* [FFmpeg-devel] [PATCH] libavcodec/hevc: reuse scale_store on idct32x32_neon
@ 2023-02-24 12:31 xufuji456
2023-02-28 11:56 ` Martin Storsjö
0 siblings, 1 reply; 2+ messages in thread
From: xufuji456 @ 2023-02-24 12:31 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: xufuji456
---
libavcodec/arm/hevcdsp_idct_neon.S | 18 ++----------------
1 file changed, 2 insertions(+), 16 deletions(-)
diff --git a/libavcodec/arm/hevcdsp_idct_neon.S b/libavcodec/arm/hevcdsp_idct_neon.S
index 75795e6a6a..41ca3b83a8 100644
--- a/libavcodec/arm/hevcdsp_idct_neon.S
+++ b/libavcodec/arm/hevcdsp_idct_neon.S
@@ -876,28 +876,14 @@ function func_tr_32x4_\name
movrel r9, trans + 32
vld1.s16 {q0}, [r9, :128]!
vld1.s16 {q1}, [r9, :128]
-
- bl tr_block1
-
add r4, sp, #2048
- vld1.s16 {q14-q15}, [r4, :128]!
- butterfly32 q14, q10, q15, q11
- scale32 d22, d23, d20, d21, q1, q14, q10, q15, \shift
-
- vld1.s16 {q14-q15}, [r4, :128]!
- butterfly32 q14, q12, q15, q13
- scale32 d2, d3, d28, d29, q1, q14, q12, q15, \shift
- transpose8_4x4 d22, d20, d2, d28
- transpose8_4x4 d29, d3, d21, d23
+ bl tr_block1
mov r1, r11
mov r2, #64
mov r8, #-64
add r3, r11, #(56 + 3 * 64)
- store16 d22, d23, d20, d21, d2, d3, d28, d29, r8
-
- @ reload multiplication coefficiens to q1
- vld1.s16 {q1}, [r9, :128]
+ scale_store \shift
bl tr_block2
add r1, r11, #8
--
2.32.0 (Apple Git-132)
_______________________________________________
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] 2+ messages in thread
* Re: [FFmpeg-devel] [PATCH] libavcodec/hevc: reuse scale_store on idct32x32_neon
2023-02-24 12:31 [FFmpeg-devel] [PATCH] libavcodec/hevc: reuse scale_store on idct32x32_neon xufuji456
@ 2023-02-28 11:56 ` Martin Storsjö
0 siblings, 0 replies; 2+ messages in thread
From: Martin Storsjö @ 2023-02-28 11:56 UTC (permalink / raw)
To: FFmpeg development discussions and patches; +Cc: xufuji456
On Fri, 24 Feb 2023, xufuji456 wrote:
> ---
> libavcodec/arm/hevcdsp_idct_neon.S | 18 ++----------------
> 1 file changed, 2 insertions(+), 16 deletions(-)
>
> diff --git a/libavcodec/arm/hevcdsp_idct_neon.S b/libavcodec/arm/hevcdsp_idct_neon.S
> index 75795e6a6a..41ca3b83a8 100644
> --- a/libavcodec/arm/hevcdsp_idct_neon.S
> +++ b/libavcodec/arm/hevcdsp_idct_neon.S
> @@ -876,28 +876,14 @@ function func_tr_32x4_\name
> movrel r9, trans + 32
> vld1.s16 {q0}, [r9, :128]!
> vld1.s16 {q1}, [r9, :128]
> -
> - bl tr_block1
> -
> add r4, sp, #2048
> - vld1.s16 {q14-q15}, [r4, :128]!
> - butterfly32 q14, q10, q15, q11
> - scale32 d22, d23, d20, d21, q1, q14, q10, q15, \shift
> -
> - vld1.s16 {q14-q15}, [r4, :128]!
> - butterfly32 q14, q12, q15, q13
> - scale32 d2, d3, d28, d29, q1, q14, q12, q15, \shift
>
> - transpose8_4x4 d22, d20, d2, d28
> - transpose8_4x4 d29, d3, d21, d23
> + bl tr_block1
> mov r1, r11
> mov r2, #64
> mov r8, #-64
> add r3, r11, #(56 + 3 * 64)
> - store16 d22, d23, d20, d21, d2, d3, d28, d29, r8
> -
> - @ reload multiplication coefficiens to q1
> - vld1.s16 {q1}, [r9, :128]
> + scale_store \shift
>
> bl tr_block2
> add r1, r11, #8
> --
> 2.32.0 (Apple Git-132)
LGTM, thanks - pushed now.
// Martin
_______________________________________________
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] 2+ messages in thread
end of thread, other threads:[~2023-02-28 11:56 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-02-24 12:31 [FFmpeg-devel] [PATCH] libavcodec/hevc: reuse scale_store on idct32x32_neon xufuji456
2023-02-28 11:56 ` Martin Storsjö
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