[FFmpeg-devel] [PATCH] aarch64: increase default alignment for functions and constants

[FFmpeg-devel] [PATCH] aarch64: increase default alignment for functions and constants

[Ramiro Polla]

Use 16-byte alignment (align=4) instead of 4-byte (align=2) in the function and
const macros. This improves instruction fetch and NEON load performance on
modern AArch64 CPUs.
---
 libavutil/aarch64/asm.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavutil/aarch64/asm.S b/libavutil/aarch64/asm.S
index 50ce7d4dfd..2e4e451ec2 100644
--- a/libavutil/aarch64/asm.S
+++ b/libavutil/aarch64/asm.S
@@ -196,7 +196,7 @@ DISABLE_SVE2
         .popsection
 #endif
 
-.macro  function name, export=0, align=2
+.macro  function name, export=0, align=4
     .macro endfunc
 ELF     .size   \name, . - \name
 FUNC    .endfunc
@@ -217,7 +217,7 @@ FUNC    .func   \name
     .endif
 .endm
 
-.macro  const   name, align=2, relocate=0
+.macro  const   name, align=4, relocate=0
     .macro endconst
 ELF     .size   \name, . - \name
         .purgem endconst
-- 
2.39.5

_______________________________________________
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".

Re: [FFmpeg-devel] [PATCH] aarch64: increase default alignment for functions and constants

[Martin Storsjö]

On Fri, 16 May 2025, Ramiro Polla wrote:

> Use 16-byte alignment (align=4) instead of 4-byte (align=2) in the function and
> const macros. This improves instruction fetch and NEON load performance on
> modern AArch64 CPUs.
> ---
> libavutil/aarch64/asm.S | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)

Ok

// 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".

Re: [FFmpeg-devel] [PATCH] aarch64: increase default alignment for functions and constants

[Ramiro Polla]

On Fri, May 16, 2025 at 8:03 AM Martin Storsjö <martin@martin.st> wrote:
> On Fri, 16 May 2025, Ramiro Polla wrote:
> > Use 16-byte alignment (align=4) instead of 4-byte (align=2) in the function and
> > const macros. This improves instruction fetch and NEON load performance on
> > modern AArch64 CPUs.
> > ---
> > libavutil/aarch64/asm.S | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
>
> Ok

Thanks. Pushed.
_______________________________________________
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".