speeds up decoding of 64kbps test file by 19MHz on h300 (cf)
and 2.5MHz on c200 (pp)
Change-Id: Idacd2f8962c20c518055d586daeec6b932b7ded2
Signed-off-by: Nils Wallménius <nils@rockbox.org>
Although the mimetype for jpeg is clearly image/jpeg, many tagging
applications seem to use image/jpg, so we'll support that too.
Change-Id: Icb9063fd5a9d8aea169eaa7f74ac52b72603d148
Reviewed-on: http://gerrit.rockbox.org/318
Reviewed-by: Michael Giacomelli <mgiacomelli@gmail.com>
Reviewed-by: Thomas Martitz <kugel@rockbox.org>
Speeds up decoding of a 64kbps test file 26MHz on H300 (cf) and
2MHz on c200 (pp)
Change-Id: I2fb4fe6c0a29321087e02fbd17fd1b1eb84e7b57
Signed-off-by: Nils Wallménius <nils@rockbox.org>
Remove downsampling code from deemphasis loop as we don't use
it and remove multiplications that are not relevant when
not using custom modes. Saves 1.4MHz on h300 (cf), 4.3MHz on
c200 (pp) and 4.6 on fuzev1 (amsv1).
Change-Id: Iab3f1d737a656a563aaa351d50db987a9cff2287
Saves about 30MHz on h300 (cf) and 1.5MHz on c200 (pp) decoding a
64kbps test file. Stack usage is still below 70%.
Change-Id: Ib13df9011adb4eef4bb91a52e5a32741c8bf8988
Speeds up decoding of a 64kbps opus test file by 34MHz on h300 (cf),
24MHz on c200 (pp) and 13MHz on fuzev1 (amsv1)
Change-Id: I0dce6b3bfe6c81d0a722dfebb13891b9a428c6ba
Synchronised with opus repo on github (https://github.com/freqmod/rockbox-opus)
Status:
* Seeking ported from speex, but fails on some cases (e.g. seek to granule 0)
* ReplayGain parsing needs to be reworked, we do vorbis-style replaygain now.
http://wiki.xiph.org/OggOpus#Comment_Header explicitly forbids these in
favour of R128_TRACK_GAIN tag.
* No optimisation yet, source files still nearly identical to opus upstream
* Multi-stream opus files may not be parsed correctly
Change-Id: Ia66f1027dc1d288083e3c57b2816700078376f9a
Reviewed-on: http://gerrit.rockbox.org/300
Reviewed-by: Bertrik Sikken <bertrik@sikken.nl>
Tested-by: Bertrik Sikken <bertrik@sikken.nl>
Since gcc 4.4 the MIPS port no longer recognizes the "h" asm constraint.
It was necessary to remove this constraint in order to avoid generating
unpredictable code sequences. We can achieve the same effect using
128-bit types.
See also:GCC 4.4 release notes at http://gcc.gnu.org/gcc-4.4/
Change-Id: I713cdf57cde1a989ad960aa441ab1ccf51f1cdc6
Emulates the basic "Meier" crossfeed (2 capacitors, 3 resistors)
as discussed in
http://www.meier-audio.homepage.t-online.de/passivefilter.htm
This crossfeed blends a bit of low-pass filtered L signal into
the R signal (and vice versa) while adding about 300 us delay
to the crossfed-signal. A difference with the crossfeed already
present in rockbox, is that this algorithm keeps the total
spectrum flat (the one currently in rockbox accentuates
low-frequency signals, making it sound a bit muffled).
This implementation is quite lightweight, just 3 multiplies per
left-right pair of samples. Has a default C implementation and
optimized assembly versions for ARM and Coldfire.
The crossfeed effect is quite subtle and is noticeable mostly
one albums that have very strong left-right separation (e.g.
one instrument only on the left, another only on the right).
In the user interface, the new crossfeed option appears as
"Meier" and is not configureable. The existing crossfeed is
renamed to "Custom" as it allows itself to be customised.
There is no entry for the user manual yet.
Change-Id: Iaa100616fe0fcd7e16f08cdb9a7f41501973eee1
When it was inactive but enabled, the format change hook was dropping
through to code that it shouldn't execute without it also being active
in processing samples.
