1951381 - Update aom to new version 5f6ce718d903dca3e49c5c10db0859a394c9be84 from 2025-03-07 00:00:00

Status: RESOLVED FIXED

(Core :: Audio/Video: Playback, enhancement)

Description

[Update Bot]

This update covers 17 commits. Here are the overall diff statistics, and then the commit information.

---

media/libaom/moz.yaml | 4 +-
third_party/aom/AUTHORS | 7 +
third_party/aom/CHANGELOG | 83 +++++++++++++++++++
third_party/aom/CMakeLists.txt | 12 ++-
third_party/aom/aom/aomcx.h | 8 +
third_party/aom/aom/src/aom_image.c | 4 +
third_party/aom/aom_dsp/x86/synonyms.h | 11 --
third_party/aom/aom_dsp/x86/synonyms_avx2.h | 15 ---
third_party/aom/apps/aomenc.c | 15 +++
third_party/aom/av1/arg_defs.c | 3 +
third_party/aom/av1/arg_defs.h | 1 +
third_party/aom/av1/av1_cx_iface.c | 65 ++++++++++++--
third_party/aom/av1/encoder/arm/highbd_pickrst_neon.c | 2 +-
third_party/aom/av1/encoder/block.h | 4 +
third_party/aom/av1/encoder/encodeframe.c | 4 +
third_party/aom/av1/encoder/encoder.h | 3 +
third_party/aom/av1/encoder/ratectrl.c | 22 ++++-
third_party/aom/av1/encoder/ratectrl.h | 2 +
third_party/aom/av1/encoder/var_based_part.c | 12 ++-
third_party/aom/build/cmake/aom_install.cmake | 27 ++++++
third_party/aom/build/cmake/config.cmake.in | 14 +++
third_party/aom/build/cmake/rtcd.pl | 28 +++++-
third_party/aom/test/encode_api_test.cc | 40 +++++++++
third_party/aom/test/svc_datarate_test.cc | 2 +-
24 files changed, 335 insertions(+), 53 deletions(-)

---

5f6ce718d903dca3e49c5c10db0859a394c9be84 by Marco Paniconi <[e-mail]>

https://aomedia.googlesource.com/aom/commit/5f6ce718d903dca3e49c5c10db0859a394c9be84
Authored: 2025-02-27 12:04:50 -0800
Committed: 2025-02-28 09:42:44 -0800

rtc: Relax the max_qp setting on scene changes.

Relax the max_qp setting on scene changes, based
on buffer_level, source_sad and current qp.
For now only for resolutions 720p and above, and
this change only affects camera mode.

Reduces the dips in quality in the issue below.

Bug: b:399575647
Change-Id: I631494fca4c7612f9cda760686a3b4413c2fe446

Files Modified:

• av1/encoder/ratectrl.c

---

0995a1d2a1d2ae6d81d2e4784ce84fb2d0722c05 by bmarques1995 <[e-mail]>

https://aomedia.googlesource.com/aom/commit/0995a1d2a1d2ae6d81d2e4784ce84fb2d0722c05
Authored: 2025-02-24 12:10:12 -0300
Committed: 2025-02-27 14:59:52 -0800

Add cmake find_package option

Change-Id: I4876f5fff39ae3816f6fd4a323cc39726b25f95c

Files Added:

• build/cmake/config.cmake.in

Files Modified:

• CMakeLists.txt
• build/cmake/aom_install.cmake

---

f131bb4bc3003519fe91d107d920bcafb5bd114d by Athulya Raj Raji Mohini <[e-mail]>

https://aomedia.googlesource.com/aom/commit/f131bb4bc3003519fe91d107d920bcafb5bd114d
Authored: 2025-02-21 17:06:23 +0000
Committed: 2025-02-26 09:53:29 -0800

Fix an incorrect index in av1_highbd_pixel_proj_error_neon

Fix an issue in the tail loop of av1_highbd_pixel_proj_error_neon,
where an incorrect index is used when width is not a multiple of 8.
Bug identified in scenarios where width % 8 != 0.

