a61645870710438e1920c1cf788552f6\n\nOriginal change\u0027s description:\n\u003e Include model inference in VoiceIsolation\n\u003e\n\u003e Internally VoiceIsolation uses a TfLite model to remove unwanted audio\n\u003e from the incoming audio stream. This change puts the model inference\n\u003e component in the VoiceIsolation processing pipeline.\n\u003e\n\u003e Bug: 512016773\n\u003e Change-Id: I615c62a7860696af882ed3854e114effc6bd7b3a\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8057185\n\u003e Commit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\n\u003e Reviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\n\u003e Reviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1666268}\n\nBug: 512016773\nChange-Id: If281b16c0c486846a1da33a9d11e0f83fab71110\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8138180\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1667262}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 6d68315d003fc2221803acf2dc98fde9f74f43d7\n" }, { "commit": "c6c03aa26995c973f2a972c4ff6981492d8dc3ca", "tree": "1a130817decee9996c45bf5cb7be983a082afae6", "parents": [ "ff4581c12d0739e8550fface78efd0e85c1a583c" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Thu Jul 23 17:31:18 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 17:35:33 2026" }, "message": "Reland \"[gpu] Remove r8_egl_images_broken and disable_r8_shared_images\"\n\nThis is a reland of commit 419f14aef0e2c9684985caba56519d8933678318\n\nDiff from original:\nRemove VideoResourceUpdater test as it is no longer needed.\n\nOriginal change\u0027s description:\n\u003e [gpu] Remove r8_egl_images_broken and disable_r8_shared_images\n\u003e\n\u003e Remove r8_egl_images_broken workaround as it was only for Android\n\u003e version \u003c\u003d9 and now min version is 10. Also remove\n\u003e disable_r8_shared_images and simplify related code.\n\u003e\n\u003e Bug: 536421285\n\u003e Change-Id: I1302a701578dfd677e0673f67f6c6d54d66ed87b\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8132497\n\u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Commit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\n\u003e Reviewed-by: Ari Chivukula \u003carichiv@chromium.org\u003e\n\u003e Reviewed-by: Giovanni Ortuno Urquidi \u003cortuno@chromium.org\u003e\n\u003e Reviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1665704}\n\nBug: 536421285\nChange-Id: I5273ea4a3b88cd9ec7a138d1b679a5cb8fc511cb\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8131031\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nReviewed-by: Ari Chivukula \u003carichiv@chromium.org\u003e\nReviewed-by: Giovanni Ortuno Urquidi \u003cortuno@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1667232}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 79959759ab2b9a14ebce9360f7bbde952ad5b08c\n" }, { "commit": "ff4581c12d0739e8550fface78efd0e85c1a583c", "tree": "65b7a1f0b9a70be5c3148812bc729f0283d5a009", "parents": [ "21adb10979c0817aa760a254357055704cee6b5b" ], "author": { "name": "luci-bisection@appspot.gserviceaccount.com", "email": "luci-bisection@appspot.gserviceaccount.com", "time": "Thu Jul 23 12:56:31 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 13:00:00 2026" }, "message": "Revert \"Remove non-8-bitprofiles from D3D11/D3D12 decoder GUID mapping.\"\n\nThis reverts commit 5fb07ab13dd9c20d3f9cb156d427f6582a1d5671.\n\nReason for revert:\nLUCI Bisection has identified this change as the cause of a test failure. See the analysis: https://ci.chromium.org/ui/p/chromium/bisection/test-analysis/b/5494727047643136\n\nSample build with failed test: https://ci.chromium.org/b/8675527907126740241\nAffected test(s):\n[://media\\:media_unittests!gtest::D3D11VideoDecoderTest#DoesNotSupportsH264HIGH10Profile](https://ci.chromium.org/ui/test/chromium/:%2F%2Fmedia%5C:media_unittests%21gtest::D3D11VideoDecoderTest%23DoesNotSupportsH264HIGH10Profile?q\u003dVHash%3A8423673130d18d17)\n\nIf this is a false positive, please report it at http://b.corp.google.com/createIssue?component\u003d1199205\u0026description\u003dAnalysis%3A+https%3A%2F%2Fci.chromium.org%2Fui%2Fp%2Fchromium%2Fbisection%2Ftest-analysis%2Fb%2F5494727047643136\u0026format\u003dPLAIN\u0026priority\u003dP3\u0026title\u003dWrongly+blamed+https%3A%2F%2Fchromium-review.googlesource.com%2Fc%2Fchromium%2Fsrc%2F%2B%2F8127926\u0026type\u003dBUG\n\nOriginal change\u0027s description:\n\u003e Remove non-8-bitprofiles from D3D11/D3D12 decoder GUID mapping.\n\u003e\n\u003e This fixes a bug where H.264 10-bit, 4:2:2, 4:4:4, and\n\u003e scalable/stereo/multiview profiles were falling through to 8-bit 4:2:0\n\u003e H.264 GUIDs in Windows D3D11 and D3D12 hardware decoders.\n\u003e\n\u003e TAG\u003dagy\n\u003e CONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\u003e\n\u003e Fixed: 536415966\n\u003e Change-Id: I631be73e2e56486720f816d59359903ac8ab0033\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127926\n\u003e Auto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Reviewed-by: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\n\u003e Commit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1666751}\n\u003e\n\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nChange-Id: I232be8c3e5bdc4518c240f4c1980e9b9a40e2ba1\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8137204\nReviewed-by: Ali Hijazi \u003cahijazi@chromium.org\u003e\nOwners-Override: Ali Hijazi \u003cahijazi@chromium.org\u003e\nCommit-Queue: Ali Hijazi \u003cahijazi@chromium.org\u003e\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nCr-Commit-Position: refs/heads/main@{#1667027}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ab9f50469624559ed011b7d764b242470cb5252f\n" }, { "commit": "21adb10979c0817aa760a254357055704cee6b5b", "tree": "71cce7062154fdce23e8e92d82f40693f9d724ba", "parents": [ "09ded7e02145098c2acbaa4307df21349865b858" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Thu Jul 23 00:53:57 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 00:57:51 2026" }, "message": "Remove non-8-bitprofiles from D3D11/D3D12 decoder GUID mapping.\n\nThis fixes a bug where H.264 10-bit, 4:2:2, 4:4:4, and\nscalable/stereo/multiview profiles were falling through to 8-bit 4:2:0\nH.264 GUIDs in Windows D3D11 and D3D12 hardware decoders.\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nFixed: 536415966\nChange-Id: I631be73e2e56486720f816d59359903ac8ab0033\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127926\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666751}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5fb07ab13dd9c20d3f9cb156d427f6582a1d5671\n" }, { "commit": "09ded7e02145098c2acbaa4307df21349865b858", "tree": "65b7a1f0b9a70be5c3148812bc729f0283d5a009", "parents": [ "c93951699f41e2deaf0eedb34be69598af905f63" ], "author": { "name": "Qiu Jianlin", "email": "jianlin.qiu@intel.com", "time": "Thu Jul 23 00:05:43 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 00:10:01 2026" }, "message": "H.265 parser slice header conformance enhancement.\n\nMake sure the slices of a coded H.265 picture bearing same nalu type and\nshort term RPS.\n\nBug: 536446950\nChange-Id: I984e23fe719e43d7d59ef71dad31a6ff701da7b0\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127907\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666720}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 435823d9ef3e2fdd6dfc92b0551181c0e90c27f3\n" }, { "commit": "c93951699f41e2deaf0eedb34be69598af905f63", "tree": "5ef71db206f2ac150765694486b7e1e1ee999a7b", "parents": [ "550182ab61e53824a9b5a0cb90a5c82533462eb3" ], "author": { "name": "Qiu Jianlin", "email": "jianlin.qiu@intel.com", "time": "Thu Jul 23 00:04:04 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 23 00:09:17 2026" }, "message": "Avoid driver UAF on decoder destruction.\n\nA failure to create event for fence wait (due to OOM) shall be treated\nas fatal that cannot be recovered. Crash intentionally to avoid driver\ncontinue using the GPU resources after decoder destruction.\n\nBug: 536427991\nChange-Id: I9174583397e4bee1b5165ced64776be7ec4e5cef\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127234\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666717}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 88448df4491c973dd24170e0ec7a53d70c12e3f6\n" }, { "commit": "550182ab61e53824a9b5a0cb90a5c82533462eb3", "tree": "905cabbb2242bdd7e12026bd71d873b043b5d715", "parents": [ "491643f5084c9136e3ec929c5fd48728e3c39db4" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Wed Jul 22 23:15:03 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 23:19:15 2026" }, "message": "aaudio: Keep callback helper alive past AAudioStream_close()\n\nOn Android R and earlier, AAudio may continue dispatching data and error\ncallbacks after AAudioStream_requestStop() and AAudioStream_close() have\nreturned. AAudioDestructionHelper is registered with AAudio as bare\nvoid* user_data and was previously freed immediately after\nAAudioStream_close() ran in its destructor, which also read GUARDED_BY\nmembers without acquiring the lock.\n\nMove the close call into a dedicated CloseDeferredStream() method that\ntakes the lock, and on Android R- schedule it as a delayed task separate\nfrom (and strictly before) the task that frees the helper, so the helper\nremains valid for any callbacks dispatched during or shortly after\nclose. The helper\u0027s destructor now closes the stream as a fallback if\nthe delayed close task is dropped on sequence shutdown.