1. 10 Sep, 2020 35 commits
  2. 09 Sep, 2020 5 commits
    • Gang Wu's avatar
      use app:chromeDrawableTint to tint TextViewWithCompoundDrawables · aacbc378
      Gang Wu authored
      We should use app:chromeDrawableTint to tint
      TextViewWithCompoundDrawables instead of android:drawableTint.
      
      android:drawableTint works on API level 23 and above. But somehow
      not work for PNG drawables in API 23, and TextViewWithCompoundDrawables
      provide app:chromeDrawableTint to tint, so we can use it to handle the
      issue.
      
      Bug: 1126035
      Change-Id: Iebc9b0d0ccf9ca8510bd03d8285d19b2ba705cec
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2402102Reviewed-by: default avatarTheresa  <twellington@chromium.org>
      Commit-Queue: Gang Wu <gangwu@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#805559}
      aacbc378
    • Kent Tamura's avatar
      Enable "gn check" for //third_party/blink/renderer/modules/webcodecs · 89a6b640
      Kent Tamura authored
      webcodecs:unit_tests had a circular dependency with
      controller:blink_unittests_sources. This CL moves out
      controller:blink_bindings_test_sources from
      controller:blink_unittests_sources and adds
      controller:blink_bindings_test_sources dependency to
      webcodecs:unit_tests to remove the circular dependency.
      
      This CL has no behavior changes.
      
      Bug: 800764
      Change-Id: Ie578dfe442f2ee111f19abb7496e41a27ff66b80
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2400644
      Commit-Queue: Takuto Ikuta <tikuta@chromium.org>
      Commit-Queue: Thomas Guilbert <tguilbert@chromium.org>
      Auto-Submit: Kent Tamura <tkent@chromium.org>
      Reviewed-by: default avatarTakuto Ikuta <tikuta@chromium.org>
      Reviewed-by: default avatarThomas Guilbert <tguilbert@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#805558}
      89a6b640
    • Miguel Casas's avatar
      media/gpu/vaapi: batch submitting VABufferIDs for MJPEG decoding · 405ed4ac
      Miguel Casas authored
      This CL follows crrev.com/c/2393629 by batching up SubmitBuffer
      calls in the JPEG decode accelerator, used e.g. for MJPEG camera
      stream decoding.
      
      This was verified via chrome:tracing while capturing HD on kohaku
      and reks (Braswell) in the link [1]. Similarly to what is seen in
      the previous CL, the combination of SubmitBuffer() calls on ToT
      is ~10% slower than the batched SubmitBuffers(). A key difference
      is that since JPEG is way simpler than e.g. VP9, the submitting
      time is comparable to the ExecuteBuffers_Locked(), so this CL
      has some minor overall decode-time savings. But still the target
      here is to reduce contention on the lock and libva for video
      conferencing cases with multiple remote participants a la Meet.
      
      [1] webrtc.github.io/samples/src/content/getusermedia/resolution/
      
      Cq-Depend: chromium:2393629
      Bug: b/166646505
      Change-Id: I183cc151de24d8fdbdc62e0172be296dacbf26f2
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2393525
      Commit-Queue: Miguel Casas <mcasas@chromium.org>
      Reviewed-by: default avatarJ Kardatzke <jkardatzke@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#805557}
      405ed4ac
    • Miguel Casas's avatar
      media/gpu/vaapi: batch submitting VABufferIDs for VP8 decoding · b512bba7
      Miguel Casas authored
      This CL follows crrev.com/c/2393629 by batching up SubmitBuffer
      calls in the VP8 decode accelerator.
      
      This is verified via chrome:tracing and codepen.io/full/qzYXba that
      plays 4 1280x572 VP8 videos at the same time. Tracing is captured for
      a few seconds, basically the SubmitBuffers() call takes ~0.312ms/call
      versus the individual SubmitBuffer() calls ~0.072ms -- which gives
      an aggregate of 0.360ms, or savings of about ~20%. Since decoding
      itself takes about the same as SubmitBuffers on my BSW device (reks),
      batching gives ~10% reduction in decode time.
      
      Bug: b/166646505
      Change-Id: I11c5570bea1c049fe0e9af75239bd60ac906a843
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2401512
      Commit-Queue: Miguel Casas <mcasas@chromium.org>
      Reviewed-by: default avatarJ Kardatzke <jkardatzke@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#805556}
      b512bba7
    • David Tseng's avatar
      Add Learn Mode descriptions for some gestures · 3cddc841
      David Tseng authored
      See go/chromevox-touch_gestures
      
      Some gestures, e.g. swipe two fingers left, do not have an explicit command. Others, like three finger swipe up/down, have commands which are not also keyboard commands.
      
      In Learn Mode, these gestures are not fullyd escribed beyond just the physical gesture description.
      
      Add these descriptions now in this change.
      
      R=akihiroota@chromium.org
      
      Bug: 1124454
      Test: browser_tests --gtest_filter=ChromeVoxLearnMode*.Gesture*
      AX-Relnotes: ChromeVox Learn Mode now describes all new gestures and what they do.
      Change-Id: I19fd3b39aa15f9df0daa864c4f99f69173d4baa1
      Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2386377
      Commit-Queue: David Tseng <dtseng@chromium.org>
      Reviewed-by: default avatarAkihiro Ota <akihiroota@chromium.org>
      Cr-Commit-Position: refs/heads/master@{#805555}
      3cddc841