x/ffmpeg
1
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2026-06-11 08:13:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

avutil/x86/tx_float: add missing vzeroupper to 15xM PFA FFT

Browse Source 
The AVX2 15xM PFA FFT calls its second-dimension subtransform with dirty
YMM. That subtransform may be a legacy-SSE codelet (fft4 is SSE2 only),
causing AVX<->SSE transition penalties. Clear them after the first
dimension, before the calls.

Detected with `sde64 -ast` FATE job.

Fixes: ace42cf581
This commit is contained in:
Kacper Michajłow
2026-06-09 17:54:21 +00:00
parent 4406f5ba5b
commit f43f609bb8
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libavutil/x86/tx_float.asm
+2
View File
@@ -1874,6 +1874,8 @@ cglobal fft_pfa_15xM_float, 4, 14, 16, 320, ctx, out, in, stride, len, lut, buf,
mov lutq, [ctxq + AVTXContext.map] ; load subtransform's map mov lutq, [ctxq + AVTXContext.map] ; load subtransform's map
movsxd lenq, dword [ctxq + AVTXContext.len] ; load subtransform's length movsxd lenq, dword [ctxq + AVTXContext.len] ; load subtransform's length
vzeroupper
.dim2: .dim2:
call tgt5q ; call the FFT call tgt5q ; call the FFT
lea inq, [inq + lenq*8] lea inq, [inq + lenq*8]