James Zern [Mon, 22 Aug 2022 22:21:46 +0000 (22:21 +0000)]
Merge "highbd_quantize_neon.c: remove unneeded assert.h" into main
James Zern [Mon, 22 Aug 2022 22:21:28 +0000 (22:21 +0000)]
Merge "vp9,search_new_mv: descale rather than scale sse" into main
James Zern [Mon, 22 Aug 2022 22:21:00 +0000 (22:21 +0000)]
Merge changes Iabed118b,I60a384b2 into main
* changes:
use VPX_NO_UNSIGNED_SHIFT_CHECK with entropy functions
compiler_attributes.h: add VPX_NO_UNSIGNED_SHIFT_CHECK
Konstantinos Margaritis [Mon, 22 Aug 2022 19:46:50 +0000 (19:46 +0000)]
[NEON] Add vpx_highbd_subtract_block function
Total gain for 12-bit encoding:
* ~1% for best and rt profile
Change-Id: I4039120dc570baab1ae519a5e38b1acff38d81f0
Konstantinos Margaritis [Fri, 19 Aug 2022 22:00:42 +0000 (22:00 +0000)]
[NEON] Added vpx_highbd_sad* functions
Total gain for 12-bit encoding:
* ~7.8% for best profile
* ~10% for rt profile
Change-Id: I89eda5c4372a5b628c9df84cdeb4c8486fc44789
James Zern [Mon, 22 Aug 2022 17:48:40 +0000 (10:48 -0700)]
highbd_quantize_neon.c: remove unneeded assert.h
Change-Id: I041f5fb23b856a2b519669b5bf8a40d3772b4a6e
James Zern [Mon, 22 Aug 2022 17:45:52 +0000 (17:45 +0000)]
Merge "[NEON] Added vpx_highbd_quantize_b* functions" into main
Scott LaVarnway [Mon, 22 Aug 2022 10:36:09 +0000 (10:36 +0000)]
Merge "Fix TEST_P(SADx4Test, DISABLED_Speed)" into main
Konstantinos Margaritis [Fri, 12 Aug 2022 17:41:11 +0000 (17:41 +0000)]
[NEON] Added vpx_highbd_quantize_b* functions
Total gain for 12-bit encoding:
* ~4.8% for best profile
* ~6.2% for rt profile
Change-Id: I61e646ab7aedf06a25db1365d6d1cf7b05101c21
James Zern [Sat, 20 Aug 2022 00:00:06 +0000 (00:00 +0000)]
Merge "loopfilter.c: normalize flat func param type" into main
James Zern [Thu, 18 Aug 2022 18:35:06 +0000 (11:35 -0700)]
use VPX_NO_UNSIGNED_SHIFT_CHECK with entropy functions
these shift values off the most significant bit as part of the process;
vp8_regular_quantize_b_sse4_1 is included here for a special case of
mask creation
quiets warnings of the form:
vp8/decoder/dboolhuff.h:81:11: runtime error: left shift of
2373679303235599696 by 3 places cannot be represented in type
'VP8_BD_VALUE' (aka 'unsigned long')
vp8/encoder/bitstream.c:257:18: runtime error: left shift of
2147493041
by 1 places cannot be represented in type 'unsigned int'
vp8/encoder/x86/quantize_sse4.c:114:18: runtime error: left shift of
4294967294 by 1 places cannot be represented in type 'unsigned int'
vp9/encoder/vp9_pickmode.c:1632:41: runtime error: left shift of
4294967295 by 1 places cannot be represented in type 'unsigned int'
Bug: b/
229626362
Change-Id: Iabed118b2a094232783e5ad0e586596d874103ca
James Zern [Thu, 18 Aug 2022 02:20:25 +0000 (19:20 -0700)]
compiler_attributes.h: add VPX_NO_UNSIGNED_SHIFT_CHECK
and use it on MD5Transform(); this behavior is well defined and is only
a warning with -fsanitize=integer, not -fsanitize=undefined.
quiets warnings of the form:
md5_utils.c:163:3: runtime error: left shift of
143704723 by 7 places
cannot be represented in type 'unsigned int'
Bug: b/
229626362
Change-Id: I60a384b2c2556f5ce71ad8ebce050329aba0b4e4
James Zern [Thu, 18 Aug 2022 18:57:32 +0000 (11:57 -0700)]
vp9,search_new_mv: descale rather than scale sse
this changes from scaling best sse to downscaling base sse in
comparisons.
this quiets an integer sanitizer warning of the form:
vp9/encoder/vp9_pickmode.c:1632:41: runtime error: left shift of
4294967295 by 1 places cannot be represented in type 'unsigned int'
Bug: b/
229626362
Change-Id: Iee2920474ba700a46177d4514ba6ef7691958069
James Zern [Thu, 18 Aug 2022 18:56:25 +0000 (11:56 -0700)]
update_thresh_freq_fact_row_mt: normalize param types
make source_variance unsigned; this matches update_thresh_freq_fact()
and the type of the MACROBLOCK member.
