vpx_idct32x32_1024_add_ssse3() is actually a sse2 function and faster
than vpx_idct32x32_1024_add_sse2(). Replace the slow one. All are
code relocations, no new code.
Change-Id: I5dac0e98cc411a4ce05660406921118986638d19
It's almost identical with vpx_idct8x8_64_add_sse2(), except little
difference in instructions order.
Change-Id: Ie60dabc35eaa6ebae7c755e6cff00a710aad284f
Replace by CAST_TO_BYTEPTR/SHORTPTR.
The rule is: if a short ptr is casted to a byte ptr, any offset
operation on the byte ptr must be doubled. We do this by casting to
short ptr first, adding offset, then casting back to byte ptr.
BUG=webm:1388
Change-Id: I9e18a73ba45ddae58fc9dae470c0ff34951fe248
Provides over 15x speedup for width > 8.
Due to smaller loads and shifting for width == 8 it gets about 8x
speedup.
For width == 4 it's only about 4x speedup because there is a lot of
shuffling and shifting to get the data properly situated.
BUG=webm:1390
Change-Id: Ice0b3dbbf007be3d9509786a61e7f35e94bdffa8
For 8-bit the subtrahend is small enough to fit into uint32_t.
This is the same that was done for:
c0241664a Resolve -Wshorten-64-to-32 in variance.
For 10/12-bit apply:
63a37d16f Prevent negative variance
Change-Id: Iab35e3f3f269035e17c711bd6cc01272c3137e1d
- Refer to patch: 48fca113d inv_txfm_ssse3,butterfly: fix win32 abi
compatibility.
- Change four butterfly() calls to butterfly_self(), to simplify the
operations.
Change-Id: Ib2a8cfe6cddcaf0a59e6e6270d8380055ea42ef3
only the first 3 parameters can be aligned to 16 as required by __m128i,
make them all pointers for consistency.
since:
07c48ccfe Improve idct32x32_34_add SSSE3 intrinsics performance
BUG=webm:1384
Change-Id: I0324f701e723a27cb470036a180693ba8829d01d
- Split the inv txfm into three parts to avoid stack spillover.
- Function level speed improves ~12%.
- Use function and macro to remove some repeated code.
Change-Id: I14f5f072334fd766808cb52bf648df792e7379ee
- Split the transform into first half and second half.
- Reschedule the instructions to avoid stack spillover.
- Function level speed improves ~16%.
Change-Id: I166889840d23aa8a273eca00f6fbdae8b4566f35
- Replace the corresponding assembly code.
- No user level speed performance degrade.
- Unit tests passed.
Change-Id: Idd0c5a4bad4976f1617c34100cb46e75e3b961e5
The previous implementation confused bit/bytes/elements. It was using
'32' as the multiplier but that was mistakenly adopted because a 32x32
transform embedded the stride.
Change-Id: Ieeb867a332416b9a40580b5e7c9b20088e9e691a
This matches bitdepth_conversion_sse2.asm and produces substantially
better assembly. The old way had lots of 'movzwl' and 'shl' and storing
back to memory before loading into an xmm register.
Change-Id: Ib33e35354dfd691a4f8b1e39f4dbcbb14cd5302b
