summaryrefslogtreecommitdiff
path: root/apps/codecs/demac (follow)
Commit message (Collapse)AuthorAge
* Correction to clobber lists of several codec's inline assembly.Andree Buschmann2010-05-29
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@26376 a1c6a512-1295-4272-9138-f99709370657
* Move c/h files implementing/defining standard library stuff into a new libc ↵Thomas Martitz2010-05-06
| | | | | | | | directory, also standard'ify some parts of the code base (almost entirely #include fixes). This is to a) to cleanup firmware/common and firmware/include a bit, but also b) for Rockbox as an application which should use the host system's c library and headers, separating makes it easy to exclude our files from the build. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25850 a1c6a512-1295-4272-9138-f99709370657
* Move (small) data into DRAM on PP5020, it's ~4.5% faster that way. Closes ↵Jens Arnold2010-03-10
| | | | | | about half of the performance gap towards PP5022. The (relatively large) buffers for decoded data stay in IRAM, as does the reciprocal table. Clarify some comments. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25108 a1c6a512-1295-4272-9138-f99709370657
* libdemac: Separate filtering calls per channel in preparation for the ↵Jens Arnold2010-03-03
| | | | | | dual-core split on PP. This also means less inlining, and hence speeds up decoding on single core slightly, due to better caching behaviour. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25005 a1c6a512-1295-4272-9138-f99709370657
* Get rid of .rept in inline asm() blocks where possible. Using .rept causes ↵Jens Arnold2010-03-03
| | | | | | gcc to wrongly estimate the size of the asm(), leading to (potential) compilation problems. This is necessary for the upcoming restructuring, and should fix ARMv6+ sim builds as well. No functional change. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25004 a1c6a512-1295-4272-9138-f99709370657
* Remove all tabs within codec path.Andree Buschmann2010-02-22
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24862 a1c6a512-1295-4272-9138-f99709370657
* Clarify comments in ARMv6 divider regarding special-case handling of large ↵Andrew Mahone2010-02-20
| | | | | | (high bit set) numerators. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24783 a1c6a512-1295-4272-9138-f99709370657
* libdemac: Add x86/x86_64 MMX asm for the filters. Not relevant for target ↵Jens Arnold2010-02-15
| | | | | | but speeds up decoding on x86/x86_64 sims. Average speedup ranges from 25% for -c2000 to 3 times for -c5000; on Intel Atom it's even 45% for -c2000 to 6 times for -c5000. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24663 a1c6a512-1295-4272-9138-f99709370657
* Remove leftover from before make system rework.Jens Arnold2010-02-11
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24602 a1c6a512-1295-4272-9138-f99709370657
* APE: Fused vector math for the filters on ARMv5te. Speedup on Cowon D2 is ↵Jens Arnold2010-02-10
| | | | | | ~4% for -c2000..-c4000 (less for -c5000). Thanks to Frank Gevaerts for testing. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24590 a1c6a512-1295-4272-9138-f99709370657
* APE: Fused vector math for the filters on ARMv6. Speedup is ~2.5% for ↵Jens Arnold2010-02-08
| | | | | | -c2000, ~7% for -c3000 and higher. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24569 a1c6a512-1295-4272-9138-f99709370657
* Put back the insane buffer where it belongs on non-ARM, and simplify the ↵Jens Arnold2010-02-04
| | | | | | selection. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24512 a1c6a512-1295-4272-9138-f99709370657
* Fix yellow: add newline at EOF in udiv32_arm-pre.SAndrew Mahone2010-02-04
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24508 a1c6a512-1295-4272-9138-f99709370657
* Use all available codec iram for reciprocal table in APE codec on ARMv4. ↵Andrew Mahone2010-02-04
| | | | | | Done by linking first with the table empty to determine free space, then sizing table to fill it. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24507 a1c6a512-1295-4272-9138-f99709370657
* Move udiv32_arm.S into libdemac, as this divider is specialized for the APE ↵Andrew Mahone2010-02-04
| | | | | | codec and an optimized divider is already provided for general use in codeclib. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24506 a1c6a512-1295-4272-9138-f99709370657
* APE codec: Speed up decoding of -c2000 and higher on ARMv4 and coldfire by ↵Jens Arnold2010-02-02
| | | | | | fusing vector math for the filters. Speedup is roughly 3.5% for -c2000, 8% for -c3000 and 12% for -c4000. To be extended to other architectures. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24473 a1c6a512-1295-4272-9138-f99709370657