quiets integer sanitizer warnings of the form:
vp9/encoder/vp9_pickmode.c:2710:58: runtime error: implicit conversion
from type 'unsigned int' of value
4294967295 (32-bit, unsigned) to type
'int' changed the value to -1 (32-bit, signed)
Bug: b/
229626362
Change-Id: I812c6ca914507bf25cad323dea3d91a3a2ea4f1d
James Zern [Thu, 18 Aug 2022 18:45:49 +0000 (11:45 -0700)]
loopfilter.c: normalize flat func param type
flat/flat2 are stored as int8_t as returned by the filter_mask*
functions.
this quiets integer sanitizer warnings of the form:
vpx_dsp/loopfilter.c:197:28: runtime error: implicit conversion from
type 'int8_t' (aka 'signed char') of value -1 (8-bit, signed) to type
'uint8_t' (aka 'unsigned char') changed the value to 255 (8-bit,
unsigned)
Bug: b/
229626362
Change-Id: Iacb6ae052d4cb2b6e0ebccbacf59ece9501d3b5f
James Zern [Thu, 18 Aug 2022 23:16:33 +0000 (23:16 +0000)]
Merge changes Icfc59932,I3d1ca618,Id3966912,I56f74981,Ia9a5dc5e, ... into main
* changes:
vpx_encoder.h: make flag constants unsigned
vp8,VP8_COMP: normalize segment_encode_breakout type
webmdec,WebmInputContext: make timestamp_ns signed
highbd_quantize_intrin_sse2: quiet int sanitizer warnings
load_unaligned_u32: use an int w/_mm_cvtsi32_si128
variance_sse2.c: add some missing casts
Scott LaVarnway [Thu, 18 Aug 2022 19:19:08 +0000 (12:19 -0700)]
Fix TEST_P(SADx4Test, DISABLED_Speed)
The reference code was being timed instead of the optimized code.
Change-Id: I67eb08dcda80e20eaa075dc2c91b7e8ef5c0cdfb
Scott LaVarnway [Wed, 17 Aug 2022 10:59:57 +0000 (10:59 +0000)]
Merge "Add vp9_highbd_quantize_fp_32x32_neon()." into main
James Zern [Wed, 17 Aug 2022 04:59:20 +0000 (21:59 -0700)]
vpx_encoder.h: make flag constants unsigned
this matches the type for vpx_codec_frame_flags_t and
vpx_codec_er_flags_t and quiets int sanitizer warnings of the form:
implicit conversion from type 'int' of value -9 (32-bit, signed) to type
'unsigned int' changed the value to
4294967287 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: Icfc5993250f37cedb300c7032cab28ce4bec1f86
James Zern [Wed, 17 Aug 2022 00:46:24 +0000 (17:46 -0700)]
vp8,VP8_COMP: normalize segment_encode_breakout type
use unsigned int as the API value is of this type; this quiets some
integer sanitizer warnings of the form:
implicit conversion from type 'unsigned int' of value
2147483648
(32-bit, unsigned) to type 'int' changed the value to -
2147483648
(32-bit, signed)
Bug: b/
229626362
Change-Id: I3d1ca618bf1b3cd57a5dca65a3067f351c1473f8
James Zern [Wed, 17 Aug 2022 00:54:14 +0000 (17:54 -0700)]
webmdec,WebmInputContext: make timestamp_ns signed
this matches the type returned from libwebm, which uses -1 as an error;
quiets integer sanitizer warnings of the form:
implicit conversion from type 'long long' of value -1 (64-bit, signed)
to type 'uint64_t' (aka 'unsigned long') changed the value to
18446744073709551615 (64-bit, unsigned)
Bug: b/
229626362
Change-Id: Id3966912f802aee3c0f7852225b55f3057c3e76a
James Zern [Wed, 17 Aug 2022 00:28:08 +0000 (17:28 -0700)]
highbd_quantize_intrin_sse2: quiet int sanitizer warnings
add a missing cast in ^ operations; quiets warnings of the form:
implicit conversion from type 'int' of value -1 (32-bit, signed) to type
'unsigned int' changed the value to
4294967295 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: I56f74981050b2c9d00bad20e68f1b73ce7454729
James Zern [Tue, 16 Aug 2022 23:52:06 +0000 (16:52 -0700)]
load_unaligned_u32: use an int w/_mm_cvtsi32_si128
this matches the type of the function parameter; quiets integer
sanitizer warnings of the form:
implicit conversion from type 'uint32_t' (aka 'unsigned int') of value
3215646151 (32-bit, unsigned) to type 'int' changed the value to
-
1079321145 (32-bit, signed)
Bug: b/
229626362
Change-Id: Ia9a5dc5e1f57cbf4f8f8fa457bb674ef43369d37
James Zern [Tue, 16 Aug 2022 23:48:00 +0000 (16:48 -0700)]
variance_sse2.c: add some missing casts
quiets integer sanitizer warnings of the form:
../vpx_dsp/x86/variance_sse2.c:100:10: runtime error: implicit
conversion from type 'unsigned int' of value
4294966272 (32-bit,
unsigned) to type 'int' changed the value to -1024 (32-bit, signed)
Bug: b/
229626362
Change-Id: I150cc0a6a6b85143c3bf96886686fe3a40897db5
James Zern [Tue, 16 Aug 2022 20:57:25 +0000 (13:57 -0700)]
configure: add -Wno-pass-failed for libyuv
with certain optimization flags or sanitizers enabled some code may fail
to vectorize:
third_party/libyuv/source/row_common.cc:3178:7: warning: loop not
vectorized: the optimizer was unable to perform the requested
transformation; the transformation might be disabled or specified as
part of an unsupported transformation ordering
[-Wpass-failed=transform-warning]
this was observed with integer/undefined sanitizers using clang 11/13
Bug: b/
229626362
Change-Id: I01595c641763c4cd4242e02f2cc5cbabfe69d03e
James Zern [Tue, 16 Aug 2022 20:04:09 +0000 (20:04 +0000)]
Merge "configure: add -Wc++{14,17,20}-extensions" into main
Scott LaVarnway [Tue, 16 Aug 2022 13:12:21 +0000 (06:12 -0700)]
Add vp9_highbd_quantize_fp_32x32_neon().
