On 19/04/2025 20:07, Mark Thompson wrote:
> Typical checkasm result on Alder Lake:
>
> decode_transquant_8_c: 408.7 ( 1.00x)
> decode_transquant_8_avx2: 94.2 ( 4.34x)
> decode_transquant_10_c: 413.1 ( 1.00x)
> decode_transquant_10_avx2: 87.5 ( 4.72x)
Interestingly this is signficantly better on an older Skylake. Typical result
there:
decode_transquant_8_c: 892.5 ( 1.00x)
decode_transquant_8_avx2: 143.4 ( 6.22x)
decode_transquant_10_c: 938.7 ( 1.00x)
decode_transquant_10_avx2: 137.9 ( 6.81x)
> ---
> libavcodec/apv_dsp.c | 4 +
> libavcodec/apv_dsp.h | 2 +
> libavcodec/x86/Makefile | 2 +
> libavcodec/x86/apv_dsp.asm | 243 ++++++++++++++++++++++++++++++++++
> libavcodec/x86/apv_dsp_init.c | 41 ++++++
> tests/checkasm/Makefile | 1 +
> tests/checkasm/apv_dsp.c | 113 ++++++++++++++++
> tests/checkasm/checkasm.c | 3 +
> tests/checkasm/checkasm.h | 1 +
> 9 files changed, 410 insertions(+)
> create mode 100644 libavcodec/x86/apv_dsp.asm
> create mode 100644 libavcodec/x86/apv_dsp_init.c
> create mode 100644 tests/checkasm/apv_dsp.c
>
> diff --git a/libavcodec/apv_dsp.c b/libavcodec/apv_dsp.c
> index 6b8b6ce590..108a3cd370 100644
> --- a/libavcodec/apv_dsp.c
> +++ b/libavcodec/apv_dsp.c
> @@ -133,4 +133,8 @@ static void apv_decode_transquant_c(void *output,
> av_cold void ff_apv_dsp_init(APVDSPContext *dsp)
> {
> dsp->decode_transquant = apv_decode_transquant_c;
> +
> +#if ARCH_X86_64
> + ff_apv_dsp_init_x86_64(dsp);
> +#endif
> }
> diff --git a/libavcodec/apv_dsp.h b/libavcodec/apv_dsp.h
> index 13f401c4cc..a7b5377357 100644
> --- a/libavcodec/apv_dsp.h
> +++ b/libavcodec/apv_dsp.h
> @@ -34,4 +34,6 @@ typedef struct APVDSPContext {
>
> void ff_apv_dsp_init(APVDSPContext *dsp);
>
> +void ff_apv_dsp_init_x86_64(APVDSPContext *dsp);
> +
> #endif /* AVCODEC_APV_DSP_H */
> diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
> index 5d53515381..821c410a0f 100644
> --- a/libavcodec/x86/Makefile
> +++ b/libavcodec/x86/Makefile
> @@ -44,6 +44,7 @@ OBJS-$(CONFIG_ADPCM_G722_DECODER) += x86/g722dsp_init.o
> OBJS-$(CONFIG_ADPCM_G722_ENCODER) += x86/g722dsp_init.o
> OBJS-$(CONFIG_ALAC_DECODER) += x86/alacdsp_init.o
> OBJS-$(CONFIG_APNG_DECODER) += x86/pngdsp_init.o
> +OBJS-$(CONFIG_APV_DECODER) += x86/apv_dsp_init.o
> OBJS-$(CONFIG_CAVS_DECODER) += x86/cavsdsp.o
> OBJS-$(CONFIG_CFHD_DECODER) += x86/cfhddsp_init.o
> OBJS-$(CONFIG_CFHD_ENCODER) += x86/cfhdencdsp_init.o
> @@ -149,6 +150,7 @@ X86ASM-OBJS-$(CONFIG_ADPCM_G722_DECODER) += x86/g722dsp.o
> X86ASM-OBJS-$(CONFIG_ADPCM_G722_ENCODER) += x86/g722dsp.o
> X86ASM-OBJS-$(CONFIG_ALAC_DECODER) += x86/alacdsp.o
> X86ASM-OBJS-$(CONFIG_APNG_DECODER) += x86/pngdsp.o
> +X86ASM-OBJS-$(CONFIG_APV_DECODER) += x86/apv_dsp.o
> X86ASM-OBJS-$(CONFIG_CAVS_DECODER) += x86/cavsidct.o
> X86ASM-OBJS-$(CONFIG_CFHD_ENCODER) += x86/cfhdencdsp.o
> X86ASM-OBJS-$(CONFIG_CFHD_DECODER) += x86/cfhddsp.o
> diff --git a/libavcodec/x86/apv_dsp.asm b/libavcodec/x86/apv_dsp.asm
> new file mode 100644
> index 0000000000..0329089f45
> --- /dev/null
> +++ b/libavcodec/x86/apv_dsp.asm
> @@ -0,0 +1,243 @@
> +;************************************************************************
> +;* This file is part of FFmpeg.
