wuggy/Dactyloidae
wuggy/Dactyloidae
1
0
Fork 0
mirror of https://repo.dactyloidae.xyz/Dactyloidae/UXP.git synced 2026-09-28 11:27:32 +09:00
Code Issues Projects Releases Packages Wiki Activity

Issue #1890 - Update libopus

Browse source 
This resolves #1890
This commit is contained in:
Moonchild 2022-05-04 17:03:37 +00:00 • committed by roytam1
commit a8b1099a1e
228 changed files with 10990 additions and 5670 deletions
Download patch file Download diff file Expand all files Collapse all files

32
media/libopus/celt/x86/x86cpu.h
View file

@ -56,38 +56,10 @@
int opus_select_arch(void);
# endif
/*gcc appears to emit MOVDQA's to load the argument of an _mm_cvtepi8_epi32()
or _mm_cvtepi16_epi32() when optimizations are disabled, even though the
actual PMOVSXWD instruction takes an m32 or m64. Unlike a normal memory
reference, these require 16-byte alignment and load a full 16 bytes (instead
of 4 or 8), possibly reading out of bounds.
We can insert an explicit MOVD or MOVQ using _mm_cvtsi32_si128() or
_mm_loadl_epi64(), which should have the same semantics as an m32 or m64
reference in the PMOVSXWD instruction itself, but gcc is not smart enough to
optimize this out when optimizations ARE enabled.
Clang, in contrast, requires us to do this always for _mm_cvtepi8_epi32
(which is fair, since technically the compiler is always allowed to do the
dereference before invoking the function implementing the intrinsic).
However, it is smart enough to eliminate the extra MOVD instruction.
For _mm_cvtepi16_epi32, it does the right thing, though does *not* optimize out
the extra MOVQ if it's specified explicitly */
# if defined(__clang__) || !defined(__OPTIMIZE__)
# define OP_CVTEPI8_EPI32_M32(x) \
#define OP_CVTEPI8_EPI32_M32(x) \
(_mm_cvtepi8_epi32(_mm_cvtsi32_si128(*(int *)(x))))
# else
# define OP_CVTEPI8_EPI32_M32(x) \
(_mm_cvtepi8_epi32(*(__m128i *)(x)))
#endif
# if !defined(__OPTIMIZE__)
# define OP_CVTEPI16_EPI32_M64(x) \
#define OP_CVTEPI16_EPI32_M64(x) \
(_mm_cvtepi16_epi32(_mm_loadl_epi64((__m128i *)(x))))
# else
# define OP_CVTEPI16_EPI32_M64(x) \
(_mm_cvtepi16_epi32(*(__m128i *)(x)))
# endif
#endif