Re: [FFmpeg-devel] [PATCH] lavc/pixblockdsp: specialise aligned 16-bit get_pixels - Rémi Denis-Courmont

From: "Rémi Denis-Courmont" <remi@remlab.net>
To: ffmpeg-devel@ffmpeg.org
Subject: Re: [FFmpeg-devel] [PATCH] lavc/pixblockdsp: specialise aligned 16-bit get_pixels
Date: Thu, 25 Jul 2024 23:28:44 +0300
Message-ID: <3537221.IQgDZ5Ah3G@basile.remlab.net> (raw)
In-Reply-To: <fa2e01b9-5db7-4641-9dd8-537d34c6bd3c@gmail.com>

Le torstaina 25. heinäkuuta 2024, 21.25.11 EEST James Almer a écrit :
> On 7/25/2024 1:50 PM, Rémi Denis-Courmont wrote:
> > Le torstaina 25. heinäkuuta 2024, 19.16.21 EEST James Almer a écrit :
> >> On 7/25/2024 12:53 PM, Rémi Denis-Courmont wrote:
> >>> The current code assumes that we have unaligned rows, which hurts on
> >>> platforms with slower unaligned accesses. (Also, this lets the compiler
> >>> unroll manually, which it seems to do in practice.)
> >>> ---
> >>> 
> >>>    libavcodec/pixblockdsp.c | 9 ++++++++-
> >>>    1 file changed, 8 insertions(+), 1 deletion(-)
> >>> 
> >>> diff --git a/libavcodec/pixblockdsp.c b/libavcodec/pixblockdsp.c
> >>> index bbbeca1618..1fff244511 100644
> >>> --- a/libavcodec/pixblockdsp.c
> >>> +++ b/libavcodec/pixblockdsp.c
> >>> @@ -26,6 +26,13 @@
> >>> 
> >>>    static void get_pixels_16_c(int16_t *restrict block, const uint8_t
> >>>    *pixels,
> >>>    
> >>>                                ptrdiff_t stride)
> >> 
> >> Is there a way to hint the compiler that block is 16 byte aligned? GCC
> >> 14 at least emits unaligned loads and stores for these.
> > 
> > We don't have uint128_t, so the best we could do is cast to uint64_t *.
> > Though GCC 13 emits 64-bit loads and stores on RV64 here with the given
> > code. Is this maybe a problem with the COPY128 macro definition on x86?
> 
> AV_COPY128 with GCC x86 uses aligned load intrinsics, but at least GCC
> 14 emits movdqu instructions here for some reason.

Another approach would be to define a structure with size and alignment of 16 
bytes, but I am none too sure that compilers will like it all that much. TBH, 
I am merely aiming for 64-bit aligned loads and stores here, which is a big 
improvement over 8-bit ones.

-- 
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".