This brings puzzles up-to-date with upstream revision
2d333750272c3967cfd5cd3677572cddeaad5932, though certain changes made
by me, including cursor-only Untangle and some compilation fixes
remain. Upstream code has been moved to its separate subdirectory and
future syncs can be done by simply copying over the new sources.
Change-Id: Ia6506ca5f78c3627165ea6791d38db414ace0804
I'm not sure all the situations it affects, to be honest. The fix
aimed to address the strange symptom here:
http://forums.rockbox.org/index.php/topic,50793.0.html
It turns out that ringbuf_add_cross was used when handles were
butted up against one another with the first parameter equal to
the last, which it interprets as being an empty case when it should
be interpreted as full in the context it was used. To fix this,
introduce full/empty variants of ringbuf_add_cross and ringbuf_sub
and use them at the appropriate time.
The other way to address the problem is ensure there's always at
least a space byte between the end of one handle and the start of
another but this make the code a bit trickier to reason about than
using additional function variants.
bufopen() may yield after creating a handle and so do some more
locking so that the buffering thread doesn't mess things up by
moving anything or not seeing the yet-to-be linked-in allocation.
Add alignof() macro to use proper method to get alignment of
struct memory_handle. That should be useful in general anyway.
It's merely defined as __alignof__ but looks nicer.
Change-Id: If21739eaa33a4f6c084a28ee5b3c8fceecfd87ce
Forced audio start was left out when a third codec attempts to
start a second track transition. Only one pending transition is
allowed at a time. There wouldn't be enough PCM in the buffer to
trigger audio playback and audio would just return without giving
the pcm buffer a kick.
Fixes FS#13100 - Player failed on short tracks
Change-Id: I338b0b12022c591930451fd5ed26a2a73008835f
It seems to like not having any idea any attempt was made to load
it so just try to put it back like the option was never on.
Integrity check other tags besides the filename.
Fix a few logfs that spat out warnings and add a bit more detail
to a couple.
Move a buffer lock to the right spot.
Better aligning macro.
Add a "continue" that should have been there when filename is
seeked over during load.
Change-Id: Ib0c517a7b77d3ba0dd6530ddef19bed62116e4a8
cppcheck reported:
[chessbox/chessbox_pgn.c:51]: (portability) 'bufptr' is of type 'void *'. When using void pointers in calculations, the behaviour is undefined.
Change-Id: Ib2139489fa316fb61b4bcdc08f17b97eb6ad0be7
Many includes of fat.h are pointless. Some includes are just for
SECTOR_SIZE. Add a file 'firmware/include/fs_defines.h' for that
and to define tuneable values that were scattered amongst various
headers.
Remove some local definitions of SECTOR_SIZE since they have to be
in agreement with the rest of the fs code anyway.
(We'll see what's in fact pointless in a moment ;)
Change-Id: I9ba183bf58bd87f5c45eba7bd675c7e2c1c18ed5
* Take out pointless dircache_search; nothing can be reconstructed
with the given info in find_entry_ram(); don't even try there.
Path AND index id must be known. Work it out later.
* Timed yield must be far more often than once every 1/4 second (?!)
* Do better the memory-remaining checks for ramcache load.
* Root separator mustn't be doubled up when searching files.
Change-Id: I091813f4495f3bd0d0c4672bc674df52343b3e48
- unregisters timer on exit, preventing possible crash
- disables synchronization mechanisms when used from an IRQ
- prevents memory allocations from overflowing the audio buffer (unlikely)
Change-Id: I3c2c4ebe93c10ca9176ed0455e7aacc2d10c059e
* Remove unused bits like the radio event and simplify basic
radio interface. It can be more self-contained with rds.h only
required by radio and tuner code.
* Add post-processing to text a-la Silicon Labs AN243. The chip's
error correction can only do so much; additional checks are highly
recommended. Simply testing for two identical messages in a row
is extremely effective and I've never seen corrupted text since
doing that, even with mediocre reception.
Groups segments must arrive in order, not randomly; logic change
only accepts them in order, starting at 0.
Time readout was made a bit better but really we'd need to use
verbose mode and ensure that no errors were seen during receiving
of time and more checks would be need to have a stable PI. The
text is the important bit anyway.
* Time out of stale text.
* Text is no longer updated until a complete group has been
received, as is specified in the standard. Perhaps go back to
scrolling text lines in the radio screen?