Change-Id: Ibac609c3ad00a1c87c4b6a19d086bd4ccc0a5934

Files Modified:

• av1/encoder/arm/highbd_pickrst_neon.c

---

5bc7fd1f89a3258973367916a2e71f86a548dfd7 by Jeremy Dorfman <[e-mail]>

https://aomedia.googlesource.com/aom/commit/5bc7fd1f89a3258973367916a2e71f86a548dfd7
Authored: 2025-02-21 17:00:46 -0500
Committed: 2025-02-21 17:09:03 -0500

rtcd: fix unstable application of --require-

Previously, --require-X arguments to rtcd.pl would be applied in hash
order, which isn't stable from run to run. Consequently, something like
--require-ssse3 --require-sse4_1 could sometimes #define functions to
SSSE3 and other times to SSE4.1 randomly. Instead, force a stable order
by sorting before calling require().

Secondarily, x86_64 and ARM64 would override the command line options
with SSE2 and NEON respectively, leading to a mildly broken RTCD with
--require-X arguments present. Instead, don't provide default
architecture requirements if --require-X are on the command line.

Change-Id: I428deefaf1961b5d5e8465ed1e259875aef03ec0

Files Modified:

• build/cmake/rtcd.pl

---

99fcd816eeaa7da46688bc4b9f4f9e71be13c2e8 by Marco Paniconi <[e-mail]>

https://aomedia.googlesource.com/aom/commit/99fcd816eeaa7da46688bc4b9f4f9e71be13c2e8
Authored: 2025-02-12 15:56:42 -0800
Committed: 2025-02-13 10:10:31 -0800

rtc-screen: Increase the sad thresh for scroll detection

For the frame-level scroll detection on superblocks for
high motion content: increase the sad thresholds to avoid
mis-detection, and make it dependent on the motion magnitude.
This reduces the psnr dips in the issue below.

Bug: b:383306740
Change-Id: I0eb37de53285d6307b790cf3d8978e492d132ee0

Files Modified:

• av1/encoder/ratectrl.c

---

c2f3c96841132c7973b4349057d01f4a4d0cb31e by Yunqing Wang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/c2f3c96841132c7973b4349057d01f4a4d0cb31e
Authored: 2025-02-11 10:50:25 -0800
Committed: 2025-02-11 10:53:06 -0800

Add extra comment for AV1E_SET_ENABLE_LOW_COMPLEXITY_DECODE

Change-Id: Id8fd5e0d08bc9f8b84cb9bf09ab08a9a3159ace1

Files Modified:

• aom/aomcx.h

---

af629abd92b7247e5d8c678e294d68bbefc53dc4 by Marco Paniconi <[e-mail]>

https://aomedia.googlesource.com/aom/commit/af629abd92b7247e5d8c678e294d68bbefc53dc4
Authored: 2025-02-07 15:23:48 -0800
Committed: 2025-02-10 17:21:56 -0800

rtc-screen: Bring back the num_col/row_blscroll_last_tl0

Removed in:
https://aomedia-review.googlesource.com/c/aom/+/196922

This brings back some speedup for screen content.

Change-Id: Ia6b59be7199a36a78a139ea9b3af4dc09b4ce25c

Files Modified:

• av1/encoder/block.h
• av1/encoder/encodeframe.c
• av1/encoder/ratectrl.c
• av1/encoder/ratectrl.h
• av1/encoder/var_based_part.c

---

8ff23cf31ea344c574600f52d6d2b34413bf3058 by Yunqing Wang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/8ff23cf31ea344c574600f52d6d2b34413bf3058
Authored: 2025-02-07 15:01:59 -0800
Committed: 2025-02-10 14:49:20 -0800

Add enable_low_complexity_decode mode

Added enable_low_complexity_decode encode option and related API
control for later development.