\n\nThis is similar in spirit to commit\n279912fa38186a0480f4a6068d3ff7bc422e0afc but posts both delayed tasks up\nfront (no nested PostDelayedTask) and tolerates the owning task runner\nbeing torn down with the tasks still pending, which avoids the\nAudioAndroidInputTest crashes that caused that change to be reverted.\n\nAdds aaudio_stream_wrapper_unittest.cc covering the deferred teardown\nsequence.\n\nBug: 501331457\nChange-Id: I497e6d0bd0c57fda6756a5d3ebaf9bef7b08da31\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127809\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666690}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 79ae7859fd23bf4affea42a7f5d89d0a6d1ba1a3\n" }, { "commit": "491643f5084c9136e3ec929c5fd48728e3c39db4", "tree": "8fd9839bb4552ec9b97ac8d902979bf243ee6fe7", "parents": [ "94e2c74917d12ba000c91a0680cd06ffc265ea07" ], "author": { "name": "Hailey Wang", "email": "haileywang@google.com", "time": "Wed Jul 22 21:05:22 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 21:09:38 2026" }, "message": "Revert \"Enable kAAudioPerStreamDeviceSelection by default\"\n\nThis reverts commit b725e6c6c12d58a0c801d2537170a30f1d8464dc.\n\nReason for revert: Causing test failure in external/wpt/audio-output/idlharness.https.window.html\n\nFailure Link: see b/537839417\n\nOriginal change\u0027s description:\n\u003e Enable kAAudioPerStreamDeviceSelection by default\n\u003e\n\u003e This CL enables the kAAudioPerStreamDeviceSelection flag by default, but\n\u003e gates the feature for desktop versions of android.\n\u003e\n\u003e This matches the current behavior of the Finch experiments, currently\n\u003e rolled out to 100% on desktop Android devices.\n\u003e\n\u003e Bug: 425608358\n\u003e Change-Id: Id8f929dfd19a64c683941b04ab5a71d83cdcc109\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8111797\n\u003e Commit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\n\u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1665782}\n\nBug: 425608358\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nChange-Id: I092beb5095736cec86775ad1a32fadf23d3000aa\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8135260\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nCommit-Queue: Hailey Wang \u003chaileywang@google.com\u003e\nOwners-Override: Hailey Wang \u003chaileywang@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666603}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: de92b900b2385efee8e9b2cba2bcdaa595487c4b\n" }, { "commit": "94e2c74917d12ba000c91a0680cd06ffc265ea07", "tree": "bde20392938f68e77e9ed3d2de7be64365f9ad1a", "parents": [ "ecd56cbbb3796e211726584ac33dbf2684fee65f" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 22 17:41:34 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 17:45:31 2026" }, "message": "Fix integer overflow in bitstream converters size calculation\n\nUse base::CheckedNumeric\u003cuint32_t\u003e when accumulating NAL unit sizes in\nCalculateNeededOutputBufferSize() for both\nH264ToAnnexBBitstreamConverter and H265ToAnnexBBitstreamConverter to\navoid unsigned integer overflow.\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nBug: 536438315\nChange-Id: I10062e171b912e22680da792c663e610cf4f7eca\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128620\nReviewed-by: Syed AbuTalib \u003clowkey@google.com\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666460}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 61632d4c0d8d08ce3edd41c8bed831459319704e\n" }, { "commit": "ecd56cbbb3796e211726584ac33dbf2684fee65f", "tree": "07f8766d5dbdb386c297efeb1b1557b08f6ffffc", "parents": [ "a067eaea9cce4faf626917a215a15e1a73273949" ], "author": { "name": "luci-bisection@appspot.gserviceaccount.com", "email": "luci-bisection@appspot.gserviceaccount.com", "time": "Wed Jul 22 14:36:36 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 14:40:06 2026" }, "message": "Revert \"Include model inference in VoiceIsolation\"\n\nThis reverts commit edb3d697a61645870710438e1920c1cf788552f6.\n\nReason for revert:\nLUCI Bisection has identified this change as the culprit of a build failure. See the analysis: https://ci.chromium.org/ui/p/chromium/bisection/compile-analysis/b/8675568641486648657\n\nSample failed build: https://ci.chromium.org/b/8675568641486648657\n\nIf this is a false positive, please report it at http://b.corp.google.com/createIssue?component\u003d1199205\u0026description\u003dAnalysis%3A+https%3A%2F%2Fchromium-review.googlesource.com%2Fc%2Fchromium%2Fsrc%2F%2B%2F8057185\u0026format\u003dPLAIN\u0026priority\u003dP3\u0026title\u003dWrongly+blamed+https%3A%2F%2Fci.chromium.org%2Fui%2Fp%2Fchromium%2Fbisection%2Fcompile-analysis%2Fb%2F8675568641486648657\u0026type\u003dBUG\n\nOriginal change\u0027s description:\n\u003e Include model inference in VoiceIsolation\n\u003e\n\u003e Internally VoiceIsolation uses a TfLite model to remove unwanted audio\n\u003e from the incoming audio stream. This change puts the model inference\n\u003e component in the VoiceIsolation processing pipeline.\n\u003e\n\u003e Bug: 512016773\n\u003e Change-Id: I615c62a7860696af882ed3854e114effc6bd7b3a\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8057185\n\u003e Commit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\n\u003e Reviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\n\u003e Reviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1666268}\n\u003e\n\nBug: 512016773\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nChange-Id: Id0ee1e698cf94c79db1d411378e166c45b48dbb1\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8134760\nCommit-Queue: luci-bisection@appspot.gserviceaccount.com \u003cluci-bisection@appspot.gserviceaccount.com\u003e\nOwners-Override: luci-bisection@appspot.gserviceaccount.com \u003cluci-bisection@appspot.gserviceaccount.com\u003e\nBot-Commit: luci-bisection@appspot.gserviceaccount.com \u003cluci-bisection@appspot.gserviceaccount.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666273}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 75a19930b2f307513e6283bef659e79d48042086\n" }, { "commit": "a067eaea9cce4faf626917a215a15e1a73273949", "tree": "e3c9719d71e17f86d60d23a2879d0e0bd9aebf34", "parents": [ "84c4995ecaf8b05cb80d537c736db639aecb6d46" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Wed Jul 22 14:17:46 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 14:22:18 2026" }, "message": "Include model inference in VoiceIsolation\n\nInternally VoiceIsolation uses a TfLite model to remove unwanted audio\nfrom the incoming audio stream. This change puts the model inference\ncomponent in the VoiceIsolation processing pipeline.\n\nBug: 512016773\nChange-Id: I615c62a7860696af882ed3854e114effc6bd7b3a\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8057185\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nReviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\nReviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1666268}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: edb3d697a61645870710438e1920c1cf788552f6\n" }, { "commit": "84c4995ecaf8b05cb80d537c736db639aecb6d46", "tree": "07f8766d5dbdb386c297efeb1b1557b08f6ffffc", "parents": [ "04c708868f8936d79959815fe97464e8ce4a3972" ], "author": { "name": "Paul Sobek", "email": "sobek@google.com", "time": "Wed Jul 22 12:57:43 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 13:01:13 2026" }, "message": "Enable kSkipMediaCodecReallocation by default\n\nThis CL promotes the kSkipMediaCodecReallocation feature to be\nenabled by default.\n\nFinch experiment results showed reduction in media codec realloc.\n\nBug: 489426333\nChange-Id: Icd2215011aa1726e282946279b9e3aa1d70eda2d\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8131818\nCommit-Queue: Paul Sobek \u003csobek@google.com\u003e\nReviewed-by: Alex Mitra \u003calexmitra@chromium.org\u003e\nAuto-Submit: Paul Sobek \u003csobek@google.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666209}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 3ffdea370de8d4a4d9908ded7c4cb17605f6f5f4\n" }, { "commit": "04c708868f8936d79959815fe97464e8ce4a3972", "tree": "687d861eb93314b093323f0eac7b3579e8886d68", "parents": [ "fe6126c4246f1b2a53777b2606fe1d8f07f38fc3" ], "author": { "name": "Gyuyoung Kim", "email": "gyuyoung@igalia.com", "time": "Wed Jul 22 08:36:10 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 08:39:42 2026" }, "message": "midi: Enable kMidiMacUmp switch in Apple platforms\n\nhttps://crrev.com/c/8016282 introduced kMidiMacUmp switch only for Mac,\nbut the switch has used by MidiManagerMac::InitializeCoreMIDI() that\nworks on Apple platforms. Thus, we need to enable the switch for the\nother Apple platforms. This is just to avoid the build error without any\nbehavior changes.\n\nBug: 327611574\nChange-Id: I3ac017f66a636cec2e017f465c399f769db544d4\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8129720\nCommit-Queue: Gyuyoung Kim \u003cgyuyoung@igalia.com\u003e\nReviewed-by: Takashi Toyoshima \u003ctoyoshim@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666094}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: fe86926b29af55e7a24e4b12674229b5dfd2289f\n" }, { "commit": "fe6126c4246f1b2a53777b2606fe1d8f07f38fc3", "tree": "2928d3fdfdf34a548ab21335ce55e3aca6a4b9c9", "parents": [ "a39223a4e900cbf6eb138513bc65232d4f5aabc5" ], "author": { "name": "Yu-Hsuan Hsu", "email": "yuhsuan@google.com", "time": "Wed Jul 22 08:29:26 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 08:35:15 2026" }, "message": "media/audio/cras: Use persistent proxy for CrasInputStream\n\nThis CL refactors CrasInputStream to use a single persistent proxy\n(CrasAudioInputStreamProxy) created in the constructor, aligning it with\nthe robust design of CrasUnifiedStream.