Up to 2.6x faster than vp9_highbd_quantize_fp_32x32_c() for full
calculations.
Bug: b/
237714063
Change-Id: Icfeff2ad4dcd57d0ceb47fe04789710807b9cbad
James Zern [Tue, 16 Aug 2022 00:26:41 +0000 (17:26 -0700)]
simple_encode.cc: clear -Wextra-semi-stmt warnings
fixes warnings of the form:
../vp9/simple_encode.cc:755:48: warning: empty expression statement has
no effect; remove unnecessary ';' to silence this warning
[-Wextra-semi-stmt]
SET_STRUCT_VALUE(config, oxcf, ret, key_freq);
Bug: b/
229626362
Change-Id: I1c9b0ae9927cdd7c31da000633bcb6e2b8242cd4
James Zern [Mon, 15 Aug 2022 23:21:53 +0000 (23:21 +0000)]
Merge "examples/svc_encodeframe.c: rm empty {}s in switch" into main
Scott LaVarnway [Mon, 15 Aug 2022 21:34:42 +0000 (21:34 +0000)]
Merge "VPX: Add vp9_highbd_quantize_fp_neon()." into main
Scott LaVarnway [Fri, 12 Aug 2022 14:15:33 +0000 (07:15 -0700)]
vp9_quantize_fp_32x32_neon() cleanup.
No change in performance.
Bug: b/
237714063
Change-Id: If6ad5fc27de4babe0bfff3fdbb4b7fd99a0544ab
Scott LaVarnway [Mon, 15 Aug 2022 13:14:49 +0000 (06:14 -0700)]
VPX: Add vp9_highbd_quantize_fp_neon().
Up to 4.1x faster than vp9_highbd_quantize_fp_c() for full
calculations.
~1.3% overall encoder improvement for the test clip used.
Bug: b/
237714063
Change-Id: I8c6466bdbcf1c398b1d8b03cab4165c1d8556b0c
James Zern [Sun, 14 Aug 2022 22:43:36 +0000 (15:43 -0700)]
examples/svc_encodeframe.c: rm empty {}s in switch
these have been unnecessary since:
0e97e7049 remove spatial svc experiment
Bug: b/
229626362
Change-Id: I57528af4dcb9092b752161c8eaba2e2808c29c5f
James Zern [Sat, 13 Aug 2022 05:24:39 +0000 (22:24 -0700)]
configure: add -Wc++{14,17,20}-extensions
the snapshot of googletest and the test files themselves are targeting
c++11 currently; these warnings are supported by recent versions of
clang
Change-Id: I5d36b3bd4058ba1610f0c8b27cad27aadee85939
Scott LaVarnway [Fri, 12 Aug 2022 00:27:02 +0000 (00:27 +0000)]
Merge "VPX: vp9_quantize_fp_neon() cleanup." into main
Scott LaVarnway [Wed, 10 Aug 2022 12:29:50 +0000 (05:29 -0700)]
VPX: vp9_quantize_fp_neon() cleanup.
No change in performance.
Bug: b/
237714063
Change-Id: I868cda7acb0de840fbc85b23f3e36c50b39c331b
James Zern [Wed, 10 Aug 2022 22:38:45 +0000 (22:38 +0000)]
Merge "vp9_cx_iface,encoder_encode: only calc ts when img!=NULL" into main
Scott LaVarnway [Tue, 9 Aug 2022 20:20:52 +0000 (20:20 +0000)]
Merge "VPX: Fix vp9_quantize_fp_avx2() VS build error." into main
Scott LaVarnway [Mon, 8 Aug 2022 22:09:32 +0000 (15:09 -0700)]
VPX: Fix vp9_quantize_fp_avx2() VS build error.
Add build fix for _mm256_extract_epi16() being undefined.
Bug: b/
237714063
Change-Id: I855b1828ce1b6b2b2f063fe097999481881bf074
Cheng Chen [Mon, 8 Aug 2022 17:12:39 +0000 (10:12 -0700)]
Use level defined min gf interval
Assume the level definition of min_gf_interval is the minimum allowed
gf_interval. We take this level comformant min_gf_interval instead of
+1.
