1921863 - Update libvpx to new version 32de9c2becdd5d96d84d37be9f7fb9de43f24a4d from 2024-10-02 00:00:00

Status: RESOLVED FIXED

(Core :: Audio/Video, enhancement)

Description

[Update Bot]

This update covers 13 commits. Here are the overall diff statistics, and then the commit information.

---

media/libvpx/config/generic/vp9_rtcd.h | 15 +
media/libvpx/config/generic/vpx_config.c | 2 +-
media/libvpx/config/linux/arm/vp9_rtcd.h | 3 +
media/libvpx/config/linux/arm/vpx_config.c | 2 +-
media/libvpx/config/linux/arm64/vp9_rtcd.h | 3 +
media/libvpx/config/linux/arm64/vpx_config.c | 2 +-
media/libvpx/config/linux/ia32/vp9_rtcd.h | 27 +
media/libvpx/config/linux/ia32/vpx_config.c | 2 +-
media/libvpx/config/linux/x64/vp9_rtcd.h | 27 +
media/libvpx/config/linux/x64/vpx_config.c | 2 +-
media/libvpx/config/mac/arm64/vp9_rtcd.h | 15 +
media/libvpx/config/mac/arm64/vpx_config.c | 2 +-
media/libvpx/config/mac/ia32/vp9_rtcd.h | 27 +
media/libvpx/config/mac/ia32/vpx_config.c | 2 +-
media/libvpx/config/mac/x64/vp9_rtcd.h | 27 +
media/libvpx/config/mac/x64/vpx_config.c | 2 +-
media/libvpx/config/win/aarch64/vp9_rtcd.h | 3 +
media/libvpx/config/win/aarch64/vpx_config.c | 2 +-
media/libvpx/config/win/ia32/vp9_rtcd.h | 27 +
media/libvpx/config/win/ia32/vpx_config.c | 2 +-
media/libvpx/config/win/x64/vp9_rtcd.h | 27 +
media/libvpx/config/win/x64/vpx_config.c | 2 +-
media/libvpx/libvpx/test/convolve_test.cc | 368 ++++++++-
media/libvpx/libvpx/vp8/encoder/onyx_int.h | 4 +-
media/libvpx/libvpx/vp8/vp8_cx_iface.c | 17 +-
media/libvpx/libvpx/vp9/common/vp9_rtcd_defs.pl | 15 +
media/libvpx/libvpx/vp9/encoder/vp9_encoder.c | 2 +-
media/libvpx/libvpx/vp9/encoder/vp9_speed_features.c | 4 +
media/libvpx/libvpx/vp9/encoder/vp9_svc_layercontext.c | 4 +-
media/libvpx/libvpx/vp9/encoder/vp9_temporal_filter.c | 95 +-
media/libvpx/libvpx/vp9/encoder/vp9_temporal_filter.h | 35 +-
media/libvpx/libvpx/vp9/encoder/vp9_tpl_model.c | 11 +-
media/libvpx/libvpx/vp9/encoder/x86/temporal_filter_avx2.c | 496 +++++++++++
media/libvpx/libvpx/vp9/vp9_cx_iface.c | 19 +-
media/libvpx/libvpx/vp9/vp9cx.mk | 2 +
media/libvpx/libvpx/vpx/vpx_tpl.h | 4 +-
media/libvpx/libvpx/vpx_dsp/arm/highbd_vpx_convolve8_sve2.c | 32 +-
media/libvpx/libvpx/vpx_dsp/arm/vpx_convolve8_neon_dotprod.c | 20 +-
media/libvpx/libvpx/vpx_dsp/arm/vpx_convolve8_neon_i8mm.c | 20 +-
media/libvpx/libvpx/vpx_ports/aarch64_cpudetect.c | 34 +-
media/libvpx/moz.yaml | 4 +-
media/libvpx/sources.mozbuild | 6 +
42 files changed, 1270 insertions(+), 145 deletions(-)

