It was possible for interrupts of higher priority than the current IRQ
level to attempt to restart the interface while it was still active on
a transfer. The list modification also wasn't protected within the I2C
ISR itself.
Change-Id: I70635c307a1443bba6801c588cf1efde299db9a4
The solution is a bit hacky as it simply call make in libs
directory as pre-dependency. Clean doesn't touch libs.
Change-Id: Ib447a48fd87cc41228944f17444474a55d393543
It's not needed as picture flow has it's own buffer.
This reverts commit 9076b433d1.
Detailed explanation from Thomas Martiz (thanks!):
buflib buffers can be passed to yielding functions just fine. Problems
only arise if the are concurrent allocations, for example if two threads
allocate from the same context simultaneously or if the callee does it's
own allocations. This can't happen in the pictureflow case, it has it's
own context and a single thread allocating from it.
Therefore the problem isn't yield() itself, but possible concurrent
buflib_alloc() calls that result from the thread switch. This is because
compaction only ever happens on allocation (and not in a backgroud
thread or so).
This patch enables "Settings -> Sound Settings -> DAC's filter roll-off" for the
iBasso DX50.
Confirmed working on iBasso DX50, firmware 1.6.
Change-Id: I901615541d8fa4f6550c5156cf8c6069f5bdf9d0
- square sine tick and tock sounds (more annoying, more useful;-)
- optical indication of tics on display
- unification of mode of operation for SWCODEC and HWCODEC (tested on simulator)
Both playback and display drawing happen in main loop, always.
- operating in two modes now:
-- 1. classic dumb metronome
--- active when openened as application without file to open
--- the usual functionality with tapping and bpm change
--- controls indicated on display
-- 2. track mode with programmable series of parts
--- active when started as viewer for a .tempo file
--- differing meters (4/4, 3/4, 6/8, etc.)
--- patterns (tick/tock/silence on each beat)
--- smooth tempo changes in those tracks
This version had lots of testing regarding metronome accuracy,
resulting in the realization that PLL A and PLL B differ
on the Clip+, causing drift. There is still drift when the timer
intervall is too small, so I settled on 2 ms as compromise.
This is the final version, after adding documentation and extensive
help from Sebastian Leonhardt testing it on slower hardware (YH820),
where it works up to 650 actual bpm with display indication.
Latest change: Documentation nitpicks.
Change-Id: I764c8252526db188352385c5462f9453d882beb9
Depends on http://gerrit.rockbox.org/r/#/c/1043/.
This patch adds a new setting in Settings -> General -> System: USB Mode.
Usable in Quickscreen and Shortcuts.
Possible settings are:
- Mass Storage: The default, on USB connect export the internal and external
drives as usual. Rockbox will exit gracefully in this case, since the internal
drive (/mnt/sdcard) is exported as mass storage device and Android prevents
Rockbox from continued execution.
- Charge Only: USB charge only, do not export the internal and external drives.
This will not close Rockbox.
- ADB: Enable the Android Debug Bridge. This will not close Rockbox.
Charge only and ADB are actually the same, since it is not yet established if
charge only is doable without adb and major hooks into Android.
German translation provided.
This may be genric for Android based devices but is only enabled
for iBasso Devices. Other maintainers may choose do adopt this.
Change-Id: I616247c29946c6595ebcf8f0c14b2410c9f0498b
Depends on http://gerrit.rockbox.org/r/#/c/1043/.
This patch adds a new setting in Settings -> General
-> System: Freq Scaling Governor
Usable in Quickscreen and Shortcuts.
Possible settings are:
- Conservative: Slow frequency switching.
- Ondemand or Interactive: Fast frequency switching.
- Powersave: Allways lowest frequency.
- Performance: Allways highest frequency.
German translation provided.
This may be genric for Android kernel based devices but is only enabled
for iBasso Devices. Other maintainers may choose do adopt this.
