Approximates division using multiply and shift.
Speeds up both sizes (8x8 and 16x16) by 30 times.
Fix the call sites to use the RTCD function.
Delete sse2 and mips implementation. They were based on a previous
implementation of the filter. It was changed in Dec 2015:
ece4fd5d22
BUG=webm:1378
Change-Id: I0818e767a802966520b5c6e7999584ad13159276
The scaling filter with zero shift will give sub-sampling for
2x downsampling. Allow for a phase shift to get an averaging filter.
Usage is for source scaling in 1 pass SVC mode for 1:2 downscale.
Reduces aliasing in downsampled image.
Keep the phase to 0/off for now.
Change-Id: Ic547ea0748d151b675f877527e656407fcf4d51e
vp9_highbd_block_error_8bit_c was a very simple wrapper around
vp9_block_error_c. The SSE2 implemention was practically identical to
the non-HBD one. It was missing some minor improvements which only
went into the original version.
In quick speed tests, the AVX implementation showed minimal
improvement over SSE2 when it does not detect overflow. However, when
overflow is detected the function is run a second time. The
OperationCheck test seems to trigger this case and reverses any
speed benefits by running ~60% slower. AVX2 on the other hand is
always 30-40% faster.
Change-Id: I9fcb9afbcb560f234c7ae1b13ddb69eca3988ba1
vp9[_highbd]_quantize]_fp[_32x32] and vp9_fdct8x8_quant do not make use
of these parameters.
scan is used for C code and iscan is used for SIMD implementations.
Change-Id: I908a0ff7d3febac33da97e0596e040ec7bc18ca5
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
development has moved to the nextgenv2 branch and a snapshot from here
was used to seed aomedia
BUG=b/29457125
Change-Id: Iedaca11ec7870fb3a4e50b2c9ea0c2b056a0d3c0
This reverts commit be12fefa4b
and commit 057c1c4034.
Also, the mismatch between the avx version and the
c version has been fixed.
BUG=https://bugs.chromium.org/p/webm/issues/detail?id=1168
For a rt encode using 1080p@60fps material, up to 11% performance
improvement overall was seen.
Change-Id: Icd1f216209ebc6fc0b8da885f32f356fa4355ed0
Function level timing test shows about 27% time saving on
a Xeon E5-2680 v2 desktop.
Rename vp9_dct_sse2.c to vp9_dct_intrin_sse2.c for vp9 and
rename dct_sse2.c to dct_intrin_sse2.c for vp10 to avoid
duplicate basenames.
Actually vp9_fwht4x4_mmx/sse2() and vp10_fwht4x4_mmx/sse2()
are identical. TODO: They should be unified later if there is
no intention to keep a duplicate.
Change-Id: I3e537b7bbd9ba417c606cd7c68c4dbbfa583f77d
vp9_diamond_search_sad_avx was disabled in:
057c1c4 disable vp9_diamond_search_sad_avx
this removes a missing prototype warning as the prototype is no longer
included in vp9_rtcd.h. the file can be restored if someone gets around
to fixing the issue.
BUG=https://bugs.chromium.org/p/webm/issues/detail?id=1168
Change-Id: Ia9fda4b81c53dc5fba7c31d780d761f886940b52
Block size passed into denoiser filter is always >= BLOCK_8X8 (in
vp9_pick_inter_mode), it is not necessary to check smaller block
size. Passed the bitexact test on clips with different resolutions and
noise levels.
Change-Id: I19fa3195d18c27d9e5de60dc11cff1522ef3714e
The bit to error transformation got doubled as a result of going from
8-bit to 9-bit costs (change d13385c).
Use defines to derive the scale numbers and comment some of the fields.
derf: -0.023 BDRATE
hevcmr: +0.067 BDRATE
stdhd: +0.098 BDRATE
(These are substantially smaller than than the original gains from 8 to
9 bit costing.)
Change-Id: I6a2b3b029b2f1415e4f90a05709b2333ec0eea9b
the final sum may use up to 26 bits
+ add a unit test
+ disable the sse2 as the result will rollover; this will be fixed in a
future commit
Change-Id: I2a49811dfaa06abfd9fa1e1e65ed7cd68e4c97ce
This function now has an AVX intrinsics version which is about 80%
faster compared to the C implementation. This provides a 2-4% total
speed-up for encode, depending on encoding parameters. The function
utilizes 3 properties of the cost function lookup table, constructed
in 'cal_nmvjointsadcost' and 'cal_nmvsadcosts'.
For the joint cost:
- mvjointsadcost[1] == mvjointsadcost[2] == mvjointsadcost[3]
For the component costs:
- For all i: mvsadcost[0][i] == mvsadcost[1][i]
(equal per component cost)
- For all i: mvsadcost[0][i] == mvsadcost[0][-i]
(Cost function is even)
These must hold, otherwise the AVX version of the function cannot be used.
Change-Id: I6c2791d43022822a9e6ab43cd124a773946d0bdc
This reverts commit f1342a7b07.
This breaks 32-bit builds:
runtime error: load of misaligned address 0xf72fdd48 for type 'const
__m128i' (vector of 2 'long long' values), which requires 16 byte
alignment
+ _mm_set1_epi64x is incompatible with some versions of visual studio
Change-Id: I6f6fc3c11403344cef78d1c432cdc9147e5c1673
This function now has an AVX intrinsics version which is about 80%
faster compared to the C implementation. This provides a 2-4% total
speed-up for encode, depending on encoding parameters. The function
utilizes 3 properties of the cost function lookup table, constructed
in 'cal_nmvjointsadcost' and 'cal_nmvsadcosts'.
For the joint cost:
- mvjointsadcost[1] == mvjointsadcost[2] == mvjointsadcost[3]
For the component costs:
- For all i: mvsadcost[0][i] == mvsadcost[1][i]
(equal per component cost)
- For all i: mvsadcost[0][i] == mvsadcost[0][-i]
(Cost function is even)
These must hold, otherwise the AVX version of the function cannot be used.
Change-Id: I184055b864c5a2dc37b2d8c5c9012eb801e9daf6
A new version of vp9_highbd_error_8bit is now available which is
optimized with AVX assembly. AVX itself does not buy us too much, but
the non-destructive 3 operand format encoding of the 128bit SSEn integer
instructions helps to eliminate move instructions. The Sandy Bridge
micro-architecture cannot eliminate move instructions in the processor
front end, so AVX will help on these machines.
Further 2 optimizations are applied:
1. The common case of computing block error on 4x4 blocks is optimized
as a special case.
2. All arithmetic is speculatively done on 32 bits only. At the end of
the loop, the code detects if overflow might have happened and if so,
the whole computation is re-executed using higher precision arithmetic.
This case however is extremely rare in real use, so we can achieve a
large net gain here.
The optimizations rely on the fact that the coefficients are in the
range [-(2^15-1), 2^15-1], and that the quantized coefficients always
have the same sign as the input coefficients (in the worst case they are
0). These are the same assumptions that the old SSE2 assembly code for
the non high bitdepth configuration relied on. The unit tests have been
updated to take this constraint into consideration when generating test
input data.
Change-Id: I57d9888a74715e7145a5d9987d67891ef68f39b7
If high bit depth configuration is enabled, but encoding in profile 0,
the code now falls back on optimized SSE2 assembler to compute the
block errors, similar to when high bit depth is not enabled.
Change-Id: I471d1494e541de61a4008f852dbc0d548856484f
Rename updated version of x86inc.asm
Use "private_prefix" instead of "program_name" and make vpx the default
prefix.
Change-Id: I4883a99b2aee8e5dc9f2c16a2e6f4b5d6e4de458
This commit moves the module inverse transform functions from vp9
to vpx_dsp folder. The hybrid transform wrapper functions stay in
the vp9 folder, since it involves codec-specific data structures.
Change-Id: Ib066367c953d3d024c73ba65157bbd70a95c9ef8