\n\nPreviously, CrasInputStream recreated the proxy on every Start() and\nonly destroyed it in Close(). If the stream was restarted (Stop() then\nStart() without Close()), the old proxy was deleted. A late callback\nfrom the previous run could still reference the old proxy, causing a\npotential use-after-free (UAF) on the proxy object itself.\n\nBy creating the proxy once in the constructor and using Attach()/Detach()\nduring Start()/Stop(), we ensure the proxy survives restart cycles and\nsafely drops late callbacks.\n\nAlso adds an unittest to verify that restarting the stream works\ncorrectly and continues to deliver callbacks.\n\nBug: 519613438\n\nChange-Id: I7d8306075d025644f63852489905aa341b76e6b8\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8004835\nAuto-Submit: Yu-Hsuan Hsu \u003cyuhsuan@chromium.org\u003e\nReviewed-by: Li-Yu Yu \u003caaronyu@google.com\u003e\nCommit-Queue: Li-Yu Yu \u003caaronyu@google.com\u003e\nCommit-Queue: Yu-Hsuan Hsu \u003cyuhsuan@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666091}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e9c05a23db323b9226fa5cdd869cd51cd060f83b\n" }, { "commit": "a39223a4e900cbf6eb138513bc65232d4f5aabc5", "tree": "b3414d0937f31ded1a27769ca827fafa3369b9f9", "parents": [ "33b39ed03025d691cb30d68108166ba38edd5781" ], "author": { "name": "Kelsen Liu", "email": "kelsenliu@chromium.org", "time": "Wed Jul 22 07:05:08 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 07:09:30 2026" }, "message": "media: Spanify WebM TracksBuilder serialization\n\nChange the serialization helpers and Track::Write()/WriteTracks() from\na raw pointer+size pair to base::SpanWriter\u003cuint8_t\u003e, removing\nUNSAFE_TODO pointer arithmetic.\n\nBug: 40284755\nChange-Id: I43f10e30d643e64a726cdcf64d2aa4e21bcfc480\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8122354\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Kelsen Liu \u003ckelsenliu@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1666060}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: a3d4a2ef2ddef172f6ba4e36e199ad2a5f82b572\n" }, { "commit": "33b39ed03025d691cb30d68108166ba38edd5781", "tree": "6bab5cef31548dd1e65bcec0efafbbf00a93ec84", "parents": [ "688621234c80589d04b6ea02b3311a05cf00b507" ], "author": { "name": "Takashi Toyoshima", "email": "toyoshim@chromium.org", "time": "Wed Jul 22 03:02:52 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 03:05:57 2026" }, "message": "midi: Refactor Mac MIDI manager to use platform-neutral UMP utility\n\nThis CL refactors the macOS MIDI manager to support the new CoreMIDI\nMIDIEventList (Universal MIDI Packet, UMP) APIs introduced in macOS 11,\nwhile fixing deprecation warnings of the old MIDIPacketList APIs.\n\nTo achieve this, we introduce a platform-neutral UMP utility\n`ump_message_util` that handles:\n- Parsing UMP word streams into legacy MIDI 1.0 messages.\n- Translating legacy MIDI 1.0 messages into UMP word streams.\n\nThis utility is platform-independent, allowing it to be compiled and\ntested on all platforms (including Linux). We also add comprehensive\nunit tests for this utility and two new LibFuzzer targets to ensure\nthe robustness of the parser and encoder.\n\nThe macOS MIDI manager (`MidiManagerMac`) is updated to use this utility\nto convert between UMP (used by the new CoreMIDI APIs) and legacy MIDI\n(used by the Web MIDI API).\n\nBug: 327611574\nChange-Id: Iad3b9b3635a76fe8c31cdd161b4027cd102f87cd\nBypass-Check-License: rename\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8016282\nReviewed-by: Giovanni Ortuno Urquidi \u003cortuno@chromium.org\u003e\nCommit-Queue: Takashi Toyoshima \u003ctoyoshim@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665988}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1bdace95d67c330798fa56a265eefd3701d08ff6\n" }, { "commit": "688621234c80589d04b6ea02b3311a05cf00b507", "tree": "fbd9f9800ec25b19baaf72afd9a32419b3d8a1c5", "parents": [ "0c8377b1a111b7773b9aedc5a58fcf8df3390636" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 22 01:43:04 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 01:46:35 2026" }, "message": "Ensure visible_rect \u003c\u003d coded_size in various decoders and accelerators\n\nIn D3D11VideoDecoder::DoDecode(), consume GetVisibleRect() on config\nchange to preserve the visible_rect \u003c\u003d coded_size invariant.\n\nIn H264Decoder::ProcessSPS(), fall back to full picture size if the SPS\ncrop rect is empty or exceeds picture dimensions.\n\nIn H265Decoder::ProcessSPS(), add CHECK assertions verifying\nvisible_rect fits within picture size and is non-empty (as guaranteed by\nH265Parser).\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nFixed: 536432454\nChange-Id: Ibae29bb933cb42c6e2a3ea58e61522bf928c3551\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128840\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665967}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 294fb05b7bdb5a08b209d9bf69af42dbe50b3a23\n" }, { "commit": "0c8377b1a111b7773b9aedc5a58fcf8df3390636", "tree": "35c0eb26606d404574da7c361b5a244a36c63106", "parents": [ "7e830458e058a6e16ff9c0bf69736fa6079c92b8" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 22 00:57:23 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 01:01:12 2026" }, "message": "Fix size narrowing size_t to int in FFmpegAudio/VideoDecoder\n\nUse base::checked_cast\u003cint\u003e when assigning buffer_span.size() to\npacket-\u003esize in FFmpegAudioDecoder::FFmpegDecode and\nFFmpegVideoDecoder::FFmpegDecode to prevent safe narrowing overflow.\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nFixed: 536443672\nChange-Id: Iccb23d69a29797a1e25a1ece47634f99dfd8de71\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127445\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665945}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 02be67a16a69e0b02ff03afd68f600a6b24a9c50\n" }, { "commit": "7e830458e058a6e16ff9c0bf69736fa6079c92b8", "tree": "1ce029c7e5db393f384381b9020b63ae98ef36d3", "parents": [ "45cad9bbf8032f8502a334d33c7896759d01e775" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Wed Jul 22 00:50:17 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 00:53:49 2026" }, "message": "Reject 0 channel AudioEncoderConfig over mojo\n\nThis CL adds a verification to reject incoming AudioEncoderConfigs with\n0 channels, rather than letting them CHECK further downstream.\n\nThis CL also adds UTs to test the relevant mojo traits.\n\nFixed: 536420562\nChange-Id: I227f9186130ed6fba21638015006bd821fdac3ad\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127302\nReviewed-by: Will Harris \u003cwfh@chromium.org\u003e\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nAuto-Submit: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCommit-Queue: Will Harris \u003cwfh@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665943}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e9a8200c4c70873afc9321e9602fa95883861128\n" }, { "commit": "45cad9bbf8032f8502a334d33c7896759d01e775", "tree": "f1a629953a1b9e344b535b9fbcc9d8e541a1a5df", "parents": [ "7ad4e5642720cbc9932b127e438d125b732afd2d" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 22 00:17:00 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 00:20:37 2026" }, "message": "Check CreateFormat return in VideoToolboxH265Accelerator::SubmitDecode\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nFixed: 536417275\nChange-Id: Id515ac68f2a99725b90ed0835f2ea2004059594d\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128210\nReviewed-by: Syed AbuTalib \u003clowkey@google.com\u003e\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665927}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 6586259b4fa1291c96a640b371a24634a8ecc52f\n" }, { "commit": "7ad4e5642720cbc9932b127e438d125b732afd2d", "tree": "a65f2ec157ef74f8b5a8f9ef2d36f43387be47ee", "parents": [ "5bfac96d4903bf91af1da898e709a257d1c885d9" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Wed Jul 22 00:12:10 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Wed Jul 22 00:16:12 2026" }, "message": "h265: Finish previous frame before updating alpha-layer SPS/PPS\n\nWhen an alpha-layer (non-zero nuh_layer_id) SPS or PPS NAL unit is\nencountered during H.265 decoding, FinishPrevFrameIfPresent() was not\nbeing called prior to parsing the parameter set and forwarding it to the\naccelerator. If an alpha-layer SPS or PPS arrived mid-picture, updating\nparameter sets before completing the current picture could cause state\ncollisions or unvalidated parameter set updates in hardware accelerators\n(such as VideoToolbox on macOS).\n\nThis change adds CHECK_ACCELERATOR_RESULT(FinishPrevFrameIfPresent()) to\nthe case H265NALU::SPS_NUT and case H265NALU::PPS_NUT branches within\nthe non-zero nuh_layer_id handling block in H265Decoder::Decode().\n\nAlso adds a unit test coverage in H265DecoderTest verifying that\nFinishPrevFrameIfPresent() completes the pending picture before\nprocessing the alpha-layer SPS/PPS.