Change-Id: I10296f5be9586ad3a409105db0cd03682a30e9c1
Reorganization
- Separated iBasso devices from PLATFORM_ANDROID. These are now standlone
hosted targets. Most device specific code is in the
firmware/target/hosted/ibasso directory.
- No dependency on Android SDK, only the Android NDK is needed.
32 bit Android NDK and Android API Level 16.
- Separate implementation for each device where feasible.
Code cleanup
- Rewrite of existing code, from simple reformat to complete reimplementation.
- New backlight interface, seperating backlight from touchscreen.
- Rewrite of device button handler, removing unneeded code and fixing memory
leaks.
- New Debug messages interface logging to Android adb logcat (DEBUGF, panicf,
logf).
- Rewrite of lcd device handler, removing unneeded code and fixing memory leaks.
- Rewrite of audiohw device handler/pcm interface, removing unneeded code and
fixing memory leaks, enabling 44.1/48kHz pthreaded playback.
- Rewrite of power and powermng, proper shutdown, using batterylog results
(see http://gerrit.rockbox.org/r/#/c/1047/).
- Rewrite of configure (Android NDK) and device specific config.
- Rewrite of the Android NDK specific Makefile.
Misc
- All plugins/games/demos activated.
- Update tinyalsa to latest from https://github.com/tinyalsa/tinyalsa.
Includes
- http://gerrit.rockbox.org/r/#/c/993/
- http://gerrit.rockbox.org/r/#/c/1010/
- http://gerrit.rockbox.org/r/#/c/1035/
Does not include http://gerrit.rockbox.org/r/#/c/1007/ due to new backlight
interface and new option for hold switch, touchscreen, physical button
interaction.
Rockbox needs the iBasso DX50/DX90 loader for startup, see
http://gerrit.rockbox.org/r/#/c/1099/
The loader expects Rockbox to be installed in /mnt/sdcard/.rockbox/. If
/mnt/sdcard/ is accessed as USB mass storage device, Rockbox will exit
gracefully and the loader will restart Rockbox on USB disconnect.
Tested on iBasso DX50.
Compiled (not tested) for iBasso DX90.
Compiled (not tested) for PLATFORM_ANDROID.
Change-Id: I5f5e22e68f5b4cf29c28e2b40b2c265f2beb7ab7
This loader is needed to use Rockbox on iBasso DX50/DX90 devices. It replaces
/system/bin/MangoPlayer.
It expects
- the original firmware MangoPlayer as /system/bin/MangoPlayer_original
- the chooser.bmp file as /system/chooser.bmp
- the rbmissing.bmp file as /system/rbmissing.bmp
- the usb.bmp file as /system/usb.bmp
- Rockbox in /mnt/sdcard/.rockbox/ (internal device storage accessable via USB)
On first boot or if the device is booted with the hold switch engaged (upper,
locked position) this loader will let the user choose between Rockbox and the
orginal firmware MangoPlayer. It remembers the last choice and will start the
choosen player an following boots.
In the chooser screen
- the device can be powered off with a power button long press.
- Rockbox can be startet with the "Next/Fast Forward" button or by touching the
right part of the touchscreen.
- the orginal firmware MangoPlayer can be startet with the "Previous/Rewind"
button or by touching the left part of the touchscreen.
If Rockbox can not be found/executed than a "Rockbox is missing" screen will be
shown and the orginal firmware MangoPlayer will be started after 30 seconds.
This loader will (try to) restart Rockbox or the orginal firmware MangoPlayer if
something unexpected happens.
This loader can be installed via a customized firmware image (update.img) or
manually via the Android Debug Bridge (if you get adbd started on the device).
This is a rewrite and code clean up of the original submission, eliminating the
need for a custom Android Vold. This loader "knows", when Rockbox has exited due
to a USB mass storage connection remounting /mnt/sdcard and acts accordingly,
displaying a "USB connection" screen while waiting to restart Rockbox.