Change-Id: I9c7e62f210c95b356e9716579ee4c19638de8e35
Cheng Chen [Sat, 6 Aug 2022 01:06:08 +0000 (18:06 -0700)]
L2E: Add target level in GOP unit tests
Change-Id: Icecc3031e1052bb5a94f6c5957ec5190aae990ba
Cheng Chen [Tue, 9 Aug 2022 00:33:51 +0000 (00:33 +0000)]
Merge "Fix VP9 auto level" into main
James Zern [Mon, 8 Aug 2022 18:28:27 +0000 (11:28 -0700)]
vp9_cx_iface,encoder_encode: only calc ts when img!=NULL
avoid calculating the end timestamp when performing a flush to prevent
an implicit conversion warning when applying a non-zero offset to a 0
pts used in that case:
vp9/vp9_cx_iface.c:1361:50: runtime error: implicit conversion from type
'vpx_codec_pts_t' (aka 'long') of value -15 (64-bit, signed) to type
'unsigned long' changed the value to
18446744073709551601 (64-bit,
unsigned)
Bug: b/
229626362
Change-Id: I68ba19b7d6de35cc185707dfb6b43406b7165035
Cheng Chen [Sat, 6 Aug 2022 00:32:12 +0000 (17:32 -0700)]
Fix VP9 auto level
The iteration index is wrong, causing the starting level to be chosen
is "LEVEL_5_2", which is intended for videos of a large resolution.
Change-Id: Id88836981bdcbd7494bd06193d6a433ac75a6d2e
Scott LaVarnway [Fri, 5 Aug 2022 14:40:26 +0000 (07:40 -0700)]
VPX: Add vpx_subtract_block_avx2().
~1.3x faster than vpx_subtract_block_sse2().
Based on aom_subtract_block_avx2().
Bug: b/
241580104
Change-Id: I17da036363f213d53c6546c3e858e4c3cba44a5b
James Zern [Thu, 4 Aug 2022 18:09:02 +0000 (18:09 +0000)]
Merge "vp9_active_[hv]_edge: add missing vpx_clear_system_state" into main
Scott LaVarnway [Wed, 3 Aug 2022 15:01:25 +0000 (08:01 -0700)]
VPX: Add vp9_highbd_quantize_fp_32x32_avx2().
~4x faster than vp9_highbd_quantize_fp_32x32_c() for full
calculations.
Bug: b/
237714063
Change-Id: Iff2182b8e7b1ac79811e33080d1f6cac6679382d
Scott LaVarnway [Tue, 2 Aug 2022 18:22:04 +0000 (11:22 -0700)]
VPX: Add vp9_highbd_quantize_fp_avx2().
Up to 5.37x faster than vp9_highbd_quantize_fp_c() for full
calculations.
~1.6% overall encoder improvement for the test clip used.
Bug: b/
237714063
Change-Id: I584fd1f60a3e02f1ded092de98970725fc66c5b8
Scott LaVarnway [Tue, 2 Aug 2022 15:09:40 +0000 (15:09 +0000)]
Merge "VPX: Add vp9_quantize_fp_32x32_avx2()." into main
James Zern [Mon, 1 Aug 2022 18:21:32 +0000 (18:21 +0000)]
Merge "Provide Arm SDOT optimizations for SAD functions" into main
Scott LaVarnway [Wed, 27 Jul 2022 14:37:44 +0000 (07:37 -0700)]
VPX: Add vp9_quantize_fp_32x32_avx2().
Up to 1.80x faster than vp9_quantize_fp_32x32_ssse3() for full
calculations.
Bug: b/
237714063
Change-Id: Ic4ae4724fce7ac85c7a089535b16a999e02f0a10
Wan-Teh Chang [Fri, 29 Jul 2022 22:42:15 +0000 (15:42 -0700)]
Fix off-by-one error of max w/h in validate_config
Fix the off-by-one errors of maximum g_w and g_h in validate_config().