\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nBug: 536444242\nChange-Id: I8d0eeb6b7221cd8ae4636c6e79782288d198131b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128206\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCommit-Queue: Qiu, Jianlin \u003cjianlin.qiu@intel.com\u003e\nCr-Commit-Position: refs/heads/main@{#1665919}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 7af6810a344b2cf35e47ec0c2f833ad0adfedf90\n" }, { "commit": "5bfac96d4903bf91af1da898e709a257d1c885d9", "tree": "447e21f0f1faa969f1aa055b8fb36c204ba2223f", "parents": [ "ecd48d5efdad843a104d95345dfe521587d50c63" ], "author": { "name": "Syed AbuTalib", "email": "lowkey@google.com", "time": "Tue Jul 21 23:49:02 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 23:52:53 2026" }, "message": "ValidateLayout() now takes a ChannelLayoutConfig\n\nWe no longer have to guard against DISCRETE when calling\n`ValidateLayout` if we provide a `ChannelLayoutConfig` instead of\n`ChannelLayout`, due to CLC having its own strict requirements (i.e.,\nDISCRETE can not be associated with zero channels).\n\nBug: 483461268\nChange-Id: I6ae9958c1544cb86073f0e28c511b8be9373cd0f\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7944387\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1665902}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e932d134bfa7bba27d1c35e0dcf41265c2f09afd\n" }, { "commit": "ecd48d5efdad843a104d95345dfe521587d50c63", "tree": "5989e084a04b04486151a9902860bb00894f60c9", "parents": [ "beafccc252372af39f8ede880dc1713d7d1a92dc" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Tue Jul 21 23:01:32 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 23:05:11 2026" }, "message": "media: Enforce bounds check on DecryptConfig.subsamples count\n\nThis CL introduces media::limits::kMaxSubsamplesPerBuffer (65535,\nmatching the ISO 23001-7 senc wire bounds) and enforces it directly\nduring Mojo deserialization in media_type_converters.cc. A check is\nalso added to MediaCodecBridgeImpl to protect the JNI boundary.\n\nBug: 536458594\nChange-Id: I1d8d91c5b949a0595aef26bb001ebe7f220322b6\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128096\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Elias Klim \u003celklm@chromium.org\u003e\nReviewed-by: Will Harris \u003cwfh@chromium.org\u003e\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665870}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: d5c61fe79ab3ad26afcb94b1eebc464f4edc9c66\n" }, { "commit": "beafccc252372af39f8ede880dc1713d7d1a92dc", "tree": "63ee7277a2f29366aae4247c61e738f0d317452d", "parents": [ "b456c595c4c92633ce311a82feca8e441fad47c7" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Tue Jul 21 22:13:33 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 22:16:48 2026" }, "message": "media: Catch SessionException during EME GenerateRequest\n\nThis CL adds a catch block directly for `MediaDrm.SessionException` in\ngetKeyRequest().\n\nThis catches transient resource errors without crashing the GPU process,\nallowing the Mojo promise to be gracefully rejected so the media player\ncan retry.\n\nSimilar to 71e803d4cba16140e8bbe98460e50b7eed7403b9, except we do not\nimplement transient detection, and we\u0027d rather just catch and let player\nretry on its own.\n\nBug: 537329064\nChange-Id: I68976dc010c7236b4bb1b97a059cd690a91fe737\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8129763\nReviewed-by: Frank Liberato \u003cliberato@chromium.org\u003e\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nAuto-Submit: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665830}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 7bc00f107323f31d7c3923fee08239cff367298b\n" }, { "commit": "b456c595c4c92633ce311a82feca8e441fad47c7", "tree": "ae563a40575d0c5c6af174744bca84dbf1f2aeab", "parents": [ "aa3ee56f61f3b766fbf7cb9fbe3c77cbd5b611a8" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Tue Jul 21 22:07:38 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 22:11:25 2026" }, "message": "Fix a few different validation issues with h264 parsing\n\nThis adds range validation for a few different fields to avoid\nblindly passing potentially bad data out to drivers.\n\nFixed: 536428615\nChange-Id: I8d945cc477ddca47b57dc28df7f66beb4442b106\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128120\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665828}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ad64b2c8fdc096eb8ce2e555b70a7f44bee0e207\n" }, { "commit": "aa3ee56f61f3b766fbf7cb9fbe3c77cbd5b611a8", "tree": "a8c6cc087fedf97ab6a8f16bf93d63adacbb8e28", "parents": [ "4715f001a5e2b9870aa5cd840c9a217c89185082" ], "author": { "name": "Thomas Guilbert", "email": "tguilbert@chromium.org", "time": "Tue Jul 21 21:06:55 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 21:10:51 2026" }, "message": "Enable kAAudioPerStreamDeviceSelection by default\n\nThis CL enables the kAAudioPerStreamDeviceSelection flag by default, but\ngates the feature for desktop versions of android.\n\nThis matches the current behavior of the Finch experiments, currently\nrolled out to 100% on desktop Android devices.\n\nBug: 425608358\nChange-Id: Id8f929dfd19a64c683941b04ab5a71d83cdcc109\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8111797\nCommit-Queue: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665782}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: b725e6c6c12d58a0c801d2537170a30f1d8464dc\n" }, { "commit": "4715f001a5e2b9870aa5cd840c9a217c89185082", "tree": "0d24f1933d9a15e876854124a6c47d8c75d53f44", "parents": [ "a2926b8d9a7a766964ec9088e89d55e39da53489" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Tue Jul 21 20:33:09 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 20:37:45 2026" }, "message": "Reland \"media: Migrate EncryptionScheme to Mojo traits\"\n\nThis is a reland of commit ddf49493c7141b0dc62d6f79af506924dccb5b0e\n\nThe CL is not the cause of that failure. The failure has existed\nsince b/531170669 and is flaky.\n\nOriginal change\u0027s description:\n\u003e media: Migrate EncryptionScheme to Mojo traits\n\u003e\n\u003e This CL migrates EncryptionScheme to Mojo EnumTraits.\n\u003e\n\u003e Bug: 393179188\n\u003e Change-Id: I3f5a78fc501af796df8fae2de9b2b7e885e5e6bf\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100522\n\u003e Reviewed-by: Daniel Cheng \u003cdcheng@chromium.org\u003e\n\u003e Reviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\n\u003e Commit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1664984}\n\nBug: 393179188\nChange-Id: Ife58e931db5929420d2cb04347591bc4c25cb8a8\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8126992\nReviewed-by: Daniel Cheng \u003cdcheng@chromium.org\u003e\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665745}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ab6980b3294f8e384d08d7c3f5f9e25c2441e17c\n" }, { "commit": "a2926b8d9a7a766964ec9088e89d55e39da53489", "tree": "45a8b8ffc81fe0a68c7b606f1bc3534833e2ec9a", "parents": [ "27b3047610ddfbae2113798c2cef24cc5492b06d" ], "author": { "name": "Dmytro Yeroshkin", "email": "yeroshkin@chromium.org", "time": "Tue Jul 21 20:30:57 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 20:34:27 2026" }, "message": "Revert \"[gpu] Remove r8_egl_images_broken and disable_r8_shared_images\"\n\nThis reverts commit 419f14aef0e2c9684985caba56519d8933678318.\n\nReason for revert: Broke linux builds\n\nFailure Link: https://ci.chromium.org/ui/p/chromium/builders/ci/linux-cast-x64-dbg/14324/overview\n\nOriginal change\u0027s description:\n\u003e [gpu] Remove r8_egl_images_broken and disable_r8_shared_images\n\u003e\n\u003e Remove r8_egl_images_broken workaround as it was only for Android\n\u003e version \u003c\u003d9 and now min version is 10. Also remove\n\u003e disable_r8_shared_images and simplify related code.\n\u003e\n\u003e Bug: 536421285\n\u003e Change-Id: I1302a701578dfd677e0673f67f6c6d54d66ed87b\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8132497\n\u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Commit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\n\u003e Reviewed-by: Ari Chivukula \u003carichiv@chromium.org\u003e\n\u003e Reviewed-by: Giovanni Ortuno Urquidi \u003cortuno@chromium.org\u003e\n\u003e Reviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1665704}\n\nBug: 536421285\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nChange-Id: Ib8ebc9a1a673e3915eb55e3b359b3a7d3ab5d240\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8132942\nOwners-Override: Dmytro Yeroshkin \u003cyeroshkin@chromium.org\u003e\nCommit-Queue: Dmytro Yeroshkin \u003cyeroshkin@chromium.org\u003e\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665742}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5ded30f43b4e54f511aa24cbd91d62a4f40fbc5b\n" }, { "commit": "27b3047610ddfbae2113798c2cef24cc5492b06d", "tree": "8663dbf087b6ea87737bd8f7460505e0aeedc83c", "parents": [ "097ad76f7e6452babafcfd4d3ca63b8abcc5886c" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 21 20:24:29 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 20:28:09 2026" }, "message": "[media] Reorder CommandBufferHelper member in MediaFoundationVEAWin\n\nReorder CommandBufferHelper member declaration in\nMediaFoundationVideoEncodeAcceleratorWin so that it gets destroyed\nafter pending_input_queue_ which needs helper\u0027s MemoryTracker on\ndestruction. Also clear the queue proactively on destruction.\n\nBug: 536449742\nChange-Id: Ieb79852aefc62ef262410bdfcb6af6e480b72e97\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8130565\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665731}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 83b9159adc9de65630b00dc86d9722b68227dca7\n" }, { "commit": "097ad76f7e6452babafcfd4d3ca63b8abcc5886c", "tree": "bfa9c397e2cb12f4aacd28e780e20314afb744f1", "parents": [ "f561606daf98df299d88b5d582710a49d693ff6b" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 21 19:59:44 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 20:04:45 2026" }, "message": "[gpu] Remove r8_egl_images_broken and disable_r8_shared_images\n\nRemove r8_egl_images_broken workaround as it was only for Android\nversion \u003c\u003d9 and now min version is 10. Also remove\ndisable_r8_shared_images and simplify related code.