Message ID | CAEa-L+tcAyGUmeRC_F8tQ0k5-noR7JYGRdi=rTmgET3LyvR+qw@mail.gmail.com |
---|---|
State | New |
Headers | show |
Series | [FFmpeg-devel,1/4] lavc/vp9dsp: R-V V ipred dc | expand |
Context | Check | Description |
---|---|---|
andriy/configure_x86 | warning | Failed to apply patch |
yinshiyou/configure_loongarch64 | warning | Failed to apply patch |
Due to the PATCH 1/4 update, updates here. flow gg <hlefthleft@gmail.com> 于2024年3月2日周六 15:42写道: > >
Because the previous patch was updated, so it was updated in this response flow gg <hlefthleft@gmail.com> 于2024年3月3日周日 10:01写道: > Due to the PATCH 1/4 update, updates here. > > flow gg <hlefthleft@gmail.com> 于2024年3月2日周六 15:42写道: > >> >>
Le perjantaina 22. maaliskuuta 2024, 8.02.38 EEST flow gg a écrit :
> Because the previous patch was updated, so it was updated in this response
Seemingly needs rebase since April 7.
updated it in the reply and https://github.com/hleft/FFmpeg/tree/vp8vp9 Rémi Denis-Courmont <remi@remlab.net> 于2024年4月30日周二 01:57写道: > Le perjantaina 22. maaliskuuta 2024, 8.02.38 EEST flow gg a écrit : > > Because the previous patch was updated, so it was updated in this > response > > Seemingly needs rebase since April 7. > > -- > レミ・デニ-クールモン > http://www.remlab.net/ > > > > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
Le tiistaina 30. huhtikuuta 2024, 2.36.22 EEST flow gg a écrit : > updated it in the reply and https://github.com/hleft/FFmpeg/tree/vp8vp9 VP9 checkasm does not pass on that branch. > Rémi Denis-Courmont <remi@remlab.net> 于2024年4月30日周二 01:57写道: > > > Le perjantaina 22. maaliskuuta 2024, 8.02.38 EEST flow gg a écrit : > > > Because the previous patch was updated, so it was updated in this > > > > response > > > > Seemingly needs rebase since April 7. > > > > -- > > レミ・デニ-クールモン > > http://www.remlab.net/ > > > > > > > > _______________________________________________ > > ffmpeg-devel mailing list > > ffmpeg-devel@ffmpeg.org > > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > > > To unsubscribe, visit link above, or email > > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
Sorry, this is because a 'bpp == 8' was missed. It has been fixed in this link Rémi Denis-Courmont <remi@remlab.net> 于2024年5月2日周四 22:11写道: > Le tiistaina 30. huhtikuuta 2024, 2.36.22 EEST flow gg a écrit : > > updated it in the reply and https://github.com/hleft/FFmpeg/tree/vp8vp9 > > VP9 checkasm does not pass on that branch. > > > Rémi Denis-Courmont <remi@remlab.net> 于2024年4月30日周二 01:57写道: > > > > > Le perjantaina 22. maaliskuuta 2024, 8.02.38 EEST flow gg a écrit : > > > > Because the previous patch was updated, so it was updated in this > > > > > > response > > > > > > Seemingly needs rebase since April 7. > > > > > > -- > > > レミ・デニ-クールモン > > > http://www.remlab.net/ > > > > > > > > > > > > _______________________________________________ > > > ffmpeg-devel mailing list > > > ffmpeg-devel@ffmpeg.org > > > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > > > > > To unsubscribe, visit link above, or email > > > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". > > > -- > レミ・デニ-クールモン > http://www.remlab.net/ > > > > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >
From 7abd262daa281cee412a905ea75a5f10dd0b1fbe Mon Sep 17 00:00:00 2001 From: sunyuechi <sunyuechi@iscas.ac.cn> Date: Fri, 1 Mar 2024 18:38:43 +0800 Subject: [PATCH 2/4] lavc/vp9dsp: R-V V ipred vert C908: vp9_vert_8x8_8bpp_c: 22.0 vp9_vert_8x8_8bpp_rvv_i64: 18.5 vp9_vert_16x16_8bpp_c: 71.2 vp9_vert_16x16_8bpp_rvv_i32: 50.7 vp9_vert_32x32_8bpp_c: 300.2 vp9_vert_32x32_8bpp_rvv_i32: 136.7 --- libavcodec/riscv/vp9_intra_rvv.S | 35 ++++++++++++++++++++++++++++++++ libavcodec/riscv/vp9dsp.h | 6 ++++++ libavcodec/riscv/vp9dsp_init.c | 3 +++ 3 files changed, 44 insertions(+) diff --git a/libavcodec/riscv/vp9_intra_rvv.S b/libavcodec/riscv/vp9_intra_rvv.S index b3b0470cfc..88b54f37b0 100644 --- a/libavcodec/riscv/vp9_intra_rvv.S +++ b/libavcodec/riscv/vp9_intra_rvv.S @@ -199,3 +199,38 @@ endfunc func ff_dc_top_8x8_rvv, zve64x dc8x8 top endfunc + +func ff_v_32x32_rvv, zve32x + vsetivli zero, 8, e8, mf2, ta, ma + vle32.v v8, (a3) + + .rept 31 + vse32.v v8, (a0) + add a0, a0, a1 + .endr + vse32.v v8, (a0) + + ret +endfunc + +func ff_v_16x16_rvv, zve32x + vsetivli zero, 4, e8, mf4, ta, ma + vle32.v v8, (a3) + + .rept 15 + vse32.v v8, (a0) + add a0, a0, a1 + .endr + vse32.v v8, (a0) + + ret +endfunc + +func ff_v_8x8_rvv, zve64x + ld t0, (a3) + vsetivli zero, 8, e64, m4, ta, ma + vmv.v.x v8, t0 + vsse64.v v8, (a0), a1 + + ret +endfunc diff --git a/libavcodec/riscv/vp9dsp.h b/libavcodec/riscv/vp9dsp.h index abd57bd836..ae4fb266d0 100644 --- a/libavcodec/riscv/vp9dsp.h +++ b/libavcodec/riscv/vp9dsp.h @@ -60,5 +60,11 @@ void ff_dc_129_16x16_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, const uint8_t *a); void ff_dc_129_8x8_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, const uint8_t *a); +void ff_v_32x32_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, + const uint8_t *a); +void ff_v_16x16_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, + const uint8_t *a); +void ff_v_8x8_rvv(uint8_t *dst, ptrdiff_t stride, const uint8_t *l, + const uint8_t *a); #endif // #ifndef AVCODEC_RISCV_VP9DSP_RISCV_H diff --git a/libavcodec/riscv/vp9dsp_init.c b/libavcodec/riscv/vp9dsp_init.c index 69ab39004c..9c550d40b5 100644 --- a/libavcodec/riscv/vp9dsp_init.c +++ b/libavcodec/riscv/vp9dsp_init.c @@ -36,6 +36,7 @@ static av_cold void vp9dsp_intrapred_init_rvv(VP9DSPContext *dsp, int bpp) dsp->intra_pred[TX_8X8][DC_128_PRED] = ff_dc_128_8x8_rvv; dsp->intra_pred[TX_8X8][DC_129_PRED] = ff_dc_129_8x8_rvv; dsp->intra_pred[TX_8X8][TOP_DC_PRED] = ff_dc_top_8x8_rvv; + dsp->intra_pred[TX_8X8][VERT_PRED] = ff_v_8x8_rvv; } if (bpp == 8 && flags & AV_CPU_FLAG_RVV_I32 && ff_get_rv_vlenb() >= 16) { @@ -51,6 +52,8 @@ static av_cold void vp9dsp_intrapred_init_rvv(VP9DSPContext *dsp, int bpp) dsp->intra_pred[TX_16X16][DC_129_PRED] = ff_dc_129_16x16_rvv; dsp->intra_pred[TX_32X32][TOP_DC_PRED] = ff_dc_top_32x32_rvv; dsp->intra_pred[TX_16X16][TOP_DC_PRED] = ff_dc_top_16x16_rvv; + dsp->intra_pred[TX_32X32][VERT_PRED] = ff_v_32x32_rvv; + dsp->intra_pred[TX_16X16][VERT_PRED] = ff_v_16x16_rvv; } #endif } -- 2.44.0