---

32de9c2becdd5d96d84d37be9f7fb9de43f24a4d by Bohan <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/32de9c2becdd5d96d84d37be9f7fb9de43f24a4d
Authored: 2024-09-25 09:27:36 -0700
Committed: 2024-09-25 14:52:01 -0700

Store full pixel MV in TPL stats and fix kf buffer

Change-Id: I8aaa476706df22e5ac42f58c87a1809e94585e37

Files Modified:

• vp9/encoder/vp9_tpl_model.c
• vpx/vpx_tpl.h

---

09b3d5fc5aa48752f95f4c0c37b0bd4ff55c0ba1 by Marco Paniconi <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/09b3d5fc5aa48752f95f4c0c37b0bd4ff55c0ba1
Authored: 2024-09-24 10:40:12 -0700
Committed: 2024-09-24 10:44:28 -0700

rtc-vp9: Always disable svc_use_low_part

Possible fix for issue below. It was only disabled
for screen in a previous change, but we force it off
always to check if it clears the issue.

The speed feature disabled is only used for 3 spatial
layers and at least 2 temporal. The impact on speed is
expected to be small, ~2%, so ok to disable for now and
see if it clears the issue.

Bug: 366146260
Change-Id: If7af006425e1e0ef297b9d6466507ea4c90ddb6f

Files Modified:

• vp9/encoder/vp9_speed_features.c

---

42e023d10c45a542c0f76951b9fa50fe96b476ab by Jonathan Wright <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/42e023d10c45a542c0f76951b9fa50fe96b476ab
Authored: 2024-09-17 15:33:18 +0100
Committed: 2024-09-24 14:13:43 +0000

Refactor Arm convolution transpose helpers to remove casting

Interleave elements in a different order to avoid needing to cast
between vector types of different element sizes.

Change-Id: I6b39f4beaa83dedd5b83d1547b7f0ada5650b867

Files Modified:

• vpx_dsp/arm/highbd_vpx_convolve8_sve2.c
• vpx_dsp/arm/vpx_convolve8_neon_dotprod.c
• vpx_dsp/arm/vpx_convolve8_neon_i8mm.c

---

3500e57e52b6af057ed54223e15e560a95df8479 by James Zern <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/3500e57e52b6af057ed54223e15e560a95df8479
Authored: 2024-09-23 15:26:13 -0700
Committed: 2024-09-24 01:52:35 +0000

vpx_highbd_convolve8_avg_sve2: fix C fallback typo

vpx_highbd_convolve8_c -> vpx_highbd_convolve8_avg_c

Change-Id: I8bc73c59d3e654739ee5c42a295f4ecdee6d7631

Files Modified:

• vpx_dsp/arm/highbd_vpx_convolve8_sve2.c

---

1fd9d456cd5fb2091d0b61c08e1ff340d85462cf by Diksha Singh <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/1fd9d456cd5fb2091d0b61c08e1ff340d85462cf
Authored: 2024-08-29 18:23:03 +0530
Committed: 2024-09-23 13:30:59 +0530

Add AVX2 for 12-tap filter of temporal filter prediction

This CL adds AVX2 implementation for the function
vpx_convolve8_12_c().The relevant unit tests are added.

The module level scaling w.r.t. C implementation are as follows:

For vpx_convolve8_12_avx2()
Blk Size Scaling w.r.t. C
8x8 13.43x
16x16 16.19x
32x32 18.47x

For convolve_horiz_12_avx2()
Blk Size Scaling w.r.t. C
8x8 13.11x
16x16 14.40x
32x32 16.71x

For convolve_vert_12_avx2()
Blk Size Scaling w.r.t. C
8x8 14.75x
16x16 19.93x
32x32 22.02x

For convolve_copy_12_avx2()
Blk Size Scaling w.r.t. C
8x8 5.73x
16x16 6.56x
32x32 3.60x

Change-Id: Id52b8baf7f46cb95dff8512004705cf26ebed53b

Files Added:

• vp9/encoder/x86/temporal_filter_avx2.c

Files Modified:

• test/convolve_test.cc
• vp9/common/vp9_rtcd_defs.pl
• vp9/encoder/vp9_temporal_filter.c
• vp9/encoder/vp9_temporal_filter.h
• vp9/vp9cx.mk

---

cdd4e35015eed890931905a0d4cf0a018fddd91f by Marco Paniconi <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/cdd4e35015eed890931905a0d4cf0a018fddd91f
Authored: 2024-09-20 08:36:32 -0700
Committed: 2024-09-20 09:50:41 -0700

vp8: Fix integer overflow in encode_frame_to_data_rate

Integer overflow in encode_frame_to_data_rate()
for the update:
lc->total_target_vs_actual += bits_off_for_this_layer

Fix is to use int64_t for total_target_vs_actual.

Bug: chromium:368114043
Change-Id: I9a01e1a69e26ae748e8ae23d9e1287431510388d

Files Modified:

• vp8/encoder/onyx_int.h

---

aa73610d034c441ef80d1f92250976f9f464e10e by Wan-Teh Chang <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/aa73610d034c441ef80d1f92250976f9f464e10e
Authored: 2024-09-19 00:09:36 -0700
Committed: 2024-09-19 00:09:36 -0700

Fix a typo: avg_frame_index => avg_frame_qindex

Change-Id: I8fd9f6f01ae712a9bf3dc9e34fe5f7115a305109

Files Modified:

• vp9/encoder/vp9_svc_layercontext.c

---

417204d7fd2a7b3d7242850992530485256042d2 by Marco Paniconi <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/417204d7fd2a7b3d7242850992530485256042d2
Authored: 2024-09-18 12:41:25 -0700
Committed: 2024-09-18 14:33:49 -0700

rtc-vp9: Fix to integer overflow in vp9-svc

Divide by 3 instead of multiple by 3, in comparison of
lrc->avg_frame_bandwidth vd lrc->last_avg_frame_bandwidth,
in two functions for reset rc.
Small loss in precision, so acceptable.

Similar change to:
https://chromium-review.googlesource.com/c/webm/libvpx/+/5698570

Bug: chromium:367892770
Change-Id: Ia9ef09a9f6beba930fedd496407cfa7057e39336

Files Modified:

• vp9/encoder/vp9_encoder.c
• vp9/encoder/vp9_svc_layercontext.c

---

ac68e7f99996cf1f70884b262f9d02f0dd0519d1 by James Zern <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/ac68e7f99996cf1f70884b262f9d02f0dd0519d1
Authored: 2024-09-17 13:31:46 -0700
Committed: 2024-09-18 19:43:10 +0000

aarch64_cpudetect: detect SVE/SVE2 on Windows

PF_ARM_SVE_INSTRUCTIONS_AVAILABLE and PF_ARM_SVE2_INSTRUCTIONS_AVAILABLE
are available in WinSDK 10.0.26100 and recent versions of mingw-w64.

Based on a patch by Martin Storsjö on ffmpeg-devel:
https://ffmpeg.org/pipermail/ffmpeg-devel/2024-September/333611.html

Change-Id: I34b2341a559f95aa400e84d709f3eb36da5dbb7b

Files Modified:

• vpx_ports/aarch64_cpudetect.c

---

729b78a127efe43831e76558f0d07f0c62acda81 by James Zern <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/729b78a127efe43831e76558f0d07f0c62acda81
Authored: 2024-09-17 13:13:54 -0700
Committed: 2024-09-18 19:43:10 +0000

aarch64_cpudetect: detect I8MM on Windows via SVE-I8MM

There's no direct processor feature constant for I8MM alone, but
there is a flag for SVE-I8MM (added in WinSDK 10.0.26100 and
recent versions of mingw-w64). If SVE-I8MM is available, we can
assume that I8MM is available.

While HW supporting these features isn't yet commonly running
Windows, this at least allows detecting and running the I8MM codepaths
in Windows builds in Wine (possibly running in QEMU).

