Onda VX747 sim was missing a limits #define; #include limits.h in
pcm_sw_volume.h.
Simply use the software volume control for the SIM volume control
rather than the SDL volume control when the target would have it
natively.
Change-Id: I8e924a2ff1b410f602452d2ea9b691efb82c931e
Implements double-buffered volume, balance and prescaling control in
the main PCM driver when HAVE_SW_VOLUME_CONTROL is defined ensuring
that all PCM is volume controlled and level changes are low in latency.
Supports -73 to +6 dB using a 15-bit factor so that no large-integer
math is needed.
Low-level hardware drivers do not have to implement it themselves but
parameters can be changed (currently defined in pcm-internal.h) to work
best with a particular SoC or to provide different volume ranges.
Volume and prescale calls should be made in the codec driver. It should
appear as a normal hardware interface. PCM volume calls expect .1 dB
units.
Change-Id: Idf6316a64ef4fb8abcede10707e1e6c6d01d57db
Reviewed-on: http://gerrit.rockbox.org/423
Reviewed-by: Michael Sevakis <jethead71@rockbox.org>
Tested-by: Michael Sevakis <jethead71@rockbox.org>
Will need it soon enough.
Combine the contents of all the various fixedpoint.h files.
Not moving fixedpoint.c for now since I'm not sure where it
should be and it causes some dependency issues.
Change-Id: Ideacbca2ca78f9158c2b114b113c274f68e908d5
grey_core.c and fixedpoint.c each had their own copies. grey_core.c
can use the ones fixedpoint.c. fixedpoint.c gets the more complete and
optimized version of fp_16exp from grey_core.c.
Change-Id: I43ee3add60796b06ed12012fbbd91815d64675a6
The new autodetection will always use USB IDs to identify players (among other
means). Add missing IDs for players the current implementation relies on other
means for detection only.
Change-Id: I037c23705518210b6ed93b4d13ccb96d933aa198
Both autodetection functionality and the configuration dialog assumed detection
to only return one found device. This isn't necessarily true, especially since
some players can be detected but detecting their mountpoint might be ambiguous
(only if no previous Rockbox installation is present).
Instead of returning individual results (found "ok" player, found "error"
player etc.) return a list containing an entry for each player. Current
autodetection code will never return more than one entry since it doesn't
handle multiple devices yet, and the configuration dialog will show an error if
multiple devices are found. Thus there is no user visible change yet. Both
autodetection and configuration dialog can now get extended to handle multiple
devices.
Change-Id: I79b763dbd6e7111783194bcc22ab7cc06a4061c1
When autoload bookmark is set to "Ask", navigate to a file with
bookmarks in that directory. Select the file and the bookmark list
appears. Even if you chose to cancel, the track started prior to this
patch.
Change-Id: I453999a9bc20faae97f9cf2080ef613c602ad8e1
Reviewed-on: http://gerrit.rockbox.org/416
Reviewed-by: Jonathan Gordon <rockbox@jdgordon.info>
Using uname has a couple of problems, especially when cross compiling. Instead
check the defines set by the preprocessor to figure the type of binaries it
produces. This improves support for cross compiling as it allows to (1) select
the correct default target and (2) makes it possible to use separate build
folders for different targets.
Change-Id: I69a32904dab97755034f2f0d63f8402309d479d2
Current MinGW #defines interface, causing the preprocessor messing up the
function parameter. Rename it to avoid this.
Change-Id: Id49edbfc056a42456d88c9e260e54844e8b1ca07
Previously it was sufficient to set those variables when running configure.
Since ANDROID_NDK_PATH is now also used in the Android Makefile set both
variables in the generated Makefile as well -- not everyone has (or wants) to
set them globally in the environment.
Change-Id: Ib64ffeea379c58c9d361b3fbfefb67084dce666c
The target all is set in libtools.make. Setting it here breaks make knowing
about the correct default target when cross compiling.
Change-Id: Icdd022094d28e4dcd442e48bb6acd1fd8f9a6ad6
Instead of trying every mountpoint during autodetection allow filtering out
filesystems that are not supported when retrieving the system mountpoints.
Change-Id: Ic23a5c804cb7c78c146dbc1af7443c67ce12464e
In the modern everything-can-be-skinned world, themes are beginning
to hit the limit of 8 fonts (which includes sysfont). Increasing this
to 12 will allow some headroom, at the cost of a bit more RAM usage
(about 20 bytes per extra possible font, so about 80 bytes total here)
Change-Id: Iddf4374b7ccf92b400614a7309bf3d2147ba98ab
The old way actually mis-used the API (I misunderstood the docs) because
it specified the marker position as a "low buffer watermark" but instead of a
future playback head position.
The replacement is a simple thread that writes the data regardless of the
filling level of the buffer (write() will just block) and polls the playback
state periodically.