\n\nBug: 536421285\nChange-Id: I1302a701578dfd677e0673f67f6c6d54d66ed87b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8132497\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Ari Chivukula \u003carichiv@chromium.org\u003e\nReviewed-by: Giovanni Ortuno Urquidi \u003cortuno@chromium.org\u003e\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665704}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 419f14aef0e2c9684985caba56519d8933678318\n" }, { "commit": "f561606daf98df299d88b5d582710a49d693ff6b", "tree": "31fc86cce82e01ff902ebe9ff7ec44165f8fb6bc", "parents": [ "77201d6553084c5e5cbefbbed4caa037b897f108" ], "author": { "name": "Ilya Nikolaevskiy", "email": "ilnik@chromium.org", "time": "Tue Jul 21 19:57:41 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 20:01:35 2026" }, "message": "Add debug logging of video color space to video capture pipeline\n\nThis is a temporary debug logging. base::Feature is used instead of the command-line flag because it\u0027s easier to pipe this to all the parts of the pipeline.\n\nFuchsia-Binary-Size: Increase is temporary.\nBug: b:507155624\nChange-Id: Id708f6ae342cebaacae961e9fcd0050d358e3384\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8124454\nCommit-Queue: Jeremy Roman \u003cjbroman@chromium.org\u003e\nReviewed-by: Colin Blundell \u003cblundell@chromium.org\u003e\nReviewed-by: Jeremy Roman \u003cjbroman@chromium.org\u003e\nAuto-Submit: Ilya Nikolaevskiy \u003cilnik@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665701}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: afda56f28b99b94a8e9b5d4f55237ac7af77ddb4\n" }, { "commit": "77201d6553084c5e5cbefbbed4caa037b897f108", "tree": "cdd2b24a60624024ee0a4bddad6983d4f6a9cc4e", "parents": [ "5bf26cd9cbe4285277b1d96098661a718039f4cb" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Tue Jul 21 17:30:44 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 17:33:45 2026" }, "message": "Fix 8-bit to k16F type confusion in VideoResourceUpdater\n\nIn VideoResourceUpdater::SupportedMultiPlaneChannelFormat(), when\nbits_per_channel \u003c\u003d 8 and k8 channel format is not supported, return\nstd::nullopt immediately instead of falling through to k16/k16F. Falling\nthrough allowed 8-bit frames to be mapped to k16F textures, causing\nWriteYUVPixelsForAllPlanesToTexture() to read 16-bit float pixels from\n8-bit pixel buffers, leading to a potential heap out-of-bounds read.\n\nAdditionally, add a CHECK_GT(bits_per_channel, 8u) assertion in\nWriteYUVPixelsForAllPlanesToTexture() when is_16bit_float is true, and\nadd a unit test EightBitFrameForcedToF16DoesNotOverread.\n\nTAG\u003dagy\nCONV\u003d4eee7864-0106-476f-b647-78eba7f8c730\n\nBug: 536421285\nChange-Id: I6c8ed9bcdadc7a340639113cbc8032773fa82a44\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8128372\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665604}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1ab8a80b7593885d06d31204c595d9f9358e56be\n" }, { "commit": "5bf26cd9cbe4285277b1d96098661a718039f4cb", "tree": "bb3b2cde0b9b2690dcc0ba86fe5f4807dff8f130", "parents": [ "bcf2952a731d47db06ed5884101757c92de5ad94" ], "author": { "name": "Piet Schouten", "email": "Piet.Schouten@microsoft.com", "time": "Tue Jul 21 16:49:04 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 16:52:37 2026" }, "message": "media: Remove unnecessary is_os_cdm code\n\nNow that the Widevine hardware CDM has been removed from the Media\nFoundation CDM code path, the only non-OS CDM the MediaFoundationService\ncapability query could describe is gone, so it always runs for the OS\nPlayReady CDM. Remove the is_os_cdm related code from\nMediaFoundationService and the service broker.\n\nAlso drop the Widevine \"HW_SECURE_ALL\" robustness branch in\nCreateMediaFoundationCdm. Hardware-secure robustness is now always\nPlayReady \"3000\".\n\nKeep MediaFoundationCdmModule::IsOsCdm() in the CDM creation/runtime\npath since it is still required by the path loaded Media Foundation\nClear Key test CDM.\n\nBug: 522626431\nChange-Id: Ica884aeba9f4309c94b6b5148520d27589f5b317\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8091537\nReviewed-by: Sangbaek Park \u003csangbaekpark@chromium.org\u003e\nReviewed-by: Evan Liu \u003cevliu@google.com\u003e\nCommit-Queue: Piet Schouten \u003cPiet.Schouten@microsoft.com\u003e\nReviewed-by: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665567}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 33aa725d4de47a381104c7fcf938b026108e6692\n" }, { "commit": "bcf2952a731d47db06ed5884101757c92de5ad94", "tree": "17d8e861e9d8f160f224104239b33ca4a655a6a2", "parents": [ "4f3c2c6aae20fca9b761305863ad79c0776cca7e" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Tue Jul 21 14:11:37 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 14:14:56 2026" }, "message": "[media] Add DumpWillBeCheck for format mismatch in WrapSharedImage\n\nAll callers of WrapSharedImage should be passing similar formats.\nThis change adds a DUMP_WILL_BE_CHECK with crash keys to fish out\nanything that missed before upgrading to CHECK.\n\nAlso update tests that had mismatches.\n\nBug: 492116792\nChange-Id: Ia30d85f0c55a08207ac49c99780281d40fae24ac\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8102889\nReviewed-by: Bo Liu \u003cboliu@chromium.org\u003e\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Vasiliy Telezhnikov \u003cvasilyt@chromium.org\u003e\nReviewed-by: Ken Buchanan \u003ckenrb@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665452}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1f4ef2b0adf239946fcafb5c06274a925220bfdd\n" }, { "commit": "4f3c2c6aae20fca9b761305863ad79c0776cca7e", "tree": "94bab3e80c91fe7193ca7b4ed933581e7c6cb5e8", "parents": [ "305b100fe25653f05b860f29f82729b14c35f164" ], "author": { "name": "Alex Gough", "email": "ajgo@chromium.org", "time": "Tue Jul 21 13:57:17 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 14:00:59 2026" }, "message": "Remove xoogler muyaoxu from OWNERS\n\nBug: N/A\nChange-Id: Ie106e864b94ac008d3324f141bc5481eee3e6337\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8110257\nReviewed-by: Jordan Bayles \u003cjophba@chromium.org\u003e\nReviewed-by: Jan Keitel \u003cjkeitel@google.com\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nReviewed-by: Tommy Steimel \u003csteimel@chromium.org\u003e\nCommit-Queue: Alex Gough \u003cajgo@chromium.org\u003e\nReviewed-by: Thomas Guilbert \u003ctguilbert@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665441}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1990bab08179bd8e6e067d65850ec917cc1ecc2b\n" }, { "commit": "305b100fe25653f05b860f29f82729b14c35f164", "tree": "41d0b2bbade8c50cf68a2e97fbd6557cf3abe44f", "parents": [ "4aace74d75c0cc1e59b00d07e1252e83ad494fe7" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Tue Jul 21 07:52:42 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 07:59:07 2026" }, "message": "Include STFT step for VoiceIsolation\n\nThe voice isolation model requires frequency-domain features,\nnecessitating a transformation of the incoming time-domain audio signal.\nThis component uses a STFT utilizing a Hann window with a 50% overlap\n(hop size equal to half the FFT size). For the Fourier transform it uses\npffft library.\n\nBug: 512016773\nChange-Id: I21d32bab76f52e41df75cbb94386bc5bf8120879\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8057184\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nReviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665251}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e84bd95e06ba1b26c7587ed6556d38e087083567\n" }, { "commit": "4aace74d75c0cc1e59b00d07e1252e83ad494fe7", "tree": "ef766d82b03d18964612766681b5277296b05533", "parents": [ "c6357ae6b83d73d08141cd815eeccc1a0f75b3f9" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Tue Jul 21 07:51:29 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 07:55:39 2026" }, "message": "voice_isolation_wav_tool test tool\n\nThis tool reads a wav file from disk, processes it using voice_isolation\nmodel (if available) and writes the result back to disk. This can be\nused to test voice_isolation effect on isolated wav_files or in\ncombination with audioproc_f from WebRTC.\n\nBug: 512016773\nChange-Id: I711c4f5b60e077029f6e135305e2f734278eb70f\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8052192\nReviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1665249}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 88d1028cdc6fe70c9acab900348c841bd903a95e\n" }, { "commit": "c6357ae6b83d73d08141cd815eeccc1a0f75b3f9", "tree": "85ea7e79b29ae008d7c340d2abde124da7cbaaa3", "parents": [ "1d0d5e3ccbe51698f931191fdc4abb2993cd6b86" ], "author": { "name": "Eriko Kurimoto", "email": "elkurin@chromium.org", "time": "Tue Jul 21 02:44:53 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 02:48:24 2026" }, "message": "Revert \"media: Migrate EncryptionScheme to Mojo traits\"\n\nThis reverts commit ddf49493c7141b0dc62d6f79af506924dccb5b0e.\n\nReason for revert: Culprit of the failure of\nvirtual/gpu/fast/canvas/canvas-drawImage-live-video.html\n\nFailure Link: https://ci.chromium.org/ui/p/chromium/builders/ci/linux-oi-rel/30727/overview\n\nOriginal change\u0027s description:\n\u003e media: Migrate EncryptionScheme to Mojo traits\n\u003e\n\u003e This CL migrates EncryptionScheme to Mojo EnumTraits.