* [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext @ 2024-02-11 21:24 Kieran Kunhya 2024-02-11 21:36 ` James Almer 0 siblings, 1 reply; 7+ messages in thread From: Kieran Kunhya @ 2024-02-11 21:24 UTC (permalink / raw) To: FFmpeg development discussions and patches [-- Attachment #1: Type: text/plain, Size: 50 bytes --] $subj, now with forward declaration also removed. [-- Attachment #2: 0001-h264_intrapred-Remove-ff_pred16x16_horizontal_8_mmxe.patch --] [-- Type: application/octet-stream, Size: 1961 bytes --] [-- Attachment #3: Type: text/plain, Size: 251 bytes --] _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext 2024-02-11 21:24 [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext Kieran Kunhya @ 2024-02-11 21:36 ` James Almer 2024-02-11 22:24 ` Kieran Kunhya 2024-02-11 22:34 ` Andreas Rheinhardt 0 siblings, 2 replies; 7+ messages in thread From: James Almer @ 2024-02-11 21:36 UTC (permalink / raw) To: ffmpeg-devel [-- Attachment #1: Type: text/plain, Size: 267 bytes --] On 2/11/2024 6:24 PM, Kieran Kunhya wrote: > $subj, now with forward declaration also removed. This function is trivial to convert to SSE2, so better do that than removing it. Attached. If other functions are harder to port to SSE2, then sure, they can be removed. [-- Attachment #2: 0001-x86-h264_intrapred-convert-ff_pred16x16_horizontal_8.patch --] [-- Type: text/plain, Size: 2372 bytes --] From fdf1db323c1209fcb927f7df0d041d78f6214bf7 Mon Sep 17 00:00:00 2001 From: James Almer <jamrial@gmail.com> Date: Sun, 11 Feb 2024 18:35:31 -0300 Subject: [PATCH] x86/h264_intrapred: convert ff_pred16x16_horizontal_8_mmxext to sse2 Signed-off-by: James Almer <jamrial@gmail.com> --- libavcodec/x86/h264_intrapred.asm | 4 +--- libavcodec/x86/h264_intrapred_init.c | 4 ++-- 2 files changed, 3 insertions(+), 5 deletions(-) diff --git a/libavcodec/x86/h264_intrapred.asm b/libavcodec/x86/h264_intrapred.asm index 8a38ba2bb5..49a63f900e 100644 --- a/libavcodec/x86/h264_intrapred.asm +++ b/libavcodec/x86/h264_intrapred.asm @@ -86,8 +86,6 @@ cglobal pred16x16_horizontal_8, 2,3 punpcklbw m1, m1 SPLATW m0, m0, 3 SPLATW m1, m1, 3 - mova [r0+r1*0+8], m0 - mova [r0+r1*1+8], m1 %endif mova [r0+r1*0], m0 @@ -98,7 +96,7 @@ cglobal pred16x16_horizontal_8, 2,3 RET %endmacro -INIT_MMX mmxext +INIT_XMM sse2 PRED16x16_H INIT_XMM ssse3 PRED16x16_H diff --git a/libavcodec/x86/h264_intrapred_init.c b/libavcodec/x86/h264_intrapred_init.c index ee46927a24..06cb0ea8fe 100644 --- a/libavcodec/x86/h264_intrapred_init.c +++ b/libavcodec/x86/h264_intrapred_init.c @@ -100,7 +100,7 @@ PRED16x16(horizontal, 10, sse2) /* 8-bit versions */ PRED16x16(vertical, 8, sse) -PRED16x16(horizontal, 8, mmxext) +PRED16x16(horizontal, 8, sse2) PRED16x16(horizontal, 8, ssse3) PRED16x16(dc, 8, sse2) PRED16x16(dc, 8, ssse3) @@ -170,7 +170,6 @@ av_cold void ff_h264_pred_init_x86(H264PredContext *h, int codec_id, } if (EXTERNAL_MMXEXT(cpu_flags)) { - h->pred16x16[HOR_PRED8x8 ] = ff_pred16x16_horizontal_8_mmxext; if (chroma_format_idc <= 1) h->pred8x8[HOR_PRED8x8 ] = ff_pred8x8_horizontal_8_mmxext; h->pred8x8l [TOP_DC_PRED ] = ff_pred8x8l_top_dc_8_mmxext; @@ -210,6 +209,7 @@ av_cold void ff_h264_pred_init_x86(H264PredContext *h, int codec_id, } if (EXTERNAL_SSE2(cpu_flags)) { + h->pred16x16[HOR_PRED8x8 ] = ff_pred16x16_horizontal_8_sse2; h->pred16x16[DC_PRED8x8 ] = ff_pred16x16_dc_8_sse2; h->pred8x8l [DIAG_DOWN_LEFT_PRED ] = ff_pred8x8l_down_left_8_sse2; h->pred8x8l [DIAG_DOWN_RIGHT_PRED ] = ff_pred8x8l_down_right_8_sse2; -- 2.43.0 [-- Attachment #3: Type: text/plain, Size: 251 bytes --] _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext 2024-02-11 21:36 ` James Almer @ 2024-02-11 22:24 ` Kieran Kunhya 2024-02-11 22:34 ` Andreas Rheinhardt 1 sibling, 0 replies; 7+ messages in thread From: Kieran Kunhya @ 2024-02-11 22:24 UTC (permalink / raw) To: FFmpeg development discussions and patches On Sun, 11 Feb 2024 at 21:36, James Almer <jamrial@gmail.com> wrote: > On 2/11/2024 6:24 PM, Kieran Kunhya wrote: > > $subj, now with forward declaration also removed. > > This function is trivial to convert to SSE2, so better do that than > removing it. Attached. > If