* Improve libdemac SATURATE slightly on ARMv4/5, move filter buffers and code ↵Andrew Mahone2010-01-30
| | | | | | out of IRAM for sizes that aren't near realtime and extend udiv32_arm reciprocal table. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24376 a1c6a512-1295-4272-9138-f99709370657
* Improvements to specialized dividers for APE codec:Andrew Mahone2010-01-28
| | | | | | | | | * Use Newton-Raphson divider on ARMv5e and ARMv6, about 7% speedup on Gigabeat S. * On ARMv4 targets using IRAM, remove insane filter buffer from IRAM, fill available IRAM with LUT of reciprocals for small divisors - speedup varies according to target and available IRAM, APE normal sample is approx. 109% RT on e200. * Rename apps/codecs/lib/udiv32_armv4.S to apps/codecs/lib/udiv32_arm.S, which includes dividers for all ARM targets specialized for APE. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24354 a1c6a512-1295-4272-9138-f99709370657
* Make the codecs use more IRAM on S5L870x, as we have plenty of it.Michael Sparmann2009-11-09
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@23594 a1c6a512-1295-4272-9138-f99709370657
* Don't use ldrd/strd on ARMv5 since not all revisions support them and the ↵Jens Arnold2009-07-17
| | | | | | gain from using them is minimal (basically code size only). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@21916 a1c6a512-1295-4272-9138-f99709370657
* Silence warning from 'ar' if the archive had to be created.Jens Arnold2009-03-01
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@20151 a1c6a512-1295-4272-9138-f99709370657
* Stop hiding errors by redirecting stderr to /dev/null. If we really need to ↵Daniel Stenberg2009-02-23
| | | | | | do re-introduce somewhere we should rather make it dependent on the V variable so that make V=1 would still show the error and only "normal" builds would hide it. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@20090 a1c6a512-1295-4272-9138-f99709370657
* Calculate watermark from bitrate and harddisk spinup time.Björn Stenberg2009-01-10
| | | | | | | | Use a smaller PCM buffer on targets with 2MB or less ram. (FS#9703) git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19743 a1c6a512-1295-4272-9138-f99709370657
* static/const/#include/tab police on various filesBertrik Sikken2009-01-02
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19643 a1c6a512-1295-4272-9138-f99709370657
* Make local functions static in codecs, where possible.Bertrik Sikken2008-12-29
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19612 a1c6a512-1295-4272-9138-f99709370657
* Slight speedup for the APE filters. Most noticeable on coldfire (+3.5% for ↵Jens Arnold2008-12-22
| | | | | | -c2000), but also helps on the arm targets (+0.9% for -c2000 on PP5002). This transformation is oveflow safe, as absres < 2^24 is guaranteed. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19556 a1c6a512-1295-4272-9138-f99709370657
* Fix decoding of stereo frames with silence in only one channel. * Make the ↵Jens Arnold2008-12-21
| | | | | | standalone decoder contain debugging information. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19552 a1c6a512-1295-4272-9138-f99709370657
* Fix handling of 8 bit mono and stereo APE files, and also optimise 16 and 24 ↵Jens Arnold2008-12-21
| | | | | | bit output in the standalone decoder a bit. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19517 a1c6a512-1295-4272-9138-f99709370657
* Assembler optimised mono predictor for ARM. Speedup for -c1000 mono is ~5% ↵Jens Arnold2008-12-09
| | | | | | on PP, ~8% on Gigabeat S (less for higher compression levels). Also fix some overlooked comments in the stereo predictor. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19375 a1c6a512-1295-4272-9138-f99709370657
* Implement mono predictor in assembler for coldfire, yielding a ~6% speedup ↵Jens Arnold2008-12-02
| | | | | | for mono -c1000. Apply ideas gained from it back to the stereo predictor, saving 4 instructions. No speed increase for stereo, probably due to cache aliasing effects. * 80-column police. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19296 a1c6a512-1295-4272-9138-f99709370657
* Shuffling around register allocation allows to keep decoded0 and decoded1 in ↵Jens Arnold2008-12-01
| | | | | | registers, for a slight speedup. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19287 a1c6a512-1295-4272-9138-f99709370657
* Remove extraneous semicolons, and fix a comment.Jens Arnold2008-11-30
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19268 a1c6a512-1295-4272-9138-f99709370657
* Fix APE 16-bit mono output: mono signals need to be scaled for rockbox.Jens Arnold2008-11-30