\n\u003e\n\u003e Bug: 393179188\n\u003e Change-Id: I3f5a78fc501af796df8fae2de9b2b7e885e5e6bf\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100522\n\u003e Reviewed-by: Daniel Cheng \u003cdcheng@chromium.org\u003e\n\u003e Reviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\n\u003e Commit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1664984}\n\nBug: 393179188\nChange-Id: Id1fcf15f70280d51a5565827bafbd62f04bfce7b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8125455\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nCommit-Queue: Eriko Kurimoto \u003celkurin@chromium.org\u003e\nAuto-Submit: Eriko Kurimoto \u003celkurin@chromium.org\u003e\nOwners-Override: Eriko Kurimoto \u003celkurin@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665143}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 4dbd10cd964233e2cc97ee7437f2d9ab43194a10\n" }, { "commit": "1d0d5e3ccbe51698f931191fdc4abb2993cd6b86", "tree": "0e7b4ac0c7b1a696be06402003b461d6cf81c918", "parents": [ "b2744d34fa7dcaf5855a361f41907a9b524ec8dc" ], "author": { "name": "Sangbaek Park", "email": "sangbaekpark@google.com", "time": "Tue Jul 21 02:00:21 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 02:04:30 2026" }, "message": "media: Always check Dolby Vision capability with HDR check\n\nPreviously in crrev.com/c/7998779, we introduced the feature flag\nkHardwareSecureDecryptionDolbyVisionWithHdrCheck to conditionally check\nDolby Vision codec capability with HDR check.\n\nSince we are ready to enable this check by default, this CL removes the\nfeature flag entirely and always enables the HDR display check when\nvalidating Dolby Vision capability for hardware-secure OS CDM playback.\n\nThe query sequence still checks Dolby Vision support without HDR first,\nand only queries with HDR if the former is supported (acting as a fast-\nfail optimization). Other video codecs continue to be queried exactly\nonce.\n\nWe temporarily keep reporting both histograms (WithHdrCheck and\nWithoutHdrCheck) for future validation to verify that the Dolby Vision\ncapability query with the HDR display check always works as expected.\nOnce validated, the WithoutHdrCheck histogram will be removed\n(crbug.com/536954447).\n\nBug: b:519288654, b:407123708\nChange-Id: I2fe8146c2e3d76c592ee77a8bd1bf2e306a3b2d2\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8102713\nReviewed-by: Xiaohan Wang \u003cxhwang@chromium.org\u003e\nCommit-Queue: Sangbaek Park \u003csangbaekpark@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665118}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 97b701776baa40af1976c59bd823ba462602926e\n" }, { "commit": "b2744d34fa7dcaf5855a361f41907a9b524ec8dc", "tree": "ccdf78dcfae74ca97b91b2abd38940666f88b024", "parents": [ "e8d69c6f3e5367e4fb4eaa7346b634e808ac42a6" ], "author": { "name": "Eugene Zemtsov", "email": "eugene@chromium.org", "time": "Tue Jul 21 00:49:41 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 00:52:37 2026" }, "message": "Reland \"media: Fix keyframe interval and prevent division by zero in NDK VEA\"\n\nThis reverts commit 399d356f3fa2601ba9e59d6dd1f4ba4b31d8f081.\n\nReason for revert: Removed problematic test\n\nOriginal change\u0027s description:\n\u003e Revert \"media: Fix keyframe interval and prevent division by zero in NDK VEA\"\n\u003e\n\u003e This reverts commit 5b43b81cee294ca41d30b8631278b757ba850797.\n\u003e\n\u003e Reason for revert: Suspected cause of failures on android-10-x86-rel\n\u003e\n\u003e Failure Link: https://ci.chromium.org/ui/p/chromium/builders/ci/android-10-x86-rel/8906/overview\n\u003e\n\u003e Original change\u0027s description:\n\u003e \u003e media: Fix keyframe interval and prevent division by zero in NDK VEA\n\u003e \u003e\n\u003e \u003e Convert GOP length (in frames) to seconds before setting\n\u003e \u003e AMEDIAFORMAT_KEY_I_FRAME_INTERVAL as a float, fixing a bug where it\n\u003e \u003e was set directly in frames.\n\u003e \u003e\n\u003e \u003e Also reject 0 framerate in Initialize() and RequestEncodingParametersChange(),\n\u003e \u003e and defensively guard the division in RecordFrameTimestamps() to prevent\n\u003e \u003e crashes.\n\u003e \u003e\n\u003e \u003e Added VerifyGOPLength integration test to verify the fix.\n\u003e \u003e\n\u003e \u003e Change-Id: I754be5b585cf70bb8191b2d2b3e1805aaa50279c\n\u003e \u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100695\n\u003e \u003e Commit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\n\u003e \u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e \u003e Cr-Commit-Position: refs/heads/main@{#1663577}\n\u003e\n\u003e No-Presubmit: true\n\u003e No-Tree-Checks: true\n\u003e No-Try: true\n\u003e Change-Id: I9079bdaef8354d4e79033989d5d8c5871233008a\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8115711\n\u003e Owners-Override: Alex Mitra \u003calexmitra@chromium.org\u003e\n\u003e Commit-Queue: Alex Mitra \u003calexmitra@chromium.org\u003e\n\u003e Bot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1663770}\n\nChange-Id: I520c7a0acb0527b84ea7254a63e6fb7e60715f26\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127217\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665083}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: e4ffb12b9c7c5f5699d77579cc709ab0dc4ea11f\n" }, { "commit": "e8d69c6f3e5367e4fb4eaa7346b634e808ac42a6", "tree": "afc3aa69c6474038e7eb0924270dfa9af2de6ccf", "parents": [ "38d8b49ddeb24d83c465213c3b65fb3fa3b365f0" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Tue Jul 21 00:18:39 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Tue Jul 21 00:22:15 2026" }, "message": "Bounds check av1 decoder\n\nWe have a limit of 32k pixels for any given dimension, but av1 allows\n64k. We should fail to decode if it\u0027s too big.\n\nFixed: 536428842\nChange-Id: Iedefeef7fac680674820553ceab52bf5f31ff435\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8127838\nReviewed-by: Syed AbuTalib \u003clowkey@google.com\u003e\nAuto-Submit: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Syed AbuTalib \u003clowkey@google.com\u003e\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665064}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 8f06b14492a6591362290aa978f5f61484b26464\n" }, { "commit": "38d8b49ddeb24d83c465213c3b65fb3fa3b365f0", "tree": "ecf6520ce2ce31a484c36717d7bbda3e397dabd4", "parents": [ "cdfdd39bd9d86cf1d885af819255c86070f0950c" ], "author": { "name": "Ted Meyer", "email": "tmathmeyer@chromium.org", "time": "Mon Jul 20 23:29:48 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 20 23:33:38 2026" }, "message": "[HLS] Let ManifestDemuxerStreams handle switching\n\nManifestDemuxer uses a wrapper DemuxerStream to monitor the underlying\nmedia timestamps from the ChunkDemuxer\u0027s DemuxerStreams. When switching\ntracks, instead of creating a new ManifestDemuxerStream, we reuse the\nsame track and just start appending new data segments into the\nChunkDemuxer. As a result, RendererImpl should not restart or\nre-initialize the pipeline.\n\nThis is a (hopefully) temporary stop-gap to make track switching more\nperformant while I figure out a good way to modify ChunkDemuxer to\nhandle multiple roles being added at different points during the\ndemuxer\u0027s life time and may or may not have data appended.\n\nBug: 535432430\nChange-Id: I576db2ff681b5a0e1facd9eecc5b2b4d2dca4462\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8111397\nCommit-Queue: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1665020}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 563012806bc40f970d3f45cd74b117cf5d31e0f4\n" }, { "commit": "cdfdd39bd9d86cf1d885af819255c86070f0950c", "tree": "ce38b463001b3c952d7d3e83c5b29b51f9a321f9", "parents": [ "19862185ba01a42e0753cbbecb4110f17ffd43ab" ], "author": { "name": "Vikram Pasupathy", "email": "vpasupathy@chromium.org", "time": "Mon Jul 20 22:36:52 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 20 22:41:06 2026" }, "message": "media: Migrate EncryptionScheme to Mojo traits\n\nThis CL migrates EncryptionScheme to Mojo EnumTraits.\n\nBug: 393179188\nChange-Id: I3f5a78fc501af796df8fae2de9b2b7e885e5e6bf\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100522\nReviewed-by: Daniel Cheng \u003cdcheng@chromium.org\u003e\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Vikram Pasupathy \u003cvpasupathy@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1664984}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ddf49493c7141b0dc62d6f79af506924dccb5b0e\n" }, { "commit": "19862185ba01a42e0753cbbecb4110f17ffd43ab", "tree": "e61c7919b8a92395e75926c25b5655ab7b3329d0", "parents": [ "0ba0a7a7d040e9d10baebbeb5e5c24d39ec7ebba" ], "author": { "name": "Saifuddin Hitawala", "email": "hitawala@chromium.org", "time": "Mon Jul 20 16:59:00 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 20 17:02:54 2026" }, "message": "[media] Add default output color space for V4L2VideoEncodeAccelerator\n\nWe added some default color space in crrev.com/c/8074041 but left\nthem as invalid for V4L2VEA. This change updates it so that its\noutput frames also have a valid default color space based on output\nformat config. This is because of some tast test failures seen on\ncrbug.com//535113130 where we see input/output frame mismatch here [1].\n\nAlso update some tests to also have default color spaces when creating\nmappable SI VideoFrame.