other functions are harder to port to SSE2, then sure, they can be > removed._______________________________________________ > > Thanks, applied. One down, couple dozen to go. Kieran _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext 2024-02-11 21:36 ` James Almer 2024-02-11 22:24 ` Kieran Kunhya @ 2024-02-11 22:34 ` Andreas Rheinhardt 2024-02-11 22:42 ` Kieran Kunhya 2024-02-11 22:47 ` James Almer 1 sibling, 2 replies; 7+ messages in thread From: Andreas Rheinhardt @ 2024-02-11 22:34 UTC (permalink / raw) To: ffmpeg-devel James Almer: > On 2/11/2024 6:24 PM, Kieran Kunhya wrote: >> $subj, now with forward declaration also removed. > > This function is trivial to convert to SSE2, so better do that than > removing it. Attached. > If other functions are harder to port to SSE2, then sure, they can be > removed. > Benchmarks? - Andreas _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext 2024-02-11 22:34 ` Andreas Rheinhardt @ 2024-02-11 22:42 ` Kieran Kunhya 2024-02-11 22:47 ` James Almer 1 sibling, 0 replies; 7+ messages in thread From: Kieran Kunhya @ 2024-02-11 22:42 UTC (permalink / raw) To: FFmpeg development discussions and patches On Sun, 11 Feb 2024 at 22:33, Andreas Rheinhardt < andreas.rheinhardt@outlook.com> wrote: > James Almer: > > On 2/11/2024 6:24 PM, Kieran Kunhya wrote: > >> $subj, now with forward declaration also removed. > > > > This function is trivial to convert to SSE2, so better do that than > > removing it. Attached. > > If other functions are harder to port to SSE2, then sure, they can be > > removed. > > > > Benchmarks? > > - Andreas > For me on Haswell x64: pred16x16_horizontal_8_c: 41.5 pred16x16_horizontal_8_sse2: 32.5 pred16x16_horizontal_8_ssse3: 9.0 pred16x16_horizontal_8_c: 43.5 pred16x16_horizontal_8_mmxext: 20.7 pred16x16_horizontal_8_ssse3: 12.2 Kieran _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext 2024-02-11 22:34 ` Andreas Rheinhardt 2024-02-11 22:42 ` Kieran Kunhya @ 2024-02-11 22:47 ` James Almer 1 sibling, 0 replies; 7+ messages in thread From: James Almer @ 2024-02-11 22:47 UTC (permalink / raw) To: ffmpeg-devel On 2/11/2024 7:34 PM, Andreas Rheinhardt wrote: > James Almer: >> On 2/11/2024 6:24 PM, Kieran Kunhya wrote: >>> $subj, now with forward declaration also removed. >> >> This function is trivial to convert to SSE2, so better do that than >> removing it. Attached. >> If other functions are harder to port to SSE2, then sure, they can be >> removed. >> > > Benchmarks? > > - Andreas On an Alder Lake i get pred16x16_horizontal_8_c: 26.5 pred16x16_horizontal_8_mmxext: 24.8 pred16x16_horizontal_8_sse2: 20.0 pred16x16_horizontal_8_ssse3: 12.5 _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
* [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext @ 2024-02-11 21:07 Kieran Kunhya 0 siblings, 0 replies; 7+ messages in thread From: Kieran Kunhya @ 2024-02-11 21:07 UTC (permalink / raw) To: FFmpeg development discussions and patches [-- Attachment #1: Type: text/plain, Size: 6 bytes --] $subj [-- Attachment #2: 0001-h264_intrapred-Remove-ff_pred16x16_horizontal_8_mmxe.patch --] [-- Type: application/octet-stream, Size: 1740 bytes --] [-- Attachment #3: Type: text/plain, Size: 251 bytes --] _______________________________________________ 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". ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-02-11 22:47 UTC | newest] Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2024-02-11 21:24 [FFmpeg-devel] [PATCH] h264_intrapred: Remove ff_pred16x16_horizontal_8_mmxext Kieran Kunhya 2024-02-11 21:36 ` James Almer 2024-02-11 22:24 ` Kieran Kunhya 2024-02-11 22:34 ` Andreas Rheinhardt 2024-02-11 22:42 ` Kieran Kunhya 2024-02-11 22:47 ` James Almer -- strict thread matches above, loose matches on Subject: below -- 2024-02-11 21:07 Kieran Kunhya
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel This inbox may be cloned and mirrored by anyone: git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \ ffmpegdev@gitmailbox.com public-inbox-index ffmpegdev Example config snippet for mirrors. AGPL code for this site: git clone https://public-inbox.org/public-inbox.git