Bug: webm:1774
Change-Id: I343783d06c1f53222be2366be79171b214486201
Konstantinos Margaritis [Fri, 29 Jul 2022 09:40:53 +0000 (09:40 +0000)]
Provide Arm SDOT optimizations for SAD functions
Change-Id: I497ee1c45d1fc4d643cefad7d87e5aaacd77869c
James Zern [Fri, 29 Jul 2022 02:05:22 +0000 (02:05 +0000)]
Merge changes I0c6604ef,Id7e13b3d,I7291d9bd,Ic7c0a2e7,Ic7ce0fd9, ... into main
* changes:
x86: normalize type with _mm_cvtsi128_si32
vp9_filter_block_plane_non420: fix implicit conversion warnings
variance_avx2.c: fix implicit conversion warnings
vp8,read_mb_modes_mv: fix implicit conversion warnings
vp8_find_near_mvs: fix implicit conversion warnings
encode_test_driver: normalize frame_flags type
vp9,decoder_decode: fix ubsan null/zero offset warning
y4m_input_fetch_frame: fix ubsan null/zero offset warning
James Zern [Thu, 28 Jul 2022 01:56:22 +0000 (18:56 -0700)]
vp9_active_[hv]_edge: add missing vpx_clear_system_state
this fixes runtime errors with clang -fsanitize=integer in x86 builds:
../vp9/encoder/vp9_rdopt.c:3250:17: runtime error: signed integer
overflow: 18 - -
2147483648 cannot be represented in type 'int'
../vp9/encoder/vp9_rdopt.c:3277:16: runtime error: signed integer
overflow: 26 - -
2147483648 cannot be represented in type 'int'
Bug: b/
229626362
Change-Id: Ic9a5063c840b4fce7056f61362234721add056a6
James Zern [Wed, 27 Jul 2022 22:22:37 +0000 (15:22 -0700)]
x86: normalize type with _mm_cvtsi128_si32
prefer int in most cases
w/clang -fsanitize=integer fixes warnings of the form:
implicit conversion from type 'int' of value -
809931979 (32-bit, signed)
to type 'uint32_t' (aka 'unsigned int') changed the value to
3485035317
(32-bit, unsigned)
Bug: b/
229626362
Change-Id: I0c6604efc188f2660c531eddfc7aa10060637813
James Zern [Wed, 27 Jul 2022 22:48:24 +0000 (15:48 -0700)]
vp9_filter_block_plane_non420: fix implicit conversion warnings
w/clang -fsanitize=integer fixes warnings of the form:
implicit conversion from type 'int' of value -2 (32-bit, signed) to type
'unsigned int' changed the value to
4294967294 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: Id7e13b3d494ccd1a2351db8fab6fdb6a9a771d51
James Zern [Wed, 27 Jul 2022 22:41:00 +0000 (15:41 -0700)]
variance_avx2.c: fix implicit conversion warnings
w/clang -fsanitize=integer fixes warnings of the form:
implicit conversion from type 'int' of value -1323 (32-bit, signed) to
type 'unsigned int' changed the value to
4294965973 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: I7291d9bd5cacea0d88d9f4c4624c096764f4a472
James Zern [Wed, 27 Jul 2022 22:31:00 +0000 (15:31 -0700)]
vp8,read_mb_modes_mv: fix implicit conversion warnings
w/clang -fsanitize=integer fixes warnings of the form:
implicit conversion from type 'uint32_t' (aka 'unsigned int') of value
4294443008 (32-bit, unsigned) to type 'int' changed the value to -524288
(32-bit, signed)
Bug: b/
229626362
Change-Id: Ic7c0a2e7b64a1dd6fd5cc64adcd5765318c2a956
James Zern [Wed, 27 Jul 2022 22:15:25 +0000 (15:15 -0700)]
vp8_find_near_mvs: fix implicit conversion warnings
unsigned -> int and vice versa
reported by clang -fsanitize=integer
vp8/common/findnearmv.c:108:11: runtime error: implicit conversion from
type 'uint32_t' (aka 'unsigned int') of value
4294443008 (32-bit,
unsigned) to type 'int' changed the value to -524288 (32-bit, signed)
vp8/common/findnearmv.c:110:33: runtime error: implicit conversion from
type 'int' of value -524288 (32-bit, signed) to type 'uint32_t' (aka
'unsigned int') changed the value to
4294443008 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: Ic7ce0fd98255ccf9307ac73e9fb6a8189b268214
James Zern [Wed, 27 Jul 2022 20:29:41 +0000 (13:29 -0700)]
encode_test_driver: normalize frame_flags type
use vpx_enc_frame_flags_t; this avoids int -> unsigned conversion
warnings; reported w/clang -fsanitize=integer:
test/error_resilience_test.cc:95:9: runtime error: implicit conversion
from type 'int' of value -
12845057 (32-bit, signed) to type 'unsigned
long' changed the value to
4282122239 (32-bit, unsigned)
Bug: b/
229626362
Change-Id: I0fc1dbe44a258f397cf1a05347d8cb86ee70b1b8
James Zern [Wed, 27 Jul 2022 02:26:23 +0000 (19:26 -0700)]
vp9,decoder_decode: fix ubsan null/zero offset warning
reported under clang-13. null data may be passed as a flush; move
data_end after that check
vp9/vp9_dx_iface.c:337:40: runtime error: applying zero offset to null
pointer
Bug: b/
229626362
Change-Id: I845726fd6eb6ac7a776e49272c6477a5ad30ffdf
James Zern [Wed, 27 Jul 2022 02:08:40 +0000 (19:08 -0700)]
y4m_input_fetch_frame: fix ubsan null/zero offset warning
reported under clang-13; use a while loop in file_read() to force a size
check before attempting to read. buf (aux_buf) may be may be null when
no conversion is necessary.
y4minput.c:29:43: runtime error: applying zero offset to null pointer
Bug: b/
229626362
Change-Id: Ia3250d6ff9c325faf48eaa31f4399e20837f8f7b
Scott LaVarnway [Mon, 25 Jul 2022 17:58:44 +0000 (10:58 -0700)]
VPX: vp9_quantize_fp_avx2() cleanup.
No change in performance.
Bug: b/
237714063
Change-Id: I8ea42759cc4dc57be6a29c23784997cb90ad4090
Scott LaVarnway [Wed, 27 Jul 2022 12:54:09 +0000 (12:54 +0000)]
Merge "VPX: Add vpx_highbd_quantize_b_32x32_avx2()." into main
James Zern [Wed, 27 Jul 2022 01:17:22 +0000 (18:17 -0700)]
highbd_temporal_filter_sse4: remove unused function params
this clears warnings under clang-13 of the form:
vp9/encoder/x86/highbd_temporal_filter_sse4.c|196 col 63| warning:
parameter 'v_pre' set but not used [-Wunused-but-set-parameter]
this is the high-bitdepth version of:
73b8aade8 temporal_filter_sse4: remove unused function params
Change-Id: I9b2c9bf27c16975e4855df6a2c967da4c8c63a3a
Scott LaVarnway [Mon, 25 Jul 2022 12:52:27 +0000 (05:52 -0700)]
VPX: Add vpx_highbd_quantize_b_32x32_avx2().