> +;*
> +;* FFmpeg is free software; you can redistribute it and/or
> +;* modify it under the terms of the GNU Lesser General Public
> +;* License as published by the Free Software Foundation; either
> +;* version 2.1 of the License, or (at your option) any later version.
> +;*
> +;* FFmpeg is distributed in the hope that it will be useful,
> +;* but WITHOUT ANY WARRANTY; without even the implied warranty of
> +;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> +;* Lesser General Public License for more details.
> +;*
> +;* You should have received a copy of the GNU Lesser General Public
> +;* License along with FFmpeg; if not, write to the Free Software
> +;* 51, Inc., Foundation Franklin Street, Fifth Floor, Boston, MA 02110-1301
> USA
> +;******************************************************************************
> +
> +%include "libavutil/x86/x86util.asm"
> +
> +SECTION .text
> +
> +align 32
> +const tmatrixh
> + dw 64, 89, 84, 75, 64, 50, 35, 18
> + dw 64, 75, 35, -18, -64, -89, -84, -50
> + dw 64, 50, -35, -89, -64, 18, 84, 75
> + dw 64, 18, -84, -50, 64, 75, -35, -89
> + dw 64, -18, -84, 50, 64, -75, -35, 89
> + dw 64, -50, -35, 89, -64, -18, 84, -75
> + dw 64, -75, 35, 18, -64, 89, -84, 50
> + dw 64, -89, 84, -75, 64, -50, 35, -18
> +const tmatrixv
> + dw 64, 89, 84, 75, 64, 50, 35, 18
> + dw 64, -18, -84, 50, 64, -75, -35, 89
> + dw 64, 75, 35, -18, -64, -89, -84, -50
> + dw 64, -50, -35, 89, -64, -18, 84, -75
> + dw 64, 50, -35, -89, -64, 18, 84, 75
> + dw 64, -75, 35, 18, -64, 89, -84, 50
> + dw 64, 18, -84, -50, 64, 75, -35, -89
> + dw 64, -89, 84, -75, 64, -50, 35, -18
> +
> +; Memory targets for vpbroadcastd (register version requires AVX512).
> +const one
> + dd 1
> +const sixtyfour
> + dd 64
> +
> +; void ff_apv_decode_transquant_avx2(void *output,
> +; ptrdiff_t pitch,
> +; const int16_t *input,
> +; const int16_t *qmatrix,
> +; int64_t bit_depth,
> +; int64_t qp_shift);
> +
> +INIT_YMM avx2
> +
> +cglobal apv_decode_transquant, 6, 6, 16, output, pitch, input, qmatrix,
> bit_depth, qp_shift
> +
> + ; Load input and dequantise
> +
> + lea rax, [bit_depthq - 2]
> + movq xm8, qp_shiftq
> + movq xm9, rax
> + vpbroadcastd m10, [one]
> + vpslld m10, m10, xm9
> + vpsrld m10, m10, 1
> +
> + ; m8 = scalar qp_shift
> + ; m9 = scalar bd_shift
> + ; m10 = vector 1 << (bd_shift - 1)
> + ; m11 = qmatrix load
> +%macro LOAD_AND_DEQUANT 2 ; (xmm input, constant offset)
> + vpmovsxwd m%1, [inputq + %2]
> + vpmovsxwd m11, [qmatrixq + %2]
> + vpmulld m%1, m%1, m11
> + vpslld m%1, m%1, xm8
> + vpaddd m%1, m%1, m10
> + vpsrad m%1, m%1, xm9
> + vpackssdw m%1, m%1, m%1
> +%endmacro
> +
> + LOAD_AND_DEQUANT 0, 0x00
> + LOAD_AND_DEQUANT 1, 0x10
> + LOAD_AND_DEQUANT 2, 0x20
> + LOAD_AND_DEQUANT 3, 0x30
> + LOAD_AND_DEQUANT 4, 0x40
> + LOAD_AND_DEQUANT 5, 0x50
> + LOAD_AND_DEQUANT 6, 0x60
> + LOAD_AND_DEQUANT 7, 0x70
> +
> + ; Transform columns
> + ; This produces the output in rows to avoid needing any
> + ; transpose operations
> +
> + vpunpcklwd m12, m0, m1
> + vpunpcklwd m13, m2, m3
> + vpunpcklwd m14, m4, m5
> + vpunpcklwd m15, m6, m7
> +
> +%macro TRANS_COL_STEP 2
> + vpbroadcastd m8, [tmatrixh + %2 + 0x00]
> + vpbroadcastd m9, [tmatrixh + %2 + 0x04]
> + vpbroadcastd m10, [tmatrixh + %2 + 0x08]
> + vpbroadcastd m11, [tmatrixh + %2 + 0x0c]
> + vpmaddwd m8, m8, m12
> + vpmaddwd m9, m9, m13
> + vpmaddwd m10, m10, m14
> + vpmaddwd m11, m11, m15
> + vpaddd m8, m8, m9
> + vpaddd m10, m10, m11
> + vpaddd m%1, m8, m10
> +%endmacro
> +
> + TRANS_COL_STEP 0, 0x00
> + TRANS_COL_STEP 1, 0x10
> + TRANS_COL_STEP 2, 0x20
> + TRANS_COL_STEP 3, 0x30
> + TRANS_COL_STEP 4, 0x40
> + TRANS_COL_STEP 5, 0x50
> + TRANS_COL_STEP 6, 0x60
> + TRANS_COL_STEP 7, 0x70
> +
> + ; Mid-transform normalisation
> + ; Note that outputs here are fitted to 16 bits
> +
> + vpbroadcastd m8, [sixtyfour]
> +
> +%macro NORMALISE 1
> + vpaddd m%1, m%1, m8
> + vpsrad m%1, m%1, 7
> + vpackssdw m%1, m%1, m%1
> + vpermq m%1, m%1, q3120
> +%endmacro
> +
> + NORMALISE 0
> + NORMALISE 1
> + NORMALISE 2
> + NORMALISE 3
> + NORMALISE 4
> + NORMALISE 5
> + NORMALISE 6
> + NORMALISE 7
> +
> + ; Vertical transform
> +
> + mova m12, [tmatrixv + 0x00]
> + mova m13, [tmatrixv + 0x20]
> + mova m14, [tmatrixv + 0x40]
> + mova m15, [tmatrixv + 0x60]
> +
> +%macro TRANS_ROW_STEP 1
> + vperm2i128 m%1, m%1, m%1, 1
> + vpmaddwd m8, m%1, m12
> + vpmaddwd m9, m%1, m13
> + vpmaddwd m10, m%1, m14
> + vpmaddwd m11, m%1, m15
> + vphaddd m8, m8, m9
> + vphaddd m10, m10, m11
> + vphaddd m%1, m8, m10
> +%endmacro
> +
> + TRANS_ROW_STEP 0
> + TRANS_ROW_STEP 1
> + TRANS_ROW_STEP 2
> + TRANS_ROW_STEP 3
> + TRANS_ROW_STEP 4
> + TRANS_ROW_STEP 5
> + TRANS_ROW_STEP 6
> + TRANS_ROW_STEP 7
> +
> + ; Renormalise, clip and store output
> +
> + vpbroadcastd m14, [one]
> + mov rax, 20
> + sub rax, bit_depthq ; bd_shift
> + movq xm9, rax
> + dec rax
> + movq xm13, rax
> + vpslld m8, m14, xm13
> + movq xm13, bit_depthq
> + vpslld m12, m14, xm13
> + vpsrld m10, m12, 1
> + vpsubd m12, m12, m14
> + vpxor m11, m11, m11
> +
> + cmp bit_depthq, 8
> + jne store_10
> +
> + ; m8 = vector 1 << (bd_shift - 1)
> + ; m9 = scalar bd_shift
> + ; m10 = vector 1 << (bit_depth - 1)
> + ; m11 = zero
> + ; m12 = vector (1 << bit_depth) - 1
> +
> +%macro NORMALISE_AND_STORE_8 1
> + vpaddd m%1, m%1, m8
> + vpsrad m%1, m%1, xm9
> + vpaddd m%1, m%1, m10
> + vextracti128 xm13, m%1, 0
> + vextracti128 xm14, m%1, 1
> + vpackusdw xm%1, xm13, xm14
> + vpackuswb xm%1, xm%1, xm%1
> + movq [outputq], xm%1
> + add outputq, pitchq
> +%endmacro
> +
> + NORMALISE_AND_STORE_8 0
> + NORMALISE_AND_STORE_8 1
> + NORMALISE_AND_STORE_8 2
> + NORMALISE_AND_STORE_8 3
> + NORMALISE_AND_STORE_8 4
> + NORMALISE_AND_STORE_8 5
> + NORMALISE_AND_STORE_8 6
> + NORMALISE_AND_STORE_8 7
> +
> + RET
> +