\n\n[1] https://source.chromium.org/chromium/chromium/src/+/main:media/gpu/chromeos/libyuv_image_processor_backend.cc;drc\u003dfba75fddc477ba753324b98a9ed1435f430d0679;l\u003d352\n\nBug: 535113130, 425634684\nChange-Id: I1f2401404f3731818db3ec7c6d2906c279faf108\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8120090\nCommit-Queue: Saifuddin Hitawala \u003chitawala@chromium.org\u003e\nReviewed-by: Hirokazu Honda \u003chiroh@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1664763}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5dc3daa6880baf37e367c2c38371e9c7e7e1e254\n" }, { "commit": "0ba0a7a7d040e9d10baebbeb5e5c24d39ec7ebba", "tree": "04091d2bf012a8381822269225a97d73caa4e25f", "parents": [ "92dce5d7b6b7f568b7ad1400fd67d216cc06d3b8" ], "author": { "name": "Henrik Andreasson", "email": "henrika@chromium.org", "time": "Mon Jul 20 16:55:16 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Mon Jul 20 16:58:40 2026" }, "message": "Reland \"Audio: Add UMA metric for WASAPI input initialization retry outcome\"\n\nThis reverts commit 4a50c5a69b0418307b47b63b88497659f5d275c7.\n\nReason for revert: I will ensure that presubmit works this time.\n\nOriginal change\u0027s description:\n\u003e Revert \"Audio: Add UMA metric for WASAPI input initialization retry outcome\"\n\u003e\n\u003e This reverts commit 48816f9d1d7efe611379244dee0de8e72ada3ad6.\n\u003e\n\u003e Reason for revert: causing win-presubmit bot to fail\n\u003e\n\u003e Failure Link: https://ci.chromium.org/ui/p/chromium/builders/ci/win-presubmit/21146/overview\n\u003e\n\u003e Original change\u0027s description:\n\u003e \u003e Audio: Add UMA metric for WASAPI input initialization retry outcome\n\u003e \u003e\n\u003e \u003e Logs Media.Audio.Capture.Win.InitializeSucceeded for all initialization\n\u003e \u003e attempts to measure the size of the change in an A/B comparison.\n\u003e \u003e\n\u003e \u003e Also logs Media.Audio.Capture.Win.InitDeviceInUseRetryOutcome when the\n\u003e \u003e first attempt to initialize WASAPI input stream fails with\n\u003e \u003e AUDCLNT_E_DEVICE_IN_USE to measure retry effectiveness.\n\u003e \u003e\n\u003e \u003e Finch experiment evaluation guide:\n\u003e \u003e\n\u003e \u003e 1. Overall Impact: Compare the success rate of Media.Audio.Capture.Win.InitializeSucceeded between\n\u003e \u003e Control (disabled) and Treatment (enabled) groups.\n\u003e \u003e 2. Mitigation Success Rate: For the Treatment group, compute:\n\u003e \u003e (kSucceededOnFirstRetry + kSucceededOnSecondRetry) / Total Samples from the Media.Audio.Capture.Win.InitDeviceInUseRetryOutcome histogram.\n\u003e \u003e 3. Optimization: Review the ratio of kSucceededOnFirstRetry vs.\n\u003e \u003e kSucceededOnSecondRetry to optimize the retry count or delay\n\u003e \u003e\n\u003e \u003e Bug: 532467443\n\u003e \u003e Change-Id: I389b63d5575817900c019f8cb0c74ba43f0c55c0\n\u003e \u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8064281\n\u003e \u003e Reviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\n\u003e \u003e Reviewed-by: Takashi Toyoshima \u003ctoyoshim@chromium.org\u003e\n\u003e \u003e Commit-Queue: Henrik Andreasson \u003chenrika@chromium.org\u003e\n\u003e \u003e Cr-Commit-Position: refs/heads/main@{#1661152}\n\u003e\n\u003e Bug: 534375682\n\u003e No-Presubmit: true\n\u003e No-Tree-Checks: true\n\u003e No-Try: true\n\u003e Bug: 532467443\n\u003e Change-Id: Id0e6ec394c84ff60afd520596f6e59dcb2f5ec9d\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8085508\n\u003e Auto-Submit: Nicolás Peña \u003cnpm@chromium.org\u003e\n\u003e Bot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\n\u003e Owners-Override: Nicolás Peña \u003cnpm@chromium.org\u003e\n\u003e Commit-Queue: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1661258}\n\nBug: 534375682\nBug: 532467443\nChange-Id: I4967aae16778e81a4e7d5b9021f754572d5dc9b0\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8117055\nReviewed-by: Fredrik Hernqvist \u003cfhernqvist@google.com\u003e\nCommit-Queue: Henrik Andreasson \u003chenrika@chromium.org\u003e\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nReviewed-by: Luc Nguyen \u003clucnguyen@google.com\u003e\nCr-Commit-Position: refs/heads/main@{#1664757}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 1fcef5c5ca056b62f671c5cf90ee41af128109e6\n" }, { "commit": "92dce5d7b6b7f568b7ad1400fd67d216cc06d3b8", "tree": "09d82de4a830076fd2c397aadb285a69511b7b05", "parents": [ "e62023036e4ac751289533389a5084bf2083bd1d" ], "author": { "name": "Feras Aldahlawi", "email": "frs@chromium.org", "time": "Fri Jul 17 23:44:16 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 23:47:43 2026" }, "message": "cdm: Modernized media/cdm to use std::make_unique.\n\nChange-Id: I2b59aeafb3853ba11e7c64da54f3306b6a6a6964\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8116390\nCommit-Queue: Feras Aldahlawi \u003cfrs@chromium.org\u003e\nReviewed-by: Sangbaek Park \u003csangbaekpark@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1664245}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 9eed4322430ffa7da8d77768d117e2dad2f8da81\n" }, { "commit": "e62023036e4ac751289533389a5084bf2083bd1d", "tree": "fa06a31d0b07927bcb528ee87efdb8bd02ae5e24", "parents": [ "772d5a0a35f833a75daf6db7552e35cb17c7dab8" ], "author": { "name": "Philipp Hancke", "email": "philipp.hancke@googlemail.com", "time": "Fri Jul 17 11:20:58 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 11:26:27 2026" }, "message": "getusermedia: disable MediaFoundation for Microsoft LifeCam Studio\n\non Windows since this causes issues with popular apps like\nGoogle Meet or Jitsi Meet when they request 720p - 360p works.\n\nSee\n https://devicehunt.com/view/type/usb/vendor/045E/device/0772\nfor the USB id\n\nBug: chromium:535635412\nChange-Id: I1904c8f2e69844eb16d0015de25ffe52c08e8b4b\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8102205\nReviewed-by: Guido Urdaneta \u003cguidou@chromium.org\u003e\nCommit-Queue: Philipp Hancke \u003cphilipp.hancke@googlemail.com\u003e\nCr-Commit-Position: refs/heads/main@{#1663803}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 8ee51aeab33e84d4b8fe1478a99995763fb713b1\n" }, { "commit": "772d5a0a35f833a75daf6db7552e35cb17c7dab8", "tree": "fd5dff3b7e426933ee6d9ff808b54ea79bd71afd", "parents": [ "11c6c0958d1150a72a4f0103f7ab0cb02c3319a2" ], "author": { "name": "Alex Mitra", "email": "alexmitra@chromium.org", "time": "Fri Jul 17 10:10:37 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 10:15:32 2026" }, "message": "Revert \"media: Fix keyframe interval and prevent division by zero in NDK VEA\"\n\nThis reverts commit 5b43b81cee294ca41d30b8631278b757ba850797.\n\nReason for revert: Suspected cause of failures on android-10-x86-rel\n\nFailure Link: https://ci.chromium.org/ui/p/chromium/builders/ci/android-10-x86-rel/8906/overview\n\nOriginal change\u0027s description:\n\u003e media: Fix keyframe interval and prevent division by zero in NDK VEA\n\u003e\n\u003e Convert GOP length (in frames) to seconds before setting\n\u003e AMEDIAFORMAT_KEY_I_FRAME_INTERVAL as a float, fixing a bug where it\n\u003e was set directly in frames.\n\u003e\n\u003e Also reject 0 framerate in Initialize() and RequestEncodingParametersChange(),\n\u003e and defensively guard the division in RecordFrameTimestamps() to prevent\n\u003e crashes.\n\u003e\n\u003e Added VerifyGOPLength integration test to verify the fix.\n\u003e\n\u003e Change-Id: I754be5b585cf70bb8191b2d2b3e1805aaa50279c\n\u003e Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100695\n\u003e Commit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\n\u003e Reviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\n\u003e Cr-Commit-Position: refs/heads/main@{#1663577}\n\nNo-Presubmit: true\nNo-Tree-Checks: true\nNo-Try: true\nChange-Id: I9079bdaef8354d4e79033989d5d8c5871233008a\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8115711\nOwners-Override: Alex Mitra \u003calexmitra@chromium.org\u003e\nCommit-Queue: Alex Mitra \u003calexmitra@chromium.org\u003e\nBot-Commit: rubber-stamper@appspot.gserviceaccount.com \u003crubber-stamper@appspot.gserviceaccount.com\u003e\nCr-Commit-Position: refs/heads/main@{#1663770}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 399d356f3fa2601ba9e59d6dd1f4ba4b31d8f081\n" }, { "commit": "11c6c0958d1150a72a4f0103f7ab0cb02c3319a2", "tree": "1f4ab95e6b811191ca1c05bb57dd9c8ce752d1c8", "parents": [ "d3346796ae7dd67e1767a4f93cda5cae7b063c07" ], "author": { "name": "Sam Zackrisson", "email": "saza@chromium.org", "time": "Fri Jul 17 08:34:12 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 08:39:40 2026" }, "message": "Audio service: share ML model ownership with clients\n\nThe ML models are kept in FlatBufferModels, which are immutable after construction and have no exterior dependencies and can therefore be safely read and deleted on arbitrary threads.\n\nWrapping our models in a //media refcounted interface has 3 benefits:\n1. The models can be read on background threads to avoid blocking main thread work that are synchronous to renderer-side JS API calls (used here: http://crrev.com/c/8108258),\n2. the ML model manager implementation is greatly simplified, and\n3. the WebRTC components in //media can access and own the models they use, instead of forcing higher layers (AudioProcessorHandler, VoiceIsolationHandler) to cache them.