Local variable offsets are now consistent for the functions,
removed unused parameters, reworked the assembly to eliminate
stalls/instructions.
Change-Id: Iaa37668f8a9bb8754df435f6a51c3a08d547f879
Break MFQE code into it's own file.
It is currently only valid for 16x16 and 8x8 Y blocks. It also filters
4x4 U/V blocks.
Refactor filtering and add associated assembly. Limited test cases show
--mfqe introduces a penalty of ~20% with HD content. The assembly
reduces the penalty to ~15%
Change-Id: I4b8de6b5cdff5413037de5b6c42f437033ee55bf
On Android NDK, rand() is inlined function. But, on our SSE optimization,
we need symbol for rand()
Change-Id: I42ab00e3255208ba95d7f9b9a8a3605ff58da8e1
This is a proof of concept RTCD implementation to replace the current
system of nested includes, prototypes, INVOKE macros, etc. Currently
only the decoder specific functions are implemented in the new system.
Additional functions will be added in subsequent commits.
Overview:
RTCD "functions" are implemented as either a global function pointer
or a macro (when only one eligible specialization available).
Functions which have RTCD specializations are listed using a simple
DSL identifying the function's base name, its prototype, and the
architecture extensions that specializations are available for.
Advantages over the old system:
- No INVOKE macros. A call to an RTCD function looks like an ordinary
function call.
- No need to pass vtables around.
- If there is only one eligible function to call, the function is
called directly, rather than indirecting through a function pointer.
- Supports the notion of "required" extensions, so in combination with
the above, on x86_64 if the best function available is sse2 or lower
it will be called directly, since all x86_64 platforms implement
sse2.
- Elides all references to functions which will never be called, which
could reduce binary size. For example if sse2 is required and there
are both mmx and sse2 implementations of a certain function, the
code will have no link time references to the mmx code.
- Significantly easier to add a new function, just one file to edit.
Disadvantages:
- Requires global writable data (though this is not a new requirement)
- 1 new generated source file.
Change-Id: Iae6edab65315f79c168485c96872641c5aa09d55
This patch removes the local copies of the dequantize
constants and implements John's idea as described
in "Make a local copy of the dequantized data" commit.
Change-Id: Ic6b7d681f00bf63263f71ff1e39ab2f80729e8b2
These functions are now used by the encoder.
This is WIP with the goal of creating a common idct/add for
the encoder and decoder. A boost of 1.8% was seen for
the HD rt test clip used.
[Tero] Added needed changes to ARM side.
Change-Id: Ibbb8000be09034203d7adffc457d3c3f8b06a5bf
to the dqcoeff or qcoeff buffer. The encoder would
populate the dc coeffs of the y blocks as a separate
stage (recon_dcblock) and the decoder would use a special
version of the idct. This change eliminates the extra copy
and reduces the code footprint.
[Tero] Added needed changes to armv6 and NEON assembly.
Change-Id: I83202ffdbaf83f6e5dd69f4ba2519fcf0b13b3ba
Storing vp8_bilinear_filters_mmx in an mmx file and using it in an sse2
file is bad
Moving towards allowing --disable-mmx
Change-Id: I20493b35bdedcdcfc0915e6f05fdbe6c81a4a742
Ronald recently sent me this patch that he did in April.
> From: Ronald S. Bultje <rbultje@google.com>
> Date: Thu, 28 Apr 2011 17:30:15 -0700
> Subject: [PATCH] SSE2 optimizations for
> vp8_build_intra_predictors_mby{,_s}().
HD decode tests have shown a performance boost up to 1.5%,
depending on material.
Patch set 3: Fixed encoder crash.
Change-Id: Ie1fd1fa3dc750eec1a7a20bfa2decc079dcf48c8
Instead of using the predict buffer, the decoder now writes
the predictor into the recon buffer. For blocks with eob=0,
unnecessary idcts can be eliminated. This gave a performance
boost of ~1.8% for the HD clips used.
Tero: Added needed changes to ARM side and scheduled some
assembly code to prevent interlocks.
Patch Set 6: Merged (I1bcdca7a95aacc3a181b9faa6b10e3a71ee24df3)
into this commit because of similarities in the idct
functions.
Patch Set 7: EC bug fix.
Change-Id: Ie31d90b5d3522e1108163f2ac491e455e3f955e6
The data processed by the loopfilter overlaps. At the block level, this
results in some redundant transforms. Grouping the filtering allows for
a single 16x16 transpose (and inversion) instead of three 16x8 transposes
(and three more inversions).
This implementation is x86_64 only. We retain the previous
implementation for x86.
Improvements are obviously material dependant, but it seems to be ~%1 in
tests here.
Change-Id: I467b7ec3655be98fb5f1a94b5d145e5e5a660007
Prepend idct function names with vp8_
so that under profiling they show up
associated with libvpx.
Change-Id: I4fe357b50236cb7730a4cc00164c0a3487a1d8b4
The data that the simple horizontal loopfilter reads is aligned, treat
it accordingly.
For the vertical, we only use the bottom 4 bytes, so don't read in 16
(and incur the penalty for unaligned access).
This shows a small improvement on older processors which have a
significant penalty for unaligned reads.
postproc_mmx.c is unused
Change-Id: I87b29bbc0c3b19ee1ca1de3c4f47332a53087b3d
Prepend . to local labels in assembly code. This
allows non unique labels within a file. Also
makes profiling information more informative
by keeping the function name with the loop name.
Change-Id: I7a983cb3a5ba2413d5dafd0a37936b268fb9e37f
global values were being referenced, but the GOT was not being set up.
as the GOT is only required for PIC, this issue wasn't caught in the
default configuration.
Change-Id: I8006e53776139362a76f2c80cf9d0f8458602b2f
http://code.google.com/p/webm/issues/detail?id=328
the decision to run the regular or simple loopfilter is made outside the
function and managed with pointers
stop tracking the option in two places. use filter_type exclusively
Change-Id: I39d7b5d1352885efc632c0a94aaf56b72cc2fe15
the win64 abi requires saving and restoring xmm6:xmm15. currently
SAVE_XMM and RESTORE XMM only allow for saving xmm6:xmm7. allow
specifying the highest register used and if the stack is unaligned.
Change-Id: Ica5699622ffe3346d3a486f48eef0206c51cf867
Went through the code and fixed it. Verified on Windows.
Where possible, remove dependencies on xmm[67]
Current code relies on pushing rbp to the stack to get 16 byte
alignment. This broke when rbp wasn't pushed
(vp8/encoder/x86/sad_sse3.asm). Work around this by using unaligned
memory accesses. Revisit this and the offsets in
vp8/encoder/x86/sad_sse3.asm in another change to SAVE_XMM.
Change-Id: I5f940994d3ebfd977c3d68446cef20fd78b07877
vp8_filter_block1d16_h4_ssse3 was never called
because UNSHADOW_ARGS moves the stack by 'mov rsp, rbp', the issue was
masked. however, if/when win64 used those registers for persistant data,
issues could/will arise.
Change-Id: I56d6effca0aeba1f86082689771cb10145d39651
A large number of functions were defined with external linkage, even
though they were only used from within one file. This patch changes
their linkage to static and removes the vp8_ prefix from their names,
which should make it more obvious to the reader that the function is
contained within the current translation unit. Functions that were
not referenced were removed.
These symbols were identified by:
$ nm -A libvpx.a | sort -k3 | uniq -c -f2 | grep ' [A-Z] ' \
| sort | grep '^ *1 '
Change-Id: I59609f58ab65312012c047036ae1e0634f795779