> +store_10:
> +
> +%macro NORMALISE_AND_STORE_10 1
> + vpaddd m%1, m%1, m8
> + vpsrad m%1, m%1, xm9
> + vpaddd m%1, m%1, m10
> + vpmaxsd m%1, m%1, m11
> + vpminsd m%1, m%1, m12
> + vextracti128 xm13, m%1, 0
> + vextracti128 xm14, m%1, 1
> + vpackusdw xm%1, xm13, xm14
> + mova [outputq], xm%1
> + add outputq, pitchq
> +%endmacro
> +
> + NORMALISE_AND_STORE_10 0
> + NORMALISE_AND_STORE_10 1
> + NORMALISE_AND_STORE_10 2
> + NORMALISE_AND_STORE_10 3
> + NORMALISE_AND_STORE_10 4
> + NORMALISE_AND_STORE_10 5
> + NORMALISE_AND_STORE_10 6
> + NORMALISE_AND_STORE_10 7
> +
> + RET
> diff --git a/libavcodec/x86/apv_dsp_init.c b/libavcodec/x86/apv_dsp_init.c
> new file mode 100644
> index 0000000000..f7f209cea8
> --- /dev/null
> +++ b/libavcodec/x86/apv_dsp_init.c
> @@ -0,0 +1,41 @@
> +/*
> + * This file is part of FFmpeg.
> + *
> + * FFmpeg is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation; either
> + * version 2.1 of the License, or (at your option) any later version.
> + *
> + * FFmpeg is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with FFmpeg; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301
> USA
> + */
> +
> +#include "config.h"
> +#include "libavutil/attributes.h"
> +#include "libavutil/cpu.h"
> +#include "libavutil/x86/asm.h"
> +#include "libavutil/x86/cpu.h"
> +#include "libavcodec/apv_dsp.h"
> +
> +void ff_apv_decode_transquant_avx2(void *output,
> + ptrdiff_t pitch,
> + const int16_t *input,
> + const int16_t *qmatrix,
> + int64_t bit_depth,
> + int64_t qp_shift);
> +
> +
> +av_cold void ff_apv_dsp_init_x86_64(APVDSPContext *dsp)
> +{
> + int cpu_flags = av_get_cpu_flags();
> +
> + if (EXTERNAL_AVX2_FAST(cpu_flags)) {
> + dsp->decode_transquant = ff_apv_decode_transquant_avx2;
> + }
> +}
> diff --git a/tests/checkasm/Makefile b/tests/checkasm/Makefile
> index d5c50e5599..193c1e4633 100644
> --- a/tests/checkasm/Makefile
> +++ b/tests/checkasm/Makefile
> @@ -28,6 +28,7 @@ AVCODECOBJS-$(CONFIG_AAC_DECODER) += aacpsdsp.o \
> sbrdsp.o
> AVCODECOBJS-$(CONFIG_AAC_ENCODER) += aacencdsp.o
> AVCODECOBJS-$(CONFIG_ALAC_DECODER) += alacdsp.o
> +AVCODECOBJS-$(CONFIG_APV_DECODER) += apv_dsp.o
> AVCODECOBJS-$(CONFIG_DCA_DECODER) += synth_filter.o
> AVCODECOBJS-$(CONFIG_DIRAC_DECODER) += diracdsp.o
> AVCODECOBJS-$(CONFIG_EXR_DECODER) += exrdsp.o
> diff --git a/tests/checkasm/apv_dsp.c b/tests/checkasm/apv_dsp.c
> new file mode 100644
> index 0000000000..a0272d8edc
> --- /dev/null
> +++ b/tests/checkasm/apv_dsp.c
> @@ -0,0 +1,113 @@
> +/*
> + * This file is part of FFmpeg.
> + *
> + * FFmpeg is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU Lesser General Public
> + * License as published by the Free Software Foundation; either
> + * version 2.1 of the License, or (at your option) any later version.
> + *
> + * FFmpeg is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> + * Lesser General Public License for more details.
