[FFmpeg-devel] [PATCH] avcodec/ac3: Remove unused fixed-point ARMv7 DSP

2024-04-17 Thread Geoff Hill
function table, but the ARMv7 implementations were kept. Signed-off-by: Geoff Hill --- libavcodec/arm/ac3dsp_neon.S | 63 1 file changed, 63 deletions(-) diff --git a/libavcodec/arm/ac3dsp_neon.S b/libavcodec/arm/ac3dsp_neon.S index 89d0ae8048..dc829541aa

[FFmpeg-devel] [PATCH v4 5/5] avcodec/ac3: Implement sum_square_butterfly_float for aarch64 NEON

2024-04-06 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 5 libavcodec/aarch64/ac3dsp_neon.S | 30 tests/checkasm/ac3dsp.c | 26 3 files changed, 61 insertions(+) diff --git a/libavcodec/aarch64

[FFmpeg-devel] [PATCH v4 4/5] avcodec/ac3: Implement sum_square_butterfly_int32 for aarch64 NEON

2024-04-06 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 5 + libavcodec/aarch64/ac3dsp_neon.S | 23 tests/checkasm/ac3dsp.c | 27 3 files changed, 55 insertions(+) diff --git a/libavcodec/aarch64

[FFmpeg-devel] [PATCH v4 3/5] avcodec/ac3: Implement ac3_extract_exponents for aarch64 NEON

2024-04-06 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 2 ++ libavcodec/aarch64/ac3dsp_neon.S | 14 + tests/checkasm/ac3dsp.c | 38 3 files changed, 54 insertions(+) diff --git a/libavcodec/aarch64/ac3dsp_init_aarch64

[FFmpeg-devel] [PATCH v4 2/5] avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON

2024-04-06 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 2 ++ libavcodec/aarch64/ac3dsp_neon.S | 16 + tests/checkasm/ac3dsp.c | 41 3 files changed, 59 insertions(+) diff --git a/libavcodec/aarch64/ac3dsp_init_aarch64

[FFmpeg-devel] [PATCH v4 1/5] avcodec/ac3: Implement float_to_fixed24 for aarch64 NEON

2024-04-06 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/Makefile | 2 ++ libavcodec/aarch64/ac3dsp_init_aarch64.c | 36 libavcodec/aarch64/ac3dsp_neon.S | 36 libavcodec/ac3dsp.c | 4 ++- libavcodec/ac3dsp.h

[FFmpeg-devel] [PATCH v4 0/5] avcodec/ac3: Add aarch64 NEON DSP

2024-04-06 Thread Geoff Hill
Geoff Hill (5): avcodec/ac3: Implement float_to_fixed24 for aarch64 NEON avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON avcodec/ac3: Implement ac3_extract_exponents for aarch64 NEON avcodec/ac3: Implement sum_square_butterfly_int32 for aarch64 NEON avcodec/ac3: Implement

[FFmpeg-devel] [PATCH v3 5/5] avcodec/ac3: Implement sum_square_butterfly_float for aarch64 NEON

2024-04-03 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 5 libavcodec/aarch64/ac3dsp_neon.S | 35 tests/checkasm/ac3dsp.c | 26 ++ 3 files changed, 66 insertions(+) diff --git a/libavcodec/aarch64

[FFmpeg-devel] [PATCH v3 4/5] avcodec/ac3: Implement sum_square_butterfly_int32 for aarch64 NEON

2024-04-03 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 5 + libavcodec/aarch64/ac3dsp_neon.S | 24 + tests/checkasm/ac3dsp.c | 27 3 files changed, 56 insertions(+) diff --git a/libavcodec/aarch64

[FFmpeg-devel] [PATCH v3 3/5] avcodec/ac3: Implement ac3_extract_exponents for aarch64 NEON

2024-04-03 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 2 ++ libavcodec/aarch64/ac3dsp_neon.S | 15 ++ tests/checkasm/ac3dsp.c | 36 3 files changed, 53 insertions(+) diff --git a/libavcodec/aarch64/ac3dsp_init_aarch64

[FFmpeg-devel] [PATCH v3 2/5] avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON

2024-04-03 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 2 ++ libavcodec/aarch64/ac3dsp_neon.S | 16 + tests/checkasm/ac3dsp.c | 41 3 files changed, 59 insertions(+) diff --git a/libavcodec/aarch64/ac3dsp_init_aarch64

[FFmpeg-devel] [PATCH v3 1/5] avcodec/ac3: Implement float_to_fixed24 for aarch64 NEON

2024-04-03 Thread Geoff Hill
Signed-off-by: Geoff Hill --- libavcodec/aarch64/Makefile | 2 ++ libavcodec/aarch64/ac3dsp_init_aarch64.c | 36 libavcodec/aarch64/ac3dsp_neon.S | 35 +++ libavcodec/ac3dsp.c | 4 ++- libavcodec/ac3dsp.h

[FFmpeg-devel] [PATCH v3 0/5] avcodec/ac3: Add aarch64 NEON DSP

2024-04-03 Thread Geoff Hill
: 516.2 Geoff Hill (5): avcodec/ac3: Implement float_to_fixed24 for aarch64 NEON avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON avcodec/ac3: Implement ac3_extract_exponents for aarch64 NEON avcodec/ac3: Implement sum_square_butterfly_int32 for aarch64 NEON avcodec/ac3: Implement

[FFmpeg-devel] [PATCH v2] avcodec/ac3: Implement float_to_fixed24 for aarch64 neon

2024-03-28 Thread Geoff Hill
[OK] checkasm: all 1 tests passed float_to_fixed24_c: 2450.7 float_to_fixed24_neon: 574.0 Signed-off-by: Geoff Hill --- libavcodec/aarch64/Makefile | 2 ++ libavcodec/aarch64/ac3dsp_init_aarch64.c | 36 +++ libavcodec/aarch64/ac3dsp_neon.S | 37

[FFmpeg-devel] [PATCH] avcodec/ac3: Implement float_to_fixed24 for aarch64 neon

2024-03-28 Thread Geoff Hill
Tested on AWS Graviton2 (t4g.medium), NixOS 23.11, GCC 12.3: float_to_fixed24_c: 2462.7 float_to_fixed24_neon: 513.5 Signed-off-by: Geoff Hill --- libavcodec/aarch64/Makefile | 2 ++ libavcodec/aarch64/ac3dsp_init_aarch64.c | 36 +++ libavcodec/aarch64