Change-Id: If29237cee4ce78dc42f5a8320878bab0cafe78f7
Reviewed-on: http://gerrit.rockbox.org/422
Tested-by: Dominik Riebeling <Dominik.Riebeling@gmail.com>
Reviewed-by: Thomas Martitz <kugel@rockbox.org>
Correctly descramble the first stages (0 to 2) in continuous mode.
Also fix a disassembled (but unused) crc routine used in dfu mode.
Change-Id: I20016d1c696a9bcb6584377ee9b55493783c7159
This tool can upload a firmware to the device in DFU mode. The
protocol is the same as the rk27xx devices except that it can
load a bigger (unlimited ?) firmware.
Change-Id: Ic9d4c5087629a9156f9d5d5cdc80767e6359c431
There are two reasons for this:
- QUrl::toEncoded() has problems with some characters like the colon and @.
Those are not percent encoded, causing the string getting parsed wrongly when
reading it back (see FS#12166).
- The password is cleartext in the configuration file.
While using base64 doesn't provide any real security either it's at
least better than plaintext.
Since this program is open source any fixed mechanism to obfuscate /
encrypt the password isn't much help either since anyone interested in
the password can look at the sources. The best way would be to
eventually use host OS functionality to store the password.
Change-Id: I6ac49d68211236e540b6ca16481e0e1c196532b7
If the host is ARM simply checking the architecture the compiler produces
binaries for will include unwarminder even for the PC tools, breaking them.
Fixes building checkwps and database on ARM hosts. Doesn't fix warble, there
are additional problems with that.
Change-Id: I135b1103a60b9e3720117cde1075b7d73424f384
This was spotted while playing with qemu-jz:
1) rockbox reads TECR and TESR which are described as write-only
registers. Datasheet doesn't mention what happens if they are
readed. Apparently this doesn't have fatal side effects.
It comes down to two defines from jz4740.h
__tcu_stop_counter(n) and __tcu_start_counter(n) which use
read-modify-write sequence.
2) rockbox accesses out of bound offset 0xd4 in DMA memspace.
It comes from dis_irq() in system-jz4740.c. NUM_DMA is 6 but
DMA channels are 0-5 so (irq <= IRQ_DMA_0 + NUM_DMA)) bound
check is wrong.
This are *NOT* tested on device.
Change-Id: I29dff6a4f828030877b7d50fbcc98866478b9e3d
Reviewed-on: http://gerrit.rockbox.org/338
Reviewed-by: Bertrik Sikken <bertrik@sikken.nl>
Tested-by: Purling Nayuki <cyq.yzfl@gmail.com>
Reviewed-by: Marcin Bukat <marcin.bukat@gmail.com>
Ideally this commit should go on top of the v3.13 branch
since the release is based on the v3.13 code base.
Change-Id: If13cf87463c7a3f7a9c2c96c916c543c9d2ed1a5
When creating a new HttpGet object schedule any old one for deletion instead of
dropping its reference and relying on Qt to clean it up once the parent object
is removed.
Change-Id: I66f2a6f01c6b6ad3a0dce0eb4fd3eacfaf7e733d
Schedule them for deletion instead of deleting them. That way pending events
can still be handled before the main loop deletes them. Fixes crashes on the
second download during installation.
Change-Id: I7c718dbb0200a64d1aa3bbb6c2839282fb965dec
SAPI only allowed values from 1.0 to 10.0, making it impossible to turn down
the volume of the voice. Increasing the volume by a factor of 10.0 is nothing
that is likely to be useful, so change the upper limit to 2.0 and decrease the
lower limit to 0.0.
Lame allowed values from 0.0 to 1.0, making it impossible to increase the
volume of the voice. Change the upper limit to 2.0 as well.
Change-Id: I8add103f6e4b3c8f1b11ee2c0ea478727bdc99c1
Prevents cutoff of tracks, especially short ones:
* Extend looped tracks by fade length to fade at start of loop repeat.
* No fade occurs for non-repeating track only having an intro.
* Uses id3.tail_trim field to store fade duration.
Use libGME built-in elapsed time reporting instead of custom calculation:
* libGME already reports in milliseconds.
* Don't advance time counter when Repeat == One. It just runs the progress
over the length limit.
Fix a comment about sample rate and set the reported bitrate to be
accurate for 44.1 kHz stereo.
Change-Id: I3ede22bda0f9a941a3fef751f4d678eb0027344c
This patch adds to YP-R0 (and other future targets using Linux
framebuffer) the ability to use LCD_ENABLE to save some CPU cycles
while display is powered off.
This patch also changes the way to toggle LCD power: now using
a proper ioctl call, slightly more efficient.
Change-Id: I544de77f5abd4ac1c13d3fe3a6e40a30f7c0bece
Reviewed-on: http://gerrit.rockbox.org/410
Reviewed-by: Thomas Martitz <kugel@rockbox.org>