> + *
> + * You should have received a copy of the GNU Lesser General Public
> + * License along with FFmpeg; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301
> USA
> + */
> +
> +#include <stdint.h>
> +
> +#include "checkasm.h"
> +
> +#include "libavutil/attributes.h"
> +#include "libavutil/mem_internal.h"
> +#include "libavcodec/apv_dsp.h"
> +
> +
> +static void check_decode_transquant_8(void)
> +{
> + LOCAL_ALIGNED_16(int16_t, input, [64]);
> + LOCAL_ALIGNED_16(int16_t, qmatrix, [64]);
> + LOCAL_ALIGNED_16(uint8_t, new_output, [64]);
> + LOCAL_ALIGNED_16(uint8_t, ref_output, [64]);
> +
> + declare_func(void,
> + uint8_t *output,
> + ptrdiff_t pitch,
> + const int16_t *input,
> + const int16_t *qmatrix,
> + int64_t bit_depth,
> + int64_t qp_shift);
> +
> + for (int i = 0; i < 64; i++) {
> + // Any signed 12-bit integer.
> + input[i] = rnd() % 2048 - 1024;
> +
> + // qmatrix input is premultiplied by level_scale, so
> + // range is 1 to 255 * 71. Interesting values are all
> + // at the low end of that, though.
> + qmatrix[i] = rnd() % 16 + 16;
> + }
> +
> + call_ref(ref_output, 8, input, qmatrix, 8, 4);
> + call_new(new_output, 8, input, qmatrix, 8, 4);
> +
> + for (int i = 0; i < 64; i++) {
> + if (ref_output[i] != new_output[i])
> + fail();
> + }
> +
> + bench_new(new_output, 8, input, qmatrix, 8, 4);
> +}
> +
> +static void check_decode_transquant_10(void)
> +{
> + LOCAL_ALIGNED_16( int16_t, input, [64]);
> + LOCAL_ALIGNED_16( int16_t, qmatrix, [64]);
> + LOCAL_ALIGNED_16(uint16_t, new_output, [64]);
> + LOCAL_ALIGNED_16(uint16_t, ref_output, [64]);
> +
> + declare_func(void,
> + uint16_t *output,
> + ptrdiff_t pitch,
> + const int16_t *input,
> + const int16_t *qmatrix,
> + int64_t bit_depth,
> + int64_t qp_shift);
> +
> + for (int i = 0; i < 64; i++) {
> + // Any signed 14-bit integer.
> + input[i] = rnd() % 16384 - 8192;
> +
> + // qmatrix input is premultiplied by level_scale, so
> + // range is 1 to 255 * 71. Interesting values are all
> + // at the low end of that, though.
> + qmatrix[i] = 16; //rnd() % 16 + 16;
> + }
> +
> + call_ref(ref_output, 16, input, qmatrix, 10, 4);
> + call_new(new_output, 16, input, qmatrix, 10, 4);
> +
> + for (int i = 0; i < 64; i++) {
> + if (ref_output[i] != new_output[i])
> + fail();
> + }
> +
> + bench_new(new_output, 16, input, qmatrix, 10, 4);
> +}
> +
> +void checkasm_check_apv_dsp(void)
> +{
> + APVDSPContext dsp;
> +
> + ff_apv_dsp_init(&dsp);
> +
> + if (check_func(dsp.decode_transquant, "decode_transquant_8"))
> + check_decode_transquant_8();
> +
> + if (check_func(dsp.decode_transquant, "decode_transquant_10"))
> + check_decode_transquant_10();
> +
> + report("apv_dsp");
> +}
> diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
> index 412b8b2cd1..3bb82ed0e5 100644
> --- a/tests/checkasm/checkasm.c
> +++ b/tests/checkasm/checkasm.c
> @@ -129,6 +129,9 @@ static const struct {
> #if CONFIG_ALAC_DECODER
> { "alacdsp", checkasm_check_alacdsp },
> #endif
> + #if CONFIG_APV_DECODER
> + { "apv_dsp", checkasm_check_apv_dsp },
> + #endif
> #if CONFIG_AUDIODSP
> { "audiodsp", checkasm_check_audiodsp },
> #endif
> diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
> index ad239fb2a4..a6b5965e02 100644
> --- a/tests/checkasm/checkasm.h
> +++ b/tests/checkasm/checkasm.h
> @@ -83,6 +83,7 @@ void checkasm_check_ac3dsp(void);
> void checkasm_check_aes(void);
> void checkasm_check_afir(void);
> void checkasm_check_alacdsp(void);
> +void checkasm_check_apv_dsp(void);
> void checkasm_check_audiodsp(void);
> void checkasm_check_av_tx(void);
> void checkasm_check_blend(void);
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