Change-Id: I50ca0b1dad8978aae115ac627abab14b52409268

Files Modified:

• aom/aomcx.h
• apps/aomenc.c
• av1/arg_defs.c
• av1/arg_defs.h
• av1/av1_cx_iface.c
• av1/encoder/encoder.h

---

3b624af45b86646a20b11a9ff803aeae588cdee6 by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/3b624af45b86646a20b11a9ff803aeae588cdee6
Authored: 2025-02-07 14:30:31 -0800
Committed: 2025-02-10 12:59:13 -0800

Update CHANGELOG for libaom v3.12.0

Bug: 394596176
Change-Id: Ib4e56b433fadab1d4a8d5cb4c8fd5b870f53ac6a

Files Modified:

• CHANGELOG

---

1b25d85cf223225e4199eb62d3a2c947e3d416be by James Zern <[e-mail]>

https://aomedia.googlesource.com/aom/commit/1b25d85cf223225e4199eb62d3a2c947e3d416be
Authored: 2025-02-05 15:10:01 -0800
Committed: 2025-02-10 11:42:05 -0800

aom_image: default cp/tc/mc fields to UNSPECIFIED

The image struct is memset to 0, but the AOM_CICP_*_UNSPECIFIED values
are non-zero. This change ensures they're set to something reasonable by
default.

Change-Id: I9b436c09f07e2879dcc57a85ca8e960e74d276aa

Files Modified:

• aom/src/aom_image.c

---

e107085626c2336dce00198cd52eb604bc33e0a9 by James Zern <[e-mail]>

https://aomedia.googlesource.com/aom/commit/e107085626c2336dce00198cd52eb604bc33e0a9
Authored: 2025-02-05 11:49:40 -0800
Committed: 2025-02-10 11:42:05 -0800

ensure subsampling is 0 w/AOM_CICP_MC_IDENTITY

From the spec:
6.4.2 Color config semantics
If matrix_coefficients is equal to MC_IDENTITY, it is a requirement of
bitstream conformance that subsampling_x is equal to 0 and subsampling_y
is equal to 0.

Bug: chromium:392929025
Change-Id: Ia2f9b6b3040867dc065ad9054de97993d81785be

Files Modified:

• av1/av1_cx_iface.c
• test/encode_api_test.cc

---

ef88d9c022dcfcfad404dabbf4b43d22e3d35a8f by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/ef88d9c022dcfcfad404dabbf4b43d22e3d35a8f
Authored: 2025-02-07 10:59:13 -0800
Committed: 2025-02-07 13:36:37 -0800

Fix a comment typo

Bug: b:394563983
Change-Id: Iaa7148962df5afcbfa554bad42026bae361ca150
(cherry picked from commit 96c8ce6160dea6ff5c7d102f561486d000f4584a)

Files Modified:

• test/svc_datarate_test.cc

---

96c8ce6160dea6ff5c7d102f561486d000f4584a by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/96c8ce6160dea6ff5c7d102f561486d000f4584a
Authored: 2025-02-07 10:59:13 -0800
Committed: 2025-02-07 10:59:13 -0800

Fix a comment typo

Bug: b:394563983
Change-Id: Iaa7148962df5afcbfa554bad42026bae361ca150

Files Modified:

• test/svc_datarate_test.cc

---

3d000065136e16d79c139d2f467c1e7b153e769a by Marco Paniconi <[e-mail]>

https://aomedia.googlesource.com/aom/commit/3d000065136e16d79c139d2f467c1e7b153e769a
Authored: 2025-02-06 11:36:19 -0800
Committed: 2025-02-06 21:13:36 -0800

rtc: Remove num_col/row_blscroll_last_tl0

These parameters are causing the tsan race condition
in the issue below.

Removing them for now. This may cause some slowdown in
some cases of webrtc usage, so we may add them back in
later.

Add screen 2TL test with 4 threads to catch the
issue below.