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19264 a1c6a512-1295-4272-9138-f99709370657
* Resurrect the ARM7 16-bit packed vector addition/subtraction for ARMv5, ↵Jens Arnold2008-11-28
| | | | | | giving a nice speedup for the higher compression levels (tested on Cowon D2). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19260 a1c6a512-1295-4272-9138-f99709370657
* On ARM9TDMI (e.g. Gigabeat F) it's faster to use a ldr/str pair than ↵Jens Arnold2008-11-27
| | | | | | add+ldmia/stmia for 2 registers. On ARM7TDMI a str pair is equally fast, so go for the simpler macro and use it for all ARMv4. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19250 a1c6a512-1295-4272-9138-f99709370657
* Speed up the predictor a little by using ldrd/strd on ARMv5+. This required ↵Jens Arnold2008-11-27
| | | | | | shuffling around the register allocation somewhat. Performance on ARMv4 is unaffected. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19248 a1c6a512-1295-4272-9138-f99709370657
* Get rid of unused return values, except the one from decode_chunk() which ↵Jens Arnold2008-11-26
| | | | | | will be used in the dual core split. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19236 a1c6a512-1295-4272-9138-f99709370657
* Added 'keywords' and 'eol-style' properties.Björn Stenberg2008-11-25
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19218 a1c6a512-1295-4272-9138-f99709370657
* Reorder instructions to avoid pipeline stalls on ARMv6 wherever possible ↵Jens Arnold2008-11-24
| | | | | | (sometimes using different registers to allow this). Speeds up the predictor by almost 20% on ARMv6 (overall speedup for -c1000 is 5%), and might also help a bit on ARMv5. ARMv4 speed is unaffected. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19210 a1c6a512-1295-4272-9138-f99709370657
* Branch optimisation in both C (giving hints to gcc - verified using ↵Jens Arnold2008-11-24
| | | | | | -fprofile-arcs and gcov) and asm files. Biggest effect on coldfire (-c1000: +8%, -c2000: +5%), but ARM also profits a bit (less than 1% on ARM7TDMI, around 1% on ARM1136). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19199 a1c6a512-1295-4272-9138-f99709370657
* Tweak the ARMv6 filter assembly a bit further.Jens Arnold2008-11-24
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19198 a1c6a512-1295-4272-9138-f99709370657
* Remove .a files before running ar, to avoid problems with renamed files ↵Björn Stenberg2008-11-20
| | | | | | remaining in the library. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19160 a1c6a512-1295-4272-9138-f99709370657
* New makefile solution: A single invocation of 'make' to build the entire ↵Björn Stenberg2008-11-20
| | | | | | | | | tree. Fully controlled dependencies give faster and more correct recompiles. Many #include lines adjusted to conform to the new standards. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19146 a1c6a512-1295-4272-9138-f99709370657
* Several tweaks and cleanups: * Use .rept instead of repeated macros for ↵Jens Arnold2008-11-19
| | | | | | repeating blocks. * Use MUL (variant) instead of MLA (variant) in the first step of the ARM scalarproduct() if there's no loop. * Unroll ARM assembler functions to 32 where not already done, plus the generic scalarproduct(). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19144 a1c6a512-1295-4272-9138-f99709370657
* Compile-time choice between 16 bit and 32 bit integers for the filters. 32 ↵Jens Arnold2008-11-19
| | | | | | bit filters are faster on ARMv4 (with assembler code), so use them there. Nice speedup on PP and Gigabeat F/X. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19140 a1c6a512-1295-4272-9138-f99709370657
* Centralise compile-time configuration.Jens Arnold2008-11-16
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19121 a1c6a512-1295-4272-9138-f99709370657
* Make it compile again on linux...Jens Arnold2008-11-16
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19120 a1c6a512-1295-4272-9138-f99709370657
* Make the standalone decoder actually work on Windows (need to open the ↵Jens Arnold2008-11-16
| | | | | | output file in binary mode). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19119 a1c6a512-1295-4272-9138-f99709370657
* Make the standalone decoder build on cygwin.Jens Arnold2008-11-16
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19117 a1c6a512-1295-4272-9138-f99709370657
* Move the contents of rangecoding.h into entropy.c, and remove the former. It ↵Jens Arnold2008-11-16
| | | | | | was only used there, and defined some variables in the .h git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19116 a1c6a512-1295-4272-9138-f99709370657
generated by cgit v1.1 at 2026-05-04 06:50:39 +0000