Change-Id: I2698e173437f9c7aa1ff40649a290220e2ee0439
System -> Debug (Keep Out) -> View CPU stats
Will now show the current cpufreq scaling governor, minimum, current and
maximum cpufreq scaling frequency for each CPU.
This may be genric for Android kernel based devices but is only enabled
for iBasso Devices. Other maintainers may choose do adopt this.
Change-Id: I53e212f8707bf2abaa557e297293fb559ac37058
The GPIO APIs for ypr0 and ypr1 targets was messy, requiring a
direct communication via several ioctls calls.
Since it is planned to add support to other devices, more GPIO are
going to be used. For that reason the functions shall be clear and
easy to use.
Change-Id: Ia2304335e1fed1305cc2c4320bd4c097e13079be
Some people reported strange charging times and
strange battery life.
Charging by OF: 25 hours; RB: 18 hours
It has been found that there are at least two issues here:
1) the way of getting battery charging status wasn't
really accurate. This attempts to fix that issue.
This patch also simplifies some code (opening a
device is no more needed, for example).
To technically explain, battery charging implies first a constant
current mode (where the voltage increases) and then a constant
voltage mode (where, obviously, the voltage reads more or less
the same). The old way the End Of Charge was detected was based solely
on the voltage, while now it is based on chip's EOC interrupt,
which should be more accurate.
2) OF explicitly sets a constant current 350 mA, while we usually had
55 mA (by as3543 default).
This wasn't discovered before since there is a caching problem
("an accurate guess") in the Samsung power mgmt Linux module,
thus its debugging dumper wasn't really working, reporting a 55 mA current.
Strangely this option should have been set by the bootloader
but apparently it does not.
Some testing is still needed, but I confirm that with this patch
I could run a benchmark for 25 hours (vs. 18 hours), by charging
within Rockbox of course.
Change-Id: I3bd921e86b9018d1cc3c720d15cc46896e8490b3
Catched by swiss_knife check there are some overlaping fields
and invalid characters in some names.
Change-Id: Ia26ffd2e29452f4ddd9f8229f78bb2a2cc325ab4
surround_enabled was never true, end up dsp_surround_flush didn't work; Thats why a cracking noise occurs in right channel when moving track positions.
redo pbe/surround flush in a much simpler way suits the current single buffer style.
Change-Id: I394054ddfb164b82c90b3dcf49df4442db87d8d2
perceptual bass enhancement
- a bbe-ish group delay corrction with Biophonic EQ boost.
- precut
auditory fatigue reduction
-reduce signal in frequency that may trigger temporary threshold shift
haas surround
-frequency between f(x1) and f(x2) is always bypassed.
-can apply to side only.
Change-Id: Icb6355ce9b1c99bf2c58c9385c3c411c0ae209d3
The WRITE() macro checks if the "fw" pointer
is open upon fwrite() error. Since fw was
not initialized before use, it might crash for non-zero fw.
cppcheck reported.
[utils/jz4740_tools/HXFreplace.c:187]: (error) Uninitialized variable: fw
Change-Id: I558170f9f17828c9fd515a5da36ad8b5cb30a9ad
It will be out of scope once the function is left.
cppcheck reported:
[firmware/drivers/rtc/rtc_zenvisionm.c:31]: (error) Pointer to local array variable returned.
[firmware/drivers/rtc/rtc_zenvisionm.c:38]: (error) Pointer to local array variable returned.
Change-Id: Ibf28ba9b3d20cadcaff22398e143488c86746660
The origin of the register value was never
moved in the desired register state due to a typo ('rhs' vs. 'rhd').
While looking at the code, I noticed the action taken
for the register value is another copy'n'paste error
from the ADD opcode above -> it added to the register value
instead of MOVing the current value.
Patch submitted upstream.
cppcheck reported:
[lib/unwarminder/unwarm_thumb.c:473]: (warning) Redundant assignment of 'state.regData[rhd].o' to itself.
Change-Id: I78cdbf37a191007a3bddbaa350b906dbce2fe671
