This fixes crashes in e.g. PNG decoding with SSE2 enabled. In fact, many
x86 optimizations for codecs assume that our buffer strides are 16-byte
aligned.
High bitdepth H.264 needs 32-bit transform coefficients, whereas
dnxhd does not. This creates a conflict with the templated
functions operating on DCTELEM data. This patch adds a field
allowing the caller to choose the element size in dsputil_init()
and adds the required functions.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Use of these has been broken ever since the h264 idct was changed
to always use transposed inputs. Furthermore, they were only
ever used if some *other* non-default idct was requested.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This macro can cause problems in conjunction with the bitdepth
template expansion. It was presumably added to keep source
compatibility when high bitdepth support was added. However,
emulated_edge_mc is a dsputil pointer and should not be called
directly, so there is little reason to keep such a macro.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This patch lets e.g. dsputil_init chose dsp functions with respect to
the bit depth to decode. The naming scheme of bit depth dependent
functions is <base name>_<bit depth>[_<prefix>] (i.e. the old
clear_blocks_c is now named clear_blocks_8_c).
Note: Some of the functions for high bit depth is not dependent on the
bit depth, but only on the pixel size. This leaves some room for
optimizing binary size.
Preparatory patch for high bit depth h264 decoding support.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
There are several places where a buffer is byte-swapped in 16-bit units.
This allows them to share code which can be optimised for various
architectures.
Signed-off-by: Mans Rullgard <mans@mansr.com>
This will be beneficial for use with the audio conversion API without
requiring it to depend on all of dsputil.
Signed-off-by: Mans Rullgard <mans@mansr.com>
C99 variadic macros require more arguments than there are named
parameters in the definition. This means we must use an extra
indirection to avoid having two different macros for arrays with
one resp more than one dimension.
Signed-off-by: Mans Rullgard <mans@mansr.com>
No speed improvement, but necessary for some future stuff.
Also opens up the possibility of asm chroma dc idct/dequant.
Originally committed as revision 26349 to svn://svn.ffmpeg.org/ffmpeg/trunk
Doesn't help speed as there isn't an asm implementation yet, but consistency
is a good thing.
Originally committed as revision 26348 to svn://svn.ffmpeg.org/ffmpeg/trunk
About 2.5x the speed.
NOTE: the way that the asm code handles large qmuls is a bit suboptimal.
If x264-style dequant was used (separate shift and qmul values), it might
be possible to get some extra speed.
Originally committed as revision 26336 to svn://svn.ffmpeg.org/ffmpeg/trunk
This removes duplicated definitions of 8x8 and 16x16 fullpel MC
functions with various names reducing dsputil.o by 8k on x86_64.
Originally committed as revision 24933 to svn://svn.ffmpeg.org/ffmpeg/trunk
Passing an explicit filename to this command is only necessary if the
documentation in the @file block refers to a file different from the
one the block resides in.
Originally committed as revision 22921 to svn://svn.ffmpeg.org/ffmpeg/trunk
This moves the H264-specific functions from DSPContext to the new
H264DSPContext. The code is made conditional on CONFIG_H264DSP
which is set by the codecs requiring it.
The qpel and chroma MC functions are not moved as these are used by
non-h264 code.
Originally committed as revision 22565 to svn://svn.ffmpeg.org/ffmpeg/trunk
This moves the DWT functions from snow.c and dsputil.c to a file of
their own. A new struct, DWTContext, holds the function pointers
previously part of DSPContext.
Originally committed as revision 22522 to svn://svn.ffmpeg.org/ffmpeg/trunk
These macros are redundant. All uses are replaced with the generic
DECLARE_ALIGNED macro instead.
Originally committed as revision 22233 to svn://svn.ffmpeg.org/ffmpeg/trunk
Since BGR24 is decoded as BGR32, fill its alpha channel with 255
using the appropriate predictors.
Originally committed as revision 21211 to svn://svn.ffmpeg.org/ffmpeg/trunk
29-105% faster apply_filter, 6-90% faster ape decoding on core2
(Any x86 other than core2 probably gets much less, since this is mostly due to ssse3 cachesplit avoidance and I haven't written the full gamut of other cachesplit modes.)
9-123% faster ape decoding on G4.
Originally committed as revision 20739 to svn://svn.ffmpeg.org/ffmpeg/trunk
necessary ff_cos_tabs tables are initialized.
Fixes issue 1507 (QDM2 broken since r20237 without hardcoded tables).
Originally committed as revision 20464 to svn://svn.ffmpeg.org/ffmpeg/trunk
While this "wastes" up to 2x32 bytes it makes the code slightly simpler and
less confusing.
Originally committed as revision 20449 to svn://svn.ffmpeg.org/ffmpeg/trunk
2.2x faster than C on conroe, 3.6x on penryn.
4-6% faster huffyuv decoding if using left or plane mode and yuv
Originally committed as revision 20287 to svn://svn.ffmpeg.org/ffmpeg/trunk
corresponding dsputil functions and remove their dependency on the FLAC
encoder.
Fixes Issue1486.
Originally committed as revision 20266 to svn://svn.ffmpeg.org/ffmpeg/trunk
initialized by ff_fft_init and using different code can result in slightly
different values, in addition it crashes when the tables are hardcoded.
On amd64 this slightly changes qdm2 output.
Originally committed as revision 20237 to svn://svn.ffmpeg.org/ffmpeg/trunk
--enable-hardcoded-tables was used.
Due to the size, the code for the tables is generated at compile time.
Originally committed as revision 20232 to svn://svn.ffmpeg.org/ffmpeg/trunk
The src3 and step arguments to vector_fmul_add_add() are always zero
and one, respectively. This removes these arguments from the function,
simplifies the code accordingly, and renames the function to better
match the new operation.
Originally committed as revision 20061 to svn://svn.ffmpeg.org/ffmpeg/trunk
This adds a function pointer for forward MDCT to FFTContext and
initialises it with the existing C function. ff_calc_mdct() is
changed to an inline function calling the selected version as
done for other fft/mdct functions.
Originally committed as revision 19818 to svn://svn.ffmpeg.org/ffmpeg/trunk
The DECLARE_ALIGNED_8 macro is defined to align to 16 bytes instead
the 8 suggested by the name on some CPUs. None of the uses of this
macro ever need 16-byte alignment, cases which once did having been
changed to always specify 16 bytes explicitly.
Originally committed as revision 19737 to svn://svn.ffmpeg.org/ffmpeg/trunk