Bug: b:394563983
Change-Id: I9d1abc58ecc57921ddbbd79fb1da1dc114f974ac
(cherry picked from commit 2454213a72f194522c98610ef17057c226ffe391)

Files Modified:

• av1/encoder/ratectrl.c
• av1/encoder/ratectrl.h
• av1/encoder/var_based_part.c
• test/svc_datarate_test.cc

---

5850fbf90d70da5825696fcfed3f0fcbb704a8d9 by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/5850fbf90d70da5825696fcfed3f0fcbb704a8d9
Authored: 2025-02-06 14:00:39 -0800
Committed: 2025-02-06 14:36:29 -0800

Update CHANGELOG for libaom v3.12.0

Bug: 394596176
Change-Id: I8e3c4194090a0d2941ab936065441e298f1ea906

Files Modified:

• CHANGELOG

---

68a1f6a5aaf3a5623c69b2492e2add84685ad9bd by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/68a1f6a5aaf3a5623c69b2492e2add84685ad9bd
Authored: 2025-02-05 17:01:19 -0800
Committed: 2025-02-06 10:20:16 -0800

Document new value 3 for AV1E_SET_ENABLE_CDEF

The comment in aom/aomcx.h was missed in
https://aomedia-review.googlesource.com/c/aom/+/194641.

Also change "on frame qindex" to "based on frame qindex".

Bug: aomedia:375221136
Change-Id: I110951c3f59661659ce32821620df9103cd7c7e0
(cherry picked from commit 39e039b28bbda0263f64fc24966b3345d74a58a9)

Files Modified:

• aom/aomcx.h
• av1/arg_defs.c
• av1/encoder/pickcdef.h

---

7e257ce038d11159235332af963e4383fb384ec3 by Wan-Teh Chang <[e-mail]>

https://aomedia.googlesource.com/aom/commit/7e257ce038d11159235332af963e4383fb384ec3
Authored: 2025-02-05 14:16:52 -0800
Committed: 2025-02-05 14:23:04 -0800

Update AUTHORS,CHANGELOG,CMakeLists.txt for 3.12.0

Bug: 394596176
Change-Id: Ifb8ab56f868e0a7123a4b9b4443f4ad8c8c17a4b

Files Modified:

• AUTHORS
• CHANGELOG
• CMakeLists.txt

---

Comment 1

[Update Bot]

I've submitted a try run for this commit: https://treeherder.mozilla.org/jobs?repo=try&revision=66f540aa7523291cd109f440648afa965045b815

Comment 2

[Update Bot]

Attachment: Bug 1951381 - Update aom to 5f6ce718d903dca3e49c5c10db0859a394c9be84

Comment 3

[Update Bot]

Attachment: Bug 1951381 - Apply mozilla patches for aom

Comment 4

[Update Bot]

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

Needs Close Investigation:

• test-windows11-64-24h2/opt-crashtest-ioi - 4 of 4 failed on the same (retriggered) task (failed: eypREhSTS1erwwp3BrrBxg, FSjVkXQKQxKwJ1PneRPg8Q, IQC07n7nRw60FnUU52g3uw, W1XjOir0RzC5CBtH8INfcg)
• test-android-em-7.0-x86_64-lite-qr/opt-geckoview-junit-fis - 1 of 4 failed on the same (retriggered) task (failed: b0l6KfZKSIG18Qx2S3Vxvw)

Known Issues:

• toolkit/mozapps/extensions/test/browser/browser_installtrigger_install.js
  - test-linux1804-64-qr/opt-mochitest-browser-chrome-swr-a11y-checks-7 (XAxqjjxxTziNhvOR1QVnGQ)

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/libaom/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/74b594144525
Update aom to 5f6ce718d903dca3e49c5c10db0859a394c9be84 r=padenot
https://hg.mozilla.org/integration/autoland/rev/163afaf72b80
Apply mozilla patches for aom r=padenot

Comment 6

[amarc]

https://hg.mozilla.org/mozilla-central/rev/74b594144525
https://hg.mozilla.org/mozilla-central/rev/163afaf72b80