Change-Id: Ie7899df0395d3f0d10f2bf2b55ea549dd06749a7
Several of the problem samples on the tracker use values outside this
range. Trying the larger table doesn't quite seem to fix things, but
its only a small amount of additional memory and looking at ffmpeg,
I think the larger table is correct.
Change-Id: Id046e62b68550701aa1f80c9abd0a1dcd711bd0d
No known samples are fixed by this problem, but I haven't tested many.
Backport of ffmpeg revision 26388.
Change-Id: Ife9654b7477a432834e3cab2cb43d16da071445a
This is needed on app targets as e.g. ".section .icode" leads to the
code getting linked to incorrect locations (0x0 in this case).
Change-Id: Ic28c5ae6d4f8001d211d685b5ca92d5ffff0c7b2
Use memmove instead of memcpy for overlapping copy, fixes
crackling in sims and warble on my system. Native targets
seem to have been unaffected.
Change-Id: I265d4ce373e224581bd2f5ba15c75b473ec231f2
If the settings, like samplerate, were to go out of range where
timestretch drops out of processing and then go back to the same as
when they were valid, it would fail to switch back on by itelf.
Change-Id: Ic5bcb268540b0db8e0483117b8a5a0ce5c5a9db0
Makes quite a huge difference to get rid of 64-bit math in a hot
area. Cuts about 12 MHz. Generic routine generates good code on
ARM and asm cuts no instructions there.
Change-Id: I4ac647406006c42004f9f5ab396cbf4e85688854
Wrap up the the stereo case into loops and remove unused calculations
hanging out in tdspeed_update().
A wee little bit of code style and column policing.
Change-Id: I8dd3ab4b3e7e56b55dc00c00f3e32996228cc457
Also move the definition to config.h
Change-Id: I36bb5020c5e06b2344292bc05e8c13ccc7a6a1ff
Reviewed-on: http://gerrit.rockbox.org/234
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Input functions have common setup sequences that can be placed
into an inline function instead of repeating it all repeatedly.
Change-Id: I9e62904ff0948651c64ddf160ed4400ed6dc81ff
It was only needed by the old arm toolchain that we no longer use or support.
Change-Id: Id0e6c67477f8834a637079b03cde5fbf9da68b1c
Reviewed-on: http://gerrit.rockbox.org/233
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Move code dealing with rockbox specific buflib allocations into a
rockbox specific file and implement buffer allocation with
malloc/free for warble/stand alone lib.
Based on patch by Sean Bartell.
Change-Id: I8cb85dad5890fbd34c1bb26abbb89c0b0f6b55cf
Reviewed-on: http://gerrit.rockbox.org/144
Tested-by: Nils Wallménius <nils@rockbox.org>
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Reviewed-by: Nils Wallménius <nils@rockbox.org>
This would zero the first 4 or 8 bytes of the array because it is declared as a pointer
rockbox/lib/rbcodec/codecs/libasap/asap.c:1229:44: warning: argument to 'sizeof' in 'memset' call is the same expression as the destination; did you mean to provide an explicit length? [-Wsizeof-pointer-memaccess]
memset(ast -> memory, 0, sizeof(ast -> memory));
~~~~~~~~~~~~~ ~~~~~~~^~~~~~
It's always used in MONO mode and doesn't need the IRAM sample/
resample buffers and 1280 bytes can be freed.
M5 can now have its PCM mixer downmix buffer in IRAM.
Change-Id: I0af08be5b212b7dfe382bba588a6585eb328a038
Use them to move tick counting, yielding and coldfire macsr handling
code to a rockbox specific file.
Change-Id: Id7417dc98c08a342eba45ba56b044a276e50564b
Reviewed-on: http://gerrit.rockbox.org/229
Tested-by: Nils Wallménius <nils@rockbox.org>
Reviewed-by: Nils Wallménius <nils@rockbox.org>
librbcodec users must provide these two files when the library is built.
rbcodecconfig.h provides configuration #defines and basic types, and
will be included by public librbcodec headers, so it must not conflict
with the user's code. rbcodecplatform.h provides various OS functions,
and will only be included by source files and private headers. This
system is intended to provide maximum flexibility for use on embedded
systems, where no operating system headers are included. Unix systems
can just copy rbcodecconfig-example.h and rbcodecplatform-unix.h with
minimal changes.