* Add proper character conversion to UTF-8. Only the default G0
table for the moment. The other two could be added in.
* Add variants "RDS_CFG_PROCESS" and "RDS_CFG_PUSH" to allow
the option for processed RDS data to be pushed to the driver and
still do proper post-processing (only text conversion for now for
the latter).
Change-Id: I4d83f8b2e89a209a5096d15ec266477318c66925
I noticed that after booting with the external storage removed,
playing from tagtree, inserting the card, forcing dircache to
reallocate from the debug screen, and trying to reenter tagtree,
it would data abort because the dircache reallaction to a larger
size caused the tagtree buffer to move.
Adjustment to at least one pointer (csi) was missed. Since it's
non-trivial there to determine when things should be NULL and
when they shouldn't, add check for menu too before moving it.
As for the rest, who knows.
Change-Id: Iea6538a2091b4b47083f39296555efc47edf8ba8
On Windows 64-bit, the size of long is 32-bit, thus any pointer to long cast is
not valid. In any case, one should use intptr_t and ptrdiff_t when casting
to integers. This commit attempts to fix all instances reported by GCC.
When relevant, I replaced code by the macros PTR_ADD, ALIGN_UP from system.h
Change-Id: I2273b0e8465d3c4689824717ed5afa5ed238a2dc
The implementation is not very complicated but there are a few things worth
noting. There was a previous "speaker enable" setting but it was a boolean.
I decided to replace it with a choice setting that has 2 options (on, off)
if headphones cannot be detect on this target, or 3 options (on, off, auto)
if we can detect headphones. This will break the old setting on target that
cannot detect jack but it makes the code more uniform and avoid maintaining
two settings with more #ifdef. The third option (auto) uses the LANG_AUTO
text, which I think is clear enough (disable speaker on jack plug).
In order to avoid code duplication (both in apps and firmware), I decided to
keep the audiohw_enable_speaker function as-is: it takes a boolean and doesn't
care about the speaker policy. I introduced a new audio_enable_speaker that
takes directly the mode (which follows the setting encoding): 0=off, 1=on
and 2=auto. This way one calls audio_enable_speaker and it changes the speaker
once to reflect the request mode. The apps code then uses this function in the
places where it makes sense: on setting load, setting change and jack (un)plug
event.
Change-Id: I027873f698eb4bc365d7c02b515297806355d9e2
Chessbox was overflowing GameList[240] causing the board to flip + crash
GameCnt changed to unsigned char which allows the array to roll over
to 0 after 255
define MAX_GAME_CNT 256 and GameList[MAX_GAME_CNT] along with 1 byte GameCnt
should fix this issue
dbg save routine left in for now to help identify any other problems
Added bounds checking to prevent second bug found when loading .pgn files
Change-Id: I2b615c8ecbed4368724412f80ce07346f3cf30a7
1. Slightly revised and regularized internal interface. Callback is used
for read and write to provide completion signal instead of having two
mechanisms.
2. Lower overhead for asynchronous or alterate completion callbacks. We
now only init what is required by the transfer. A couple unneeded
structure members were also nixed.
3. Fixes a bug that would neglect a semaphore wait if pumping the I2C
interrupts in a loop when not in thread state or interrupts are masked.
4. Corrects broken initialization order by defining KDEV_INIT, which
makes kernel_init() call kernel_device_init() to initialize additional
devices _after_ the kernel, threading and synchronization objects are
safe to use.
5. Locking set_cpu_frequency has to be done at the highest level in
system.c to ensure the boost counter and the frequency are both set in
agreement. Reconcile the locking inteface between PP and AMS (the only
two currently using locking there) to keep it clean.
Now works fine with voltages in GIT HEAD on my Fuze v2, type 0.
Previously, everything crashed and died instantly. action.c calling
set_cpu_frequency from a tick was part of it. The rest may have been
related to 3. and 4. Honestly, I'm not certain!
Testing by Mihail Zenkov indicates it solves our problems. This will
get the developer builds running again after the kernel assert code
push.
Change-Id: Ie245994fb3e318dd5ef48e383ce61fdd977224d4
The function is neither reentrant nor ISR callable. Instead of
using a ticked-based timeout, have the button driver provide the
unboost after a delay when waiting for a button.
