- adding one extra pixel all around the frame
- do not copy when SAO is not applied
5% improvement
cherry picked from commit 10fc29fc19a12c4d8168fbe1a954b76386db12d0
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The previous implementation targeted DTS Coherent Acoustics, which only
requires nbits == 4 (fft16()). This case was (and still is) linked directly
rather than being indirected through ff_fft_calc_vfp(), but now the full
range from radix-4 up to radix-65536 is available. This benefits other codecs
such as AAC and AC3.
The implementaion is based upon the C version, with each routine larger than
radix-16 calling a hierarchy of smaller FFT functions, then performing a
post-processing pass. This pass benefits a lot from loop unrolling to
counter the long pipelines in the VFP. A relaxed calling standard also
reduces the overhead of the call hierarchy, and avoiding the excessive
inlining performed by GCC probably helps with I-cache utilisation too.
I benchmarked the result by measuring the number of gperftools samples that
hit anywhere in the AAC decoder (starting from aac_decode_frame()) or
specifically in the FFT routines (fft4() to fft512() and pass()) for the
same sample AAC stream:
Before After
Mean StdDev Mean StdDev Confidence Change
Audio decode 2245.5 53.1 1599.6 43.8 100.0% +40.4%
FFT routines 940.6 22.0 348.1 20.8 100.0% +170.2%
Signed-off-by: Martin Storsjö <martin@martin.st>
The previous implementation targeted DTS Coherent Acoustics, which only
requires mdct_bits == 6. This relatively small size lent itself to
unrolling the loops a small number of times, and encoding offsets
calculated at assembly time within the load/store instructions of each
iteration.
In the more general case (codecs such as AAC and AC3) much larger arrays
are used - mdct_bits == [8, 9, 11]. The old method does not scale for
these cases, so more integer registers are used with non-unrolled versions
of the loops (and with some stack spillage). The postrotation filter loop
is still unrolled by a factor of 2 to permit the double-buffering of some
VFP registers to facilitate overlap of neighbouring iterations.
I benchmarked the result by measuring the number of gperftools samples
that hit anywhere in the AAC decoder (starting from aac_decode_frame())
or specifically in ff_imdct_half_c / ff_imdct_half_vfp, for the same
example AAC stream:
Before After
Mean StdDev Mean StdDev Confidence Change
aac_decode_frame 2368.1 35.8 2117.2 35.3 100.0% +11.8%
ff_imdct_half_* 457.5 22.4 251.2 16.2 100.0% +82.1%
Signed-off-by: Martin Storsjö <martin@martin.st>
These where removed by libav in
See: git show -C 2d60444331
diff --git a/libavcodec/dsputil.c b/libavcodec/me_cmp.c
similarity index 98%
rename from libavcodec/dsputil.c
rename to libavcodec/me_cmp.c
index ba71a99..9fcc937 100644
--- a/libavcodec/dsputil.c
+++ b/libavcodec/me_cmp.c
@@ -1,8 +1,4 @@
/*
- * DSP utils
- * Copyright (c) 2000, 2001 Fabrice Bellard
- * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
- *
* This file is part of Libav.
*
* Libav is free software; you can redistribute it and/or
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Fixes fate failure with --enable-memory-poisoning && make THREAD_TYPE=slice THREADS=7 fate-hevc-conformance-ENTP_C_Qualcomm_1
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Mickaël Raulet
Michael Niedermayer
Diego Biurrun
Carl Eugen Hoyos
Nidhi Makhijani
Timothy Gu
Martin Storsjö
James Almer
Ben Avison
Luca Barbato
Vittorio Giovara