Up to 11.78x faster than vpx_quantize_b_32x32_sse2() for full
calculations.
~1.7% overall encoder improvement for the test clip used.
Bug: b/
237714063
Change-Id: Ib759056db94d3487239cb2748ffef1184a89ae18
Scott LaVarnway [Wed, 20 Jul 2022 16:51:56 +0000 (09:51 -0700)]
VPX: Add vpx_highbd_quantize_b_avx2().
Up to 3.61x faster than vpx_highbd_quantize_b_sse2() for full
calculations.
~2.3% overall encoder improvement for the test clip used.
Bug: b/
237714063
Change-Id: I23f88d2a7f96aaa4103778372f4f552207f73cee
Scott LaVarnway [Mon, 25 Jul 2022 11:38:13 +0000 (11:38 +0000)]
Merge "VPX: Add vpx_quantize_b_32x32_avx2()." into main
Cheng Chen [Fri, 22 Jul 2022 20:27:51 +0000 (20:27 +0000)]
Merge "L2E: Add more unit tests for GOP API" into main
Cheng Chen [Wed, 20 Jul 2022 22:35:33 +0000 (15:35 -0700)]
L2E: Add more unit tests for GOP API
Add unit tests for a 4 frame video, which could be considered as a
corner case.
Three different GOP settings are tested and verified as valid.
(1). The first GOP has 3 coding frames, no alt ref.
The second GOP has 1 coding frame, no alt ref.
The numer of coding frames is 4.
Their frame types are: keyframe, inter_frame, inter_frame,
golden_frame.
(2). The first GOP has 4 coding frames, use alt ref.
The second GOP has 1 coding frame, which is the overlay of
the first GOP's alt ref frame.
The numer of coding frames is 5.
Their types are: keyframe, alt_ref, inter_frame, inter_frame,
overlay_frame.
(3). Only one GOP with 4 coding frames, do not use alt ref.
The numer of coding frames is 4.
Their types are: keyframe, inter_frame, inter_frame, inter_frame.
Change-Id: I4079ff5065da79834b363b1e1976f65efed3f91f
James Zern [Wed, 20 Jul 2022 00:16:05 +0000 (17:16 -0700)]
avg_intrin_avx2: rm dead store in highbd_hadamard_8x8
missed in:
53dd1e8e7 avg_intrin_{sse2,avg2}: rm dead store in hadamard_8x8
Change-Id: I378e4a388ceb193a4cfee4d9d317fc62fcc4b39e
James Zern [Tue, 19 Jul 2022 23:32:30 +0000 (16:32 -0700)]
pp_filter_test: quiet static analysis warning
in CheckLowFilterOutput(); use std::unique_ptr to avoid spurious memory
leak warning:
test/pp_filter_test.cc|466 col 3| warning: Potential leak of memory
pointed to by 'expected_output' [cplusplus.NewDeleteLeaks]
ASSERT_NE(expected_output, nullptr);
Bug: b/
229626362
Change-Id: Ie9e06c9b9442ffa134e514d2aee70841d19c8ecb
James Zern [Tue, 19 Jul 2022 23:30:59 +0000 (16:30 -0700)]
encode_api_test: quiet static analysis warning
in ConfigChangeThreadCount(); initialize cfg as the static analyzer can
assume AlwaysTrue() within EXPECT_NO_FATAL_FAILURE may return false
causing InitCodec() not to be called.
test/encode_api_test.cc|321 col 3| warning: 1st function call argument
is an uninitialized value [core.CallAndMessage]
video.SetSize(cfg.g_w, cfg.g_h);
Bug: b/
229626362
Change-Id: I54899ed0a207ca685416bed3a0e9c9644668e163
Scott LaVarnway [Tue, 12 Jul 2022 20:22:35 +0000 (13:22 -0700)]
VPX: Add vpx_quantize_b_32x32_avx2().
Up to 1.36x faster than vpx_quantize_b_32x32_avx() for full
calculations. Up to 1.29x faster for VP9_HIGHBITDEPTH builds.