Based on patch from Martin Storsjö on ffmpeg-devel:
https://ffmpeg.org/pipermail/ffmpeg-devel/2024-September/333609.html

Change-Id: I77117bee8516924fddcdecccae8bab3cf5beed96

Files Modified:

• vpx_ports/aarch64_cpudetect.c

---

6dfdc4ee1098edd7f1e9c390ae34a660b9c4751d by James Zern <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/6dfdc4ee1098edd7f1e9c390ae34a660b9c4751d
Authored: 2024-09-17 15:45:48 -0700
Committed: 2024-09-18 19:06:17 +0000

tiny_ssim: fix argc check

The program requires a minimum of 2 parameters. Previously the tool
would crash if only one input file was given.

Bug: webm:365481206
Change-Id: I875d81b2db4fcc4338061c03b23bb51b0aad58e4

Files Modified:

• tools/tiny_ssim.c

---

696c488d3581ce5258d4e01093f5f50840e2e40c by Marco Paniconi <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/696c488d3581ce5258d4e01093f5f50840e2e40c
Authored: 2024-09-13 17:21:24 -0700
Committed: 2024-09-17 23:07:24 +0000

rtc-vp9: Disable svc_use_low_part for screen

Possible fix for issue below. The speed feature disabled
is only used for 3 spatial layers and at least 2 temporal.
The impact on speed is expected to be small, ~2%, so ok
to disable for now and see if it clears the issue.

Bug: 366146260

Change-Id: I94ab991d583cc2ce758db337abbbb463a65f0767

Files Modified:

• vp9/encoder/vp9_speed_features.c

---

c6de95ce0e54c3d40fa39bb1c8cab282b7d1d877 by Jerome Jiang <[e-mail]>

https://chromium.googlesource.com/webm/libvpx/+/c6de95ce0e54c3d40fa39bb1c8cab282b7d1d877
Authored: 2024-09-06 11:18:15 -0400
Committed: 2024-09-06 13:02:12 -0400

Initialize gf_picture in vp9 tpl

Bug: b/365068397
Change-Id: Id267532928353148f916b73feb19de515db14cb9

Files Modified:

• vp9/encoder/vp9_tpl_model.c

---

Comment 1

[Update Bot]

I've submitted a try run for this commit: https://treeherder.mozilla.org/jobs?repo=try&revision=4787da7cd0c88a4941db070ce27d20b925c606be

Comment 2

[Update Bot]

Attachment: Bug 1921863 - Update libvpx to 32de9c2becdd5d96d84d37be9f7fb9de43f24a4d

Comment 3

[Update Bot]

Attachment: Bug 1921863 - Apply mozilla patches for libvpx

Comment 4

[Update Bot]

The try push is done, we found jobs with unclassified failures.

Needs Investigation (Possible Intermittents):

• browser/components/sessionstore/test/browser_label_and_icon.js
  • 2 of 4 failed on the same (retriggered) task
    - test-linux1804-64-qr/opt-mochitest-browser-chrome-swr-a11y-checks-6 (DqTBCUNSSc294O80IzYLCQ)
    - test-linux1804-64-qr/opt-mochitest-browser-chrome-swr-a11y-checks-6 (b05YXrbtSi-Q3KyGiWaH3g)

These failures could mean that the library update changed something and caused
tests to fail. You'll need to review them yourself and decide where to go from here.

In either event, I have done all I can and you will need to take it from here. If you
don't want to land my patch, you can replicate it locally for editing with
./mach vendor media/libvpx/moz.yaml

When reviewing, please note that this is external code, which needs a full and
careful inspection - not a rubberstamp.

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/ca7625a707a2
Update libvpx to 32de9c2becdd5d96d84d37be9f7fb9de43f24a4d r=chunmin
https://hg.mozilla.org/integration/autoland/rev/a789f648ce94
Apply mozilla patches for libvpx r=chunmin

Comment 6

[Serban Stanca [:SerbanS]]

https://hg.mozilla.org/mozilla-central/rev/ca7625a707a2
https://hg.mozilla.org/mozilla-central/rev/a789f648ce94