Change-Id: I350a2274d173da391fd1ca00c4202e9760d91def
Reviewed-on: http://gerrit.rockbox.org/143
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Tested-by: Nils Wallménius <nils@rockbox.org>
Moved to playback.c, since it doesn't use metadata from the music file.
Change-Id: I5c3ad7750d94b36754f64eb302f96ec163785cb9
Reviewed-on: http://gerrit.rockbox.org/142
Reviewed-by: Nils Wallménius <nils@rockbox.org>
The LSP feature in WMA requires that the noise table values be
doubled verses when it is not used. Unfortunately, the previous
code would double the same values every time a LSP file was
decoded without first resetting them to their original values.
Change the code to check if the values are already doubled, and
then double/halve them as needed. This is still a bit ugly,
in the future consider using the built in rockbox dither instead
of a lookup table.
Fixes playback when skipping back and forth between low and high
bitrate WMA.
Change-Id: I4c393092e4a789bc8f98d74274fe207400b9550e
Reviewed-on: http://gerrit.rockbox.org/226
Reviewed-by: Michael Giacomelli <giac2000@hotmail.com>
Tested-by: Michael Giacomelli <giac2000@hotmail.com>
Moves replaygain definitions to lib/rbcodec/dsp/dsp_misc.h.
Intermediate functions in misc.c handle any adjustment and calling
the rbcodec APIs.
Change-Id: I9f03561bca9aedd13760cf19c4e19aa3c68e7024
Reviewed-on: http://gerrit.rockbox.org/140
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Move a few functions to .text that probably don't see a huge benefit
from being .icode. Will scrutinize later.
Change-Id: I7bdffc326076c5cd7e6a1c57d25d31e653920327
Creates a standard buffer passing, local data passing and messaging
system for processing stages. Stages can be moved to their own source
files to reduce clutter and ease assimilation of new ones. dsp.c
becomes dsp_core.c which supports an engine and framework for effects.
Formats and change notifications are passed along with the buffer so
that they arrive at the correct time at each stage in the chain
regardless of the internal delays of a particular one.
Removes restrictions on the number of samples that can be processed at
a time and it pays attention to destination buffer size restrictions
without having to limit input count, which also allows pcmbuf to
remain fuller and safely set its own buffer limits as it sees fit.
There is no longer a need to query input/output counts given a certain
number of input samples; just give it the sizes of the source and
destination buffers.
Works in harmony with stages that are not deterministic in terms of
sample input/output ratio (like both resamplers but most notably
the timestretch). As a result it fixes quirks with timestretch hanging
up with certain settings and it now operates properly throughout its
full settings range.
Change-Id: Ib206ec78f6f6c79259c5af9009fe021d68be9734
Reviewed-on: http://gerrit.rockbox.org/200
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Tested-by: Michael Sevakis <jethead71@rockbox.org>
This function has been changed to rbcodec_format_is_atomic, which
doesn't require an enum from the kernel.
Change-Id: I1d537605087fe130a9b545509d7b8a340806dbf2
Reviewed-on: http://gerrit.rockbox.org/141
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Tested-by: Nils Wallménius <nils@rockbox.org>
dsp_set_eq_coefs now has parameters for the band settings, so it doesn't
need to access global_settings.
Change-Id: I29ac19fc353b15a79cb25f0e45132aef0881e4c9
Reviewed-on: http://gerrit.rockbox.org/138
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Differences between signed samples cover the entire unsigned 32-bit
range. "abs" will think any difference exceeding INT32_MAX is negative
which is not corrent. Test which argument is greater and subtract the
lesser from it, outputting unsigned difference.
Change-Id: I73a8e5e418d49ff73d1a7c98eeb4731946dcfe84
libs in $ROOT/lib now add to $(CORE_LIBS) and $(EXTRA_LIBS) and are
automatically linked by the core and codecs/plugins respectively.
Change-Id: Iff482c792a8c8142718f6a16a450c6e2f1497c9a
Warble uses Rockbox's codecs to play files with SDL or convert them to
WAV or raw formats. It also prints metadata and supports some of the DSP
effects. In the future, warble could be used to implement an automated
test suite for codecs, metadata, and DSP.
Change-Id: Ife1a63d2354496016277bfcbae4a9c23423ebd86
Reviewed-on: http://gerrit.rockbox.org/135
Reviewed-by: Nils Wallménius <nils@rockbox.org>
Tested-by: Nils Wallménius <nils@rockbox.org>