\n\nBug: webrtc:442444736\nChange-Id: Ic16c58d871039f9949a9544d2899f16930c7af46\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8091889\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Sam Zackrisson \u003csaza@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663738}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 3acad265244c446faa34b90f131dfd155557b79e\n" }, { "commit": "d3346796ae7dd67e1767a4f93cda5cae7b063c07", "tree": "28f86d22a4a0f7a94ae39e5e37c8b8e765826483", "parents": [ "1368ccf9f9ec13064695ed6421771c08d903e51d" ], "author": { "name": "Pablo Barrera", "email": "barrerap@google.com", "time": "Fri Jul 17 07:24:57 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 07:30:12 2026" }, "message": "Include a resampler for VoiceIsolation\n\nInternally VoiceIsolation works at a fixed sample rate and frame size.\nThis change adds a couple of `ConvertingAudioFifo` to adjust the\nincoming signal to the expected cadence and sampling frequency. This\nintroduces an algorithm delay of approximately 80ms. Likely this is not\na problem once we introduce a band splitting system later in\ndevelopment.\n\nWe considered using AudioConverter as an alternative, but this is a pull\nbased system instead of pull base. With it we can reduce latency to\naround 50ms but we thought it was not worth it due to code complexity.\nIt\u0027s also possible to use a custom implementation for both the resampler\nand the buffering steps, lowering latency to 23ms, the minimum algorithm\ndelay. Again, the complexity did not justify the latency reduction that\nwe will remove in the future using other processing components.\n\nBug: 512016773\nChange-Id: I4481f37f69556d2f55a695a5fbc1fec01042085d\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8052654\nReviewed-by: Sam Zackrisson \u003csaza@chromium.org\u003e\nCommit-Queue: Pablo Barrera González \u003cbarrerap@google.com\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663710}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: ae50a28e24edf0842b5fe17b59724477eba7b2b1\n" }, { "commit": "1368ccf9f9ec13064695ed6421771c08d903e51d", "tree": "bebde6ad94bfdaf2f097bb798f3f82ce55991204", "parents": [ "bb2c0dd75e5e9c673588bb8605674149478a81e7" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Fri Jul 17 05:55:14 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 06:01:43 2026" }, "message": "Add WPT test coverage for 16-bit high-bit-depth video frame encoding\n\nAdds `video-encoder-hbd.https.any.js` WebPlatformTest to test encoding\n16-bit high-bit-depth (HBD) and alpha VideoFrame inputs directly from\nUint16Array buffers without going through Canvas.\n\nTest coverage details:\n- Tests 9 high-bit-depth input formats: I420P10, I422P10, I444P10,\n I420P12, I422P12, I444P12, I420AP10, I422AP10, I444AP10.\n- Exercises 8-bit conversion encoding for 8-bit profiles (AV1 Main 8-bit,\n VP8, VP9 Profile 0, VP9 Profile 1, H.264 Baseline) and 10-bit HBD\n encoding for high-bit-depth profiles (VP9 Profile 2, VP9 Profile 3).\n- Re-decodes output chunks via VideoDecoder and verifies RGBA channels\n (including alpha) using VideoFrame.copyTo({ format: \u0027RGBA\u0027 }).\n- Includes capability checks via VideoEncoder and VideoDecoder\n isConfigSupported() with assert_implements_optional().\n\nFixed: 529852980\nChange-Id: Ib42c3c97a04c0733ecbb4d5d3da0b94e6b9c082a\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8093249\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663682}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 9263a4eb79e4b8889c815086444ffff6e25674b7\n" }, { "commit": "bb2c0dd75e5e9c673588bb8605674149478a81e7", "tree": "16a822d70a8235d520e4f83da011d7b8badc2892", "parents": [ "3be9d47b1e98cf0e7329119bff3864509ed83551" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Fri Jul 17 05:54:56 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 06:00:30 2026" }, "message": "Add high-bit-depth YUV format support to software video encoders\n\nAllows OpenH264, VPX, and AV1 software video encoders to receive\nhigh-bit-depth (10-bit and 12-bit), 4:2:2, 4:4:4, and alpha YUV input\nformats without rejecting them. When an input format is not natively\nsupported by the codec configuration, frames are automatically converted\nand scaled using VideoFrameConverter.\n\nAlso updates software video encoder tests to include high-bit-depth\ninput format parameters and refactors CountDifferentPixels() to support\n16-bit sample formats and inverse chroma subsampling scaling.\n\nBug: 529852980\nChange-Id: I55bc0664c9354a312701bbd43d9a3bed47f855c4\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8093014\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663681}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 2d939a5f9f493fd4bd56bc7f9cd7b6fd39359a75\n" }, { "commit": "3be9d47b1e98cf0e7329119bff3864509ed83551", "tree": "ff2f5a045eb742eb19117b9b1f1e1d17a7b0fc0f", "parents": [ "827f5c33ebe2a94b090ff41b34fe56e519bb9b41" ], "author": { "name": "Jordan Bayles", "email": "jophba@chromium.org", "time": "Fri Jul 17 02:09:27 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Fri Jul 17 02:14:34 2026" }, "message": "Fix cross-thread WeakPtr UAF in FrameInjectingDemuxer\n\nFrameInjectingDemuxer methods execute on the media sequence and bind its\nWeakPtr instances to that sequence. However, when the media player is\ntorn down, DemuxerManager destructs FrameInjectingDemuxer inline on the\nmain thread. Because base::WeakPtr invalidation and dereferencing must\noccur on the same sequence, invalidating on main while callbacks execute\non the media sequence causes a Use-After-Free (UAF) or CHECK failure.\n\nThis change updates DemuxerManager::~DemuxerManager() to destroy\nkFrameInjectingDemuxer on the media sequence via DeleteSoon, matching\nkManifestDemuxer behavior. It also adds\nweak_factory_.InvalidateWeakPtrs() in FrameInjectingDemuxer::Stop() to\nensure pending callbacks are dropped on pipeline stop, and adds unit\ntests covering FrameInjectingDemuxer.\n\nBug: 533186443\nChange-Id: I14a0d7d372ea68dc08ac7d6fc63385d88a328378\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8091971\nReviewed-by: Ted (Chromium) Meyer \u003ctmathmeyer@chromium.org\u003e\nCommit-Queue: Jordan Bayles \u003cjophba@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663632}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: a5fbd365227804d0887fd9e25253672d4bc21c52\n" }, { "commit": "827f5c33ebe2a94b090ff41b34fe56e519bb9b41", "tree": "fb84097e5e4f76b24bb79f0527d4878b2fce2548", "parents": [ "719151ce0179d477f141fb11dfeefc2e6bac1b88" ], "author": { "name": "Eugene Zemtsov", "email": "eugene@chromium.org", "time": "Thu Jul 16 23:41:01 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 16 23:48:46 2026" }, "message": "media: Fix keyframe interval and prevent division by zero in NDK VEA\n\nConvert GOP length (in frames) to seconds before setting\nAMEDIAFORMAT_KEY_I_FRAME_INTERVAL as a float, fixing a bug where it\nwas set directly in frames.\n\nAlso reject 0 framerate in Initialize() and RequestEncodingParametersChange(),\nand defensively guard the division in RecordFrameTimestamps() to prevent\ncrashes.\n\nAdded VerifyGOPLength integration test to verify the fix.\n\nChange-Id: I754be5b585cf70bb8191b2d2b3e1805aaa50279c\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8100695\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663577}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 5b43b81cee294ca41d30b8631278b757ba850797\n" }, { "commit": "719151ce0179d477f141fb11dfeefc2e6bac1b88", "tree": "c024d9f132429f708bc3ecbce4360c54f6dd1da0", "parents": [ "83f5da6fea2fe3e3bcfa3b0cb302c683d2de4e0e" ], "author": { "name": "Eugene Zemtsov", "email": "eugene@chromium.org", "time": "Thu Jul 16 21:17:41 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 16 21:23:03 2026" }, "message": "Roll ffmpeg for M152\n\nFix tests and glue code to account for newly introduced in ffmpeg AAC\ninitial_padding\n\nBug: 527501449, 504356442, 403330011\nChange-Id: I3471e8e4621c037958316ebc09cd38dd59ce110f\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8066846\nReviewed-by: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663509}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: 493d2dfa61ffb1e9cf218c24e356f31e1b76b66a\n" }, { "commit": "83f5da6fea2fe3e3bcfa3b0cb302c683d2de4e0e", "tree": "e84f8b20a38815006f3375d05326566334c8eb50", "parents": [ "0a57e5a85c83c0a67dcec44137626ef6056c2603" ], "author": { "name": "Dale Curtis", "email": "dalecurtis@chromium.org", "time": "Thu Jul 16 18:35:19 2026" }, "committer": { "name": "Copybara-Service", "email": "copybara-worker@google.com", "time": "Thu Jul 16 18:40:47 2026" }, "message": "[media] Add high-bit-depth YUV input support to VideoFrameConverter\n\nAdd support for 10-bit and 12-bit YUV input formats (YUV420P10/12,\nYUV422P10/12, YUV444P10/12, YUV420AP10, YUV422AP10, and YUV444AP10)\nto VideoFrameConverter.\n\nSpatial scaling and subsampling conversions are handled in 16-bit space\nvia ScalePlane_16 in I4xxxScale_16, followed by plane-by-plane 16-to-8\nbit down-conversion using Convert16To8Plane.\n\nAlso adds default plane layout definitions for YUV planar alpha formats\n(I420A, I422A, I444A, YUV420AP10, YUV422AP10, YUV444AP10) in VideoFrame.\n\nBug: 529852980\nChange-Id: I5955dbecb201093e47f7aa4d91cdfef2d317d339\nReviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8089421\nCommit-Queue: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCommit-Queue: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nReviewed-by: Eugene Zemtsov \u003ceugene@chromium.org\u003e\nAuto-Submit: Dale Curtis \u003cdalecurtis@chromium.org\u003e\nCr-Commit-Position: refs/heads/main@{#1663340}\nNOKEYCHECK\u003dTrue\nGitOrigin-RevId: c8052ed47f85f104a01396c9dec2285b056f3dc9\n" } ], "next": "0a57e5a85c83c0a67dcec44137626ef6056c2603" }