summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Add the thread state check into test_codec as used in mpegplayer which makes ↵Michael Sevakis2007-10-19
| | | | | | it should make it less prone to trash the codec thread after use. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15203 a1c6a512-1295-4272-9138-f99709370657
* Chessbox: ported to c200, also simplify the tile size calculation (thanks to ↵Marianne Arnold2007-10-19
| | | | | | Jens). Just retrieve it from the assigned bmp (same as e.g. Sudoku does), so that there's no need to check for lcd resolution in two different places. Update the manual. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15202 a1c6a512-1295-4272-9138-f99709370657
* Patches moved out from twiki.Björn Stenberg2007-10-19
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15201 a1c6a512-1295-4272-9138-f99709370657
* APE codec: Further optimised filtering yields 3..4% speedup for -c2000 (now ↵Jens Arnold2007-10-19
| | | | | | 135% realtime), -c3000 (now 97% realtime) and higher modes. Single 32 bit stores are faster than movem/lea in IRAM. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15200 a1c6a512-1295-4272-9138-f99709370657
* Reverse some logic with Peterson's algorithm and cut an instruction. Make ↵Michael Sevakis2007-10-19
| | | | | | C-reference reflect it. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15199 a1c6a512-1295-4272-9138-f99709370657
* dirbrowse should return a GO_TO_* value, not false on error (fixes the ↵Jonathan Gordon2007-10-19
| | | | | | problem where browsing plugins/fons/etc dirs from the menu would put you back n the filebrowser if the folder was empty instead of back in the menu) git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15198 a1c6a512-1295-4272-9138-f99709370657
* Code police strike. Every line was padded with trailing space. Trim it.Michael Sevakis2007-10-19
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15197 a1c6a512-1295-4272-9138-f99709370657
* start of a working keymap for mrobe 500Jonathan Gordon2007-10-19
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15196 a1c6a512-1295-4272-9138-f99709370657
* init spi and uart in the main buildJonathan Gordon2007-10-19
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15195 a1c6a512-1295-4272-9138-f99709370657
* APE codec: Assembler optimised vector math routines for coldfire. -c2000 is ↵Jens Arnold2007-10-18
| | | | | | now usable at 130% realtime (was 107%), -c3000 is near realtime (93%, was 64%). -c1000 doesn't change. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15194 a1c6a512-1295-4272-9138-f99709370657
* Replace thumb-interwork-4.0.3.diff patch for arm-elf-gcc with ↵Dave Chapman2007-10-18
| | | | | | rockbox-multilibs-arm-elf-gcc-4.0.3.diff. This builds libgcc.a for armv4 (current targets), armv5 (M:Robe), armv6 (Gigabeat S) and thumb interworking. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15193 a1c6a512-1295-4272-9138-f99709370657
* Just fix a typo.Mustapha Senhaji2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15192 a1c6a512-1295-4272-9138-f99709370657
* Bump version to 1.0.2 for new rbutil release.Dominik Riebeling2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15190 a1c6a512-1295-4272-9138-f99709370657
* Clean out an unused struct member from the synthObject structNils Wallménius2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15189 a1c6a512-1295-4272-9138-f99709370657
* Removed annoying 'Saving settings' splashBjörn Stenberg2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15188 a1c6a512-1295-4272-9138-f99709370657
* slightly better touchpad driver. Still not brilliant, but the ↵Jonathan Gordon2007-10-18
| | | | | | bootloader/debugger can follow a stylus with a cursor, so we can start work on getting apps/ working git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15187 a1c6a512-1295-4272-9138-f99709370657
* Make rockbox think we have a full battery untill this actually works, ↵Jonathan Gordon2007-10-18
| | | | | | otherwise apps/ doesnt like it git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15186 a1c6a512-1295-4272-9138-f99709370657
* Fix a warning in the simDave Chapman2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15185 a1c6a512-1295-4272-9138-f99709370657
* Oversight in syncing on my end, sorry.Thom Johansen2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15184 a1c6a512-1295-4272-9138-f99709370657
* Add Coldfire and ARM assembler for "reverse multiply and copy" function too. ↵Thom Johansen2007-10-18
