Age | Commit message (Collapse) | Author |
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16855 a1c6a512-1295-4272-9138-f99709370657
|
|
mutex ownership and queue_send calls are inheritable). Priorities are differential so that dispatch depends on the runnable range of priorities. Codec priority can therefore be raised in small steps (pcmbuf updated to enable). Simplify the kernel functions to ease implementation and use the same kernel.c for both sim and target (I'm tired of maintaining two ;_). 1) Not sure if a minor audio break at first buffering issue will exist on large-sector disks (the main mutex speed issue was genuinely resolved earlier). At this point it's best dealt with at the buffering level. It seems a larger filechunk could be used again. 2) Perhaps 64-bit sims will have some minor issues (finicky) but a backroll of the code of concern there is a 5-minute job. All kernel objects become incompatible so a full rebuild and update is needed.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16791 a1c6a512-1295-4272-9138-f99709370657
|
|
lcd_blit_mono() for colour targets. Plugin API became incompatible, so sort, clean up & bump. * Implemented lcd_blit_mono() for M3.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16775 a1c6a512-1295-4272-9138-f99709370657
|
|
devices replicate app.lds since their target tree doesn't follow a syntax typical to the newer arm targets - the portalplayers could be cleaned up further. boot.lds and plugin.lds still need to be cleaned up.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16651 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16639 a1c6a512-1295-4272-9138-f99709370657
|
|
Keep fat lock access from compiling for Ondios - think of a nicer way later.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16637 a1c6a512-1295-4272-9138-f99709370657
|
|
use and use before initialization. For c200/e200: Be sure fat cache and ata locks are acquired in the proper order during hot swapping. Delay hotswap monitoring until after initial file mounting (address 2nd kobj concern + possible call of fat driver before init).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16636 a1c6a512-1295-4272-9138-f99709370657
|
|
in case two SYS_HOTSWAP_INSERTED messages are received in a row. Be sure flags are reset to re-initialize the SD-card. Go back to locking out driver access during disk mount/unmount attempts.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16635 a1c6a512-1295-4272-9138-f99709370657
|
|
insert/remove events while filesystem access is blocked. Also use atomic GPIO bit manipulation where possible, and protect the interrupt level manipulation in the ISR from interrupts arriving at the same time.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16592 a1c6a512-1295-4272-9138-f99709370657
|
|
usb_wait_for_disconnect() is gone.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16452 a1c6a512-1295-4272-9138-f99709370657
|
|
driver uses the sd driver, and wants to know about changes
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16449 a1c6a512-1295-4272-9138-f99709370657
|
|
callbacks to get done faster). Use a simpler loop for ata thread. Add a balancing mutex_lock call to ata_init (oops).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16278 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16256 a1c6a512-1295-4272-9138-f99709370657
|
|
Sansa E series.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16122 a1c6a512-1295-4272-9138-f99709370657
|
|
where it becomes a reenterable corelock.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16105 a1c6a512-1295-4272-9138-f99709370657
|
|
initialized the lock on every call to ata_init - that should be a one time init or else the lock could be corrupted on connect.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16104 a1c6a512-1295-4272-9138-f99709370657
|
|
make it consistant with the ipods. apart from removing a bit of confusion, it fixes the wheel in pictureflow and possibly elsewhere.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16045 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16038 a1c6a512-1295-4272-9138-f99709370657
|
|
to 16kB). Sansa ata driver also didn't yield enough so buffering would starve other threads. Bump priority of audio thread to keep UI more responsive.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15724 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15681 a1c6a512-1295-4272-9138-f99709370657
|
|
Gen, Mini 1st Gen: >3.3x speedup, 1st..3rd Gen: ~2x speedup, Mini 2nd Gen: ~30% speedup).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15655 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15600 a1c6a512-1295-4272-9138-f99709370657
|
|
the software PWM fade in/ fade out. * Backlight handling cleanup, getting rid of one layer of 'lowlevelness'. * Use atomic GPIO bit manipulation for PP502x backlight handling.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15599 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15593 a1c6a512-1295-4272-9138-f99709370657
|
|
data and buffers on PortalPlayer processors to cache line boundaries. They're noops when PROC_NEED_CACHEALIGN isn't defined. Go safe and increase the value to 32 since I'm not sure yet if 16 is sufficient - changing that is a one-liner. Add helper to plugin API which will be needed shortly.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15523 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15493 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15477 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15476 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15474 a1c6a512-1295-4272-9138-f99709370657
|
|
the awkward *_nolock stuff.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15468 a1c6a512-1295-4272-9138-f99709370657
|
|
versions, so move those definitions into pp5020.h, and add the missing group. Make microSD hotswap detection work on Sansa c200. Note that microSD access itself does not work yet.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15460 a1c6a512-1295-4272-9138-f99709370657
|
|
updates (was only visible in mandelbrot).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15343 a1c6a512-1295-4272-9138-f99709370657
|
|
to have a safe margin.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15337 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15297 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15246 a1c6a512-1295-4272-9138-f99709370657
|
|
includes shifting the readings by 0x14 when PLL is enabled. Battery voltages and remote/scrollpad values adjusted accordingly. We also now wait for the ADC to indicate the conversion is complete before reading the data, so hopefully we will get more reliably steady readings.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15141 a1c6a512-1295-4272-9138-f99709370657
|
|
towards the possibility of other types. All SVN targets the low-lag code to speed up blocking operations. Most files are modified here simple due to a name change to actually support a real event object and a param change to create_thread. Add some use of new features but just sit on things for a bit and leave full integration for later. Work will continue on to address size on sensitive targets and simplify things if possible. Any PP target having problems with SWP can easily be changed to sw corelocks with one #define change in config.h though only PP5020 has shown an issue and seems to work without any difficulties.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15134 a1c6a512-1295-4272-9138-f99709370657
|
|
Immeasurable speedup on iPod Color, huge speedup on small H10 (a factor of 3). Should be a bit faster on big H10 too. * Big H10 changed bitmap format, so needs reconfiguring + full rebuild. * Better register naming for the mono LCD bridge. Register names for the colour LCD bridge.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15082 a1c6a512-1295-4272-9138-f99709370657
|
|
removing all hotswap support from the bootloader. It won't fully be closed until Sansapatcher is updated of course. There's still a report of the partition dump not working unless 20 bytes of padding are inserted at the start of main() but that's only a developer-enabled feature. Seems weird though.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15038 a1c6a512-1295-4272-9138-f99709370657
|
|
can be optimized more and dithering is not yet implemented.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15020 a1c6a512-1295-4272-9138-f99709370657
|
|
bit.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15018 a1c6a512-1295-4272-9138-f99709370657
|
|
buttonlight_ .
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15016 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15015 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15014 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@15012 a1c6a512-1295-4272-9138-f99709370657
|
|
intent.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14982 a1c6a512-1295-4272-9138-f99709370657
|
|
with one. Kill all the now unneeded cache flushing and i2s tweaking on e200 so clicking is no worry. Write the driver framebuffer at uncached addresses. Recording monitoring may be a little noisy in the left channel for the moment when not boosted and will be addressed. All seems to work as advertised including flash ROM dump.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14976 a1c6a512-1295-4272-9138-f99709370657
|
|
that many (if not all) c200s have the radio hardware even if you can't use it with the original firmware...
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14963 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14945 a1c6a512-1295-4272-9138-f99709370657
|
|
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@14877 a1c6a512-1295-4272-9138-f99709370657
|