Bug: b/
237714063
Change-Id: I97aa6a18d4dc2f3187b76800f91bbba7be447ef1
James Zern [Tue, 19 Jul 2022 02:00:49 +0000 (19:00 -0700)]
avg_intrin_{sse2,avg2}: rm dead store in hadamard_8x8
this quiets a couple static analysis warnings with clang 11:
vpx_dsp/x86/avg_intrin_sse2.c:278:45: warning: Although the value stored
to 'src_diff' is used in the enclosing expression, the value is never
actually read from 'src_diff' [deadcode.DeadStores]
src[7] = _mm_load_si128((const __m128i *)(src_diff += src_stride));
^ ~~~~~~~~~~
vpx_dsp/x86/avg_intrin_avx2.c:307:49: warning: Although the value stored
to 'src_diff' is used in the enclosing expression, the value is never
actually read from 'src_diff' [deadcode.DeadStores]
src[7] = _mm256_loadu_si256((const __m256i *)(src_diff += src_stride));
^ ~~~~~~~~~~
Bug: b/
229626362
Change-Id: I4b0201bd39775885df0afc03fa5da70910b9dad6
James Zern [Tue, 19 Jul 2022 01:56:03 +0000 (18:56 -0700)]
vpx_int_pro_row_c: add an assert for height
this quiets a static analysis warning with clang 11:
vpx_dsp/avg.c:353:15: warning: Assigned value is garbage or undefined
[core.uninitialized.Assign]
hbuf[idx] /= norm_factor;
^ ~~~~~~~~~~~
the same fix was applied in libaom:
1ad0889bc aom_int_pro_row_c: add an assert for height
Bug: b/
229626362
Change-Id: Ic8a249f866b33b02ec9f378581e51ac104d97169
Cheng Chen [Thu, 14 Jul 2022 18:41:11 +0000 (11:41 -0700)]
L2E: Update the description of allow_alt_ref
It is fixed per each encoding and can not be changed per GOP.
Change-Id: I5905b712437142f2274bfa674ceef6093495457f
James Zern [Thu, 14 Jul 2022 04:54:22 +0000 (21:54 -0700)]
vpxenc: fix --disable-loopfilter help alignment
Change-Id: I34444e6437ca0e735d6db07bf98bfa4741ad2c01
Konstantinos Margaritis [Wed, 13 Jul 2022 16:54:30 +0000 (16:54 +0000)]
Actually include the fix for commit
8f4d1890c.
Change-Id: I6780f610151f2e092da525ff064d4b69f74fa61b
Scott LaVarnway [Mon, 11 Jul 2022 21:48:55 +0000 (21:48 +0000)]
Merge "VPX: Add vpx_quantize_b_avx2()." into main
James Zern [Mon, 11 Jul 2022 20:23:28 +0000 (20:23 +0000)]
Merge "vp8_macros_msa.h: avoid shadowing variables in defines" into main
Scott LaVarnway [Wed, 6 Jul 2022 15:51:52 +0000 (08:51 -0700)]
VPX: Add vpx_quantize_b_avx2().
Up to 1.58x faster than vpx_quantize_b_avx() depending
on the size.
Bug: b/
237714063
Change-Id: I595a6bb32ebee63f69f27b5a15322fdeae1bf70e
James Zern [Mon, 11 Jul 2022 19:58:21 +0000 (19:58 +0000)]
Merge "Revert "Revert "[NEON] Optimize vp9_diamond_search_sad() for NEON""" into main
James Zern [Sat, 9 Jul 2022 21:49:54 +0000 (14:49 -0700)]
vp8_macros_msa.h: avoid shadowing variables in defines
this avoids a warning with certain versions of gcc; observed with:
mipsisa32r6el-linux-gnu-gcc (Debian 10.2.1-6) 10.2.1
20210110
Change-Id: I8999f487a79a9d53133816d572054b2423330bcf
Konstantinos Margaritis [Wed, 6 Jul 2022 08:42:54 +0000 (08:42 +0000)]
Revert "Revert "[NEON] Optimize vp9_diamond_search_sad() for NEON""
This reverts commit
9f1329f8ac88ea5d7c6ae5d6a57221c36cf85ac8
and fixes a dumb mistake in evaluation of vfcmv. Used vdupq_n_s16,
instead of vdupq_n_s32.
Change-Id: Ie236c878c166405c49bc0f93f6d63a6715534a0a
Scott LaVarnway [Thu, 7 Jul 2022 22:15:45 +0000 (22:15 +0000)]
Merge "VPX: Add quantize speed test for ref vs opt." into main
Scott LaVarnway [Thu, 7 Jul 2022 13:34:11 +0000 (06:34 -0700)]
VPX: Add quantize speed test for ref vs opt.
Bug: b/
237714063
Change-Id: I4304ba8d976fed3613e28442983b04a9cfc15b79
James Zern [Wed, 6 Jul 2022 22:06:51 +0000 (15:06 -0700)]
Revert "Fix bug with smaller width bigger size"
This reverts commit
5b530fc962bcb8a51bbf03f5fbc2912f21b86e70.
This fixes memory related fuzzer failures in the decoder.
Bug: webm:1642
Bug: oss-fuzz:48609
Bug: oss-fuzz:48629
Bug: oss-fuzz:48632
Bug: oss-fuzz:48638
Bug: oss-fuzz:48639
Bug: oss-fuzz:48651
Bug: oss-fuzz:48657
Bug: oss-fuzz:48659
Bug: oss-fuzz:48660
Bug: oss-fuzz:48661
Bug: oss-fuzz:48680
Bug: oss-fuzz:48686
Bug: oss-fuzz:48697
Bug: oss-fuzz:48706
Bug: oss-fuzz:48712
Bug: oss-fuzz:48717
Bug: oss-fuzz:48728
Bug: oss-fuzz:48732
Bug: oss-fuzz:48780
Bug: oss-fuzz:48781
Bug: oss-fuzz:48782
Bug: oss-fuzz:48785
Change-Id: I67a8539a3083f00eec1164fef5c6a8bc209f91fc
Jerome Jiang [Fri, 1 Jul 2022 14:37:32 +0000 (14:37 +0000)]
Merge "Fix bug with smaller width bigger size" into main
Jerome Jiang [Thu, 30 Jun 2022 17:48:56 +0000 (13:48 -0400)]
Fix bug with smaller width bigger size
Bug: webm:1642
Change-Id: I831b7701495eebeeff6bdc0b570f737bb6d536c6
Jerome Jiang [Thu, 30 Jun 2022 14:53:15 +0000 (10:53 -0400)]
ABI compatibility to CHANGELOG for prev releases.
Bug: webm:1757
Change-Id: I19576aa0bc065045dcb0eaf770ae5b0d9ac9d684
Marco Paniconi [Wed, 29 Jun 2022 17:35:36 +0000 (10:35 -0700)]
rtc: Add svc test for profile 2 10/12 bit
Add TODO to fix the superframe parser
for 10/12 bit.
Change-Id: Ib76c4daa0ff2f516510829ead6a397c89abba2f3
Marco Paniconi [Tue, 28 Jun 2022 18:22:48 +0000 (11:22 -0700)]
rtc-svc: Fix to make SVC work for Profile 1
Added datarate unittest for 4:4:4 and 4:2:2 input,
for spatial and temporal layers.
Fix is needed in vp9_set_literal_size():
the sampling_x/y should be passed into update_inital_width(),
othewise sampling_x/y = 1/1 (4:2:0) was forced.
vp9_set_literal_size() is only called by the svc and
on dynamic resize.
Fix issue with the normative optimized scaler:
UV width/height was assumed to be 1/2 of Y, for
the ssse and neon code.
Also fix to assert for the scaled width/height:
in case scaled width/height is odd it should be
incremented by 1 (make it even).
Change-Id: I3a2e40effa53c505f44ef05aaa3132e1b7f57dd5
Jerome Jiang [Tue, 28 Jun 2022 22:59:24 +0000 (22:59 +0000)]
Merge "Merge tag 'v1.12.0' into main" into main
Jerome Jiang [Tue, 28 Jun 2022 14:24:57 +0000 (10:24 -0400)]
Add vp8_ prefix for quantize_lsx.c
Duplicate name as vpx_dsp/loongarch/quantize_lsx.c
Chromium update script fails.
Bug: webm:1755
Change-Id: Ifb956c2292d909496eb2b9e1833993f1b021b07e
Jerome Jiang [Tue, 28 Jun 2022 21:07:31 +0000 (17:07 -0400)]
Merge tag 'v1.12.0' into main
Release v1.12.0 Torrent Duck
2022-06-17 v1.12.0 "Torrent Duck"
This release adds optimizations for Loongarch, adds support for vp8 in the
real-time rate control library, upgrades GoogleTest to v1.11.0, updates
libwebm to libwebm-1.0.0.28-20-g206d268, and includes numerous bug fixes.
- Upgrading:
This release is ABI compatible with the previous release.
vp8 support in the real-time rate control library.
New codec control VP8E_SET_RTC_EXTERNAL_RATECTRL is added.
Configure support for darwin21 is added.
GoogleTest is upgraded to v1.11.0.
libwebm is updated to libwebm-1.0.0.28-20-g206d268.
- Enhancement:
Numerous improvements on checking memory allocations.
Optimizations for Loongarch.
Code clean-up.
- Bug fixes:
Fix to a crash related to {vp8/vp9}_set_roi_map.
Fix to compiling failure with -Wformat-nonliteral.
Fix to integer overflow with vp9 with high resolution content.
Fix to AddNoiseTest failure with ARMv7.
Fix to libvpx Null-dereference READ in vp8.
Change-Id: I6964e96bccf016f977cc6e83dc0a192d66a19618
Jerome Jiang [Tue, 28 Jun 2022 18:38:49 +0000 (14:38 -0400)]
Replace date with version and release from README
CHANGELOG has the date.
Bug: webm:1752
Change-Id: I2888ce2afed8619f043eee1e9ca23bdf9d75e607
Cheng Chen [Fri, 24 Jun 2022 18:57:42 +0000 (11:57 -0700)]
L2E: Distinguish fixed and active gf_interval
min/max_gf_interval is fixed and can be passed from the command line.
It must satisfy the level constraints.
active_min/max_gf_interval might be changing based on
min/max_gf_interval. It is determined per GOP.
Change-Id: If456c691c97a8b4c946859c05cedd39ca7defa9c
James Zern [Fri, 17 Jun 2022 01:43:44 +0000 (18:43 -0700)]
vp9_encode_sb_row: remove a branch w/CONFIG_REALTIME_ONLY
replace the check on use_nonrd_pick_mode with an assert. this is only a
start, there are many branches that could be removed that check mode ==
REALTIME, etc. with this configuration.
Bug: webm:1773
Change-Id: I38cf9f83e7c085eb8e87d5cf6db7dc75359b611b
(cherry picked from commit
08b86d76224453ef9cbab4b10a48617715d9a14e)