| | | | | | Gives big speedup on Coldfire, small on ARM. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15183 a1c6a512-1295-4272-9138-f99709370657
* Put some arrays used in the decoder into IRAM for a few percent speedup ↵Jens Arnold2007-10-18
| | | | | | (mainly on coldfire). git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15182 a1c6a512-1295-4272-9138-f99709370657
* Use the core codec thread's stack for the mpegplayer audio thread - this ↵Dave Chapman2007-10-18
| | | | | | will free up another 9KB of IRAM for use by mpegplayer git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15181 a1c6a512-1295-4272-9138-f99709370657
* Some changes to the UART/button driver - reduce the number of interrupts - ↵Karl Kurbjun2007-10-18
| | | | | | more work is needed git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15180 a1c6a512-1295-4272-9138-f99709370657
* PP502x: Get switch_thread back out of IRAM and devise a better core wakeup ↵Michael Sevakis2007-10-18
| | | | | | system that's not timing dependant. Hopefully something simpler will be found or devised eventually that meets all requirements. Rename mailbox-related registers. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15179 a1c6a512-1295-4272-9138-f99709370657
* Woops, movem.l needs other order.Thom Johansen2007-10-18
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15178 a1c6a512-1295-4272-9138-f99709370657
* Add Coldfire assembler for window and add function, yielding a nice speedup. ↵Thom Johansen2007-10-18
| | | | | | Tweak ARM assembler version of same routine to act the same as the Coldfire version. Remove an error condition that newer ffmpegs do not bail out on. Files requiring this new behaviour have been observed. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15177 a1c6a512-1295-4272-9138-f99709370657
* Fix a problem that caused stale pcm data to be played subsequent to aStéphane Doyon2007-10-18
| | | | | | | | | | | | | | | | | | | | | | | | | voice shutup. Observed when moving through a few voiced items in very quick succession. This is for the case where music playback is not in progress, only voice is playing. The first few samples of audio data for voicing the first skipped item make it to the pcm buffer, but the shutup comes before enough pcm data has accumulated to actually start pcm playback. The condition at the top of voice_on_voice_stop() is therefore false, pcmbuf_play_stop() is not called, and the beginning of the interrupted utterance is left to wait in the pcm buffer. That data will end up prepended to the following voice clip to be played, causing a kind of stuttering effect. The fix is to remove the condition on pcm_is_playing() in voice_on_voice_stop(): always clear the pcm data, it's harmless if there wasn't any. Thanks to jhMikeS for his assistance. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15176 a1c6a512-1295-4272-9138-f99709370657
* Less stack hogging in the coldfire IDCT for mpegplayer. Speedup might or ↵Jens Arnold2007-10-18
| | | | | | might not be measurable. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15175 a1c6a512-1295-4272-9138-f99709370657
* Don't check for external rbutil.ini anymore -- doing so didn't gave a real ↵Dominik Riebeling2007-10-17
| | | | | | benefit but caused trouble instead. The file is built-in anyway. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15174 a1c6a512-1295-4272-9138-f99709370657
* Coldfire: Partially revert buffered writes, and enable them explicitly for ↵Jens Arnold2007-10-17
| | | | | | the LCD only. Turned out that with buffered writes enabled, reads after writes are not always serialized, which is a bad thing for I/O. Buffered writes to the LCD are safe because that's write-only. Fixes FS #7985. How come I didn't notice this earlier? :> git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15173 a1c6a512-1295-4272-9138-f99709370657
* Move some tables that aren't performance critical out of iram, const policeNils Wallménius2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15172 a1c6a512-1295-4272-9138-f99709370657
* Remove some dead codeDave Chapman2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15171 a1c6a512-1295-4272-9138-f99709370657
* Make a fix for the WMA seek to start bug that actually works...Thom Johansen2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15170 a1c6a512-1295-4272-9138-f99709370657
* Add myself as maintainer for a couple of things.Barry Wardell2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15169 a1c6a512-1295-4272-9138-f99709370657
* Mpegplayer: Assembler optimised motion compensation for coldfire (just the ↵Jens Arnold2007-10-17
| | | | | | variants that are assemblerised for ARM) for a nice speedup. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15168 a1c6a512-1295-4272-9138-f99709370657
* Get rid of special handling of 'The'; it's 100% language-specific and we ↵Steve Bavin2007-10-17
| | | | | | need to work towards better handling og 'The' in general. Feel free to include it in your own tagnavi_custom.config if you really need it. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15167 a1c6a512-1295-4272-9138-f99709370657
* Fix FS #7973. Hack a way to make skip to start of track work for WMA until ↵Thom Johansen2007-10-17
| | | | | | we have proper seeking. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15166 a1c6a512-1295-4272-9138-f99709370657
* Include Tracks in A to Z database search (FS#7983).Steve Bavin2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15165 a1c6a512-1295-4272-9138-f99709370657
* Try to fix the non-working $ replacement.Jens Arnold2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15164 a1c6a512-1295-4272-9138-f99709370657
* Add Karim Boucher for his initial work on the mpegplayer assembler IDCT for ↵Jens Arnold2007-10-17
| | | | | | coldfire. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15163 a1c6a512-1295-4272-9138-f99709370657
* Turns out you can boot the main build by copying apps/rockbox.bin to ↵Karl Kurbjun2007-10-17
| | | | | | /rockbox.mrboot. Something is wrong with the rockbox bootloader, but at least now we can work with the main build. Added in a warning for the core_sleep function git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15162 a1c6a512-1295-4272-9138-f99709370657
* MIDI: Scale each instrument by a calibration value, since not all patches ↵Stepan Moskovchenko2007-10-17
| | | | | | | | | | | are at the same base volume. This should make some files sound more natural, without some instruments overpowering others. Ultimately, these values need to be moved to the config file, and the table should probably be put in IRAM or something. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15161 a1c6a512-1295-4272-9138-f99709370657
* MIDI: Optimize the new pitch bending code for a small speedup.Stepan Moskovchenko2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15160 a1c6a512-1295-4272-9138-f99709370657
* Fix core_sleep for targets that have not implemented it yetKarl Kurbjun2007-10-17
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15159 a1c6a512-1295-4272-9138-f99709370657
* Mpegplayer: Convert the assembler optimised ARM IDCT into a proper assembler ↵Jens Arnold2007-10-17
| | | | | | file. Clean up the CPU selection in idct.c. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15158 a1c6a512-1295-4272-9138-f99709370657
* Apply Akio Idehara's fix for FS#7972 - Fix ARM's swp (xchg) inline assembly ↵Michael Sevakis2007-10-17
| | | | | | for gcc 4.2. Also avoids UNPREDICTABLE behavior that GCC should have always warned about. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15157 a1c6a512-1295-4272-9138-f99709370657
* Mpegplayer: Assembler optimised IDCT for coldfire, based on FS #5995 by ↵Jens Arnold2007-10-16
| | | | | | Karim Boucher. Put the IDCT block buffer in IRAM for better performance. The whole libmpeg2 decoder struct doesn't fit without throwing some libmad buffers out of IRAM, but then doesn't change performance significantly. Mpegplayer is quite usable now on X5; H300 is sort-of usable for widescreen. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15156 a1c6a512-1295-4272-9138-f99709370657
* Take out some NOCACHEBSS_ATTR's that were accidentally left. Put some ↵Michael Sevakis2007-10-16
| | | | | | threading code in IRAM that should be there on PP502x. git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15155 a1c6a512-1295-4272-9138-f99709370657
* When portable installation is finished set progress value to 100%.Dominik Riebeling2007-10-16
| | | | git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15154 a1c6a512-1295-4272-9138-f99709370657
* Mandelbrot: make increasing/decreasing iterations on X5/M5 actually usable ↵Marianne Arnold2007-10-16
| | | | | | (the buttoncombo was not possible before) and add necessary _PRE definitions to the defines and the button loop to prevent repeating the event accidentally (on some targets there was no control about how many steps one actually zoomed out on one button 'press'). Update the menu accordingly and fill out the blanks... git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15153 a1c6a512-1295-4272-9138-f99709370657
generated by cgit v1.1 at 2026-05-18 22:44:19 +0000