x86: fft: fix imdct_half() for AVX
authorRonald S. Bultje <rsbultje@gmail.com>
Thu, 2 Aug 2012 17:15:46 +0000 (12:15 -0500)
committerJustin Ruggles <justin.ruggles@gmail.com>
Thu, 2 Aug 2012 17:40:11 +0000 (13:40 -0400)
commitc728518b3cbb0daf0d0a65ba7adfcb48c5629b93
tree0042b25b0462735fb301e4d679112e50a619accd
parent150adea6da24e5342584b403dc0f901aa79f53c8
x86: fft: fix imdct_half() for AVX

Some calculations were changed in b6a3849 to use mmsize, which was not correct
for the AVX version, which uses INIT_YMM and therefore has mmsize == 32.

Fixes Bug 341.

Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
libavcodec/x86/fft_mmx.asm