HAVE_GUI_BOOST gets immediate boost after dequeuing any message,
otherwise the queue has to have at least three messages waiting
for it to trigger a boost-- essentially the behavior that existed
but now combined in one place.
Change-Id: I1d924702840f56a1a65abe41fa92b4e753c4e75a
Selective backlight allows the user to choose actions that will not
enable the backlight when pressed.
Advanced softlock allows user to choose actions that will not be
blocked by screenlock on devices without a hold button.
Both only occur in FM and WPS Contexts.
Update:
Back from the dead
-Cleaned up code, removed unnecessary calls, re-arranged last filter action
timeout conditional to work in case last_filtered_action_tick was never set
-Added entries to the manual
-Fixed back button on some menus not activating backlight
-Made menus more intuitive, no actions selected now changes menu item to off.
-Added talk fuctionality.
-Added option to disable selective backlight while on external power.
-Rewrote backlight and softlock handling code to fix issue with scrollwheels
-Menu changed to have toggle(yes/no) and settings
-Optimized selective actions lookup
-Added option to disable notification of 'buttons locked' while softlocked
-Removed uneeded code, consolidated action lookup to single function
-Fixed incorrect name on selective softlock menu
-Added option to disable touch on touchscreen devices
-Fixed backlight on original screenlock without selective screenlock active
-Added text selection in mask_select for when show_icons is off
-Fixed voice in mask_select to speak if voice is defined instead of spelling
-Added more lang defines (play skip seek)
-Added option to disable unknown keys turning on backlight
-Fixed Conditional argument In wrong place causing players without
backlight to fail to build
-Fixed Disable Unknown blocking detection of context change
-Fixed canceling menu didn't update new settings
-Added Autolock on backlight off
-Removed backlight_on_force from backlight.c, Now sets ignore next to false
and uses backlight_on
-Cleaned up autolock code added strings to lang file
-Fixed issue where rapid presses would bypass softlock
-Removed old softlock code, Cleaned selective actions code
-Changed menu to match existing RB menus
-Fixed Backlight_on_Hold blocked by backlight_ignore_next
-Fixed ignore_next for ipod
-Fixed bug allowing context with softlock to bypass selective backlight
-Changed mask_select to no longer prompt for changes to be saved
-Changed menu names
-Added ignore timeout to allow ipod scroll wheel to work properly and other
players to still work properly, removed some previous code including
ignore_event
-Increased ignore timeout to prevent sd card accesses from interrupting action
code and turning on backlight
-Changed Unknown action to unmapped action in menu, changed handling code
-Removed unneeded logic and variables for handling unfiltered actions
-Reverted unmapped action code to previous functionality
-Added manual entries (thanks JohnB)
-Removed elusive unhandled unicode character from manual, changed formatting slightly
Actions:
Volume,Play,Seek,Skip
Extras:
Disable unmapped actions
Disable selective backlight on external power
Disable touch during softlock on touchscreen devices
Disable softlock notifications (power button still notifies)
Autolock on backlight off
Method:
Adds a function to ignore backlight on next call
If selected action occurs backlight is forced on,
Filter_first_keypress stays intact.
Selective softlock allows selected actions through, bypasses the normal
softlock routine.
ToDo:
DONE
previous commit (#1) has attribution for folder_select.c which mask_select
is based from.
Change-Id: I08132ddcfd64c81751ef23b720f3ec6d68695fe4
Playlist dircache references should be back in working order.
Reenabling dircache references in the database ramcache is not
yet done as it requires quite a bit of rework. Otherwise, the
database in RAM is functional again.
Some buffer compatibility changes have been made for database
commit because the dircache buffer can no longer be stolen, only
freed by an API call.
Change-Id: Ib57c3e98cb23e798d4439e9da7ebd73826e733a4
defined in mingw environments.
Renamed defines of UNALIGNED to ROCKBOX_UNALIGNED so that they don't
conflict with definitions in mingw32 cross-compiling environments
(defined in _mingw.h).
Change-Id: I369848c0f507e6bf5ff9ab4a60663bbbda6edc52
- line drawing should eventually be moved to xlcd, but for now it's very unportable code
- also fixes a minor issue with the configuration screens
Change-Id: I897f01b7210cbbec32665c2bc67920c965ea0bec
- line drawing should eventually be moved to xlcd, but for now it's very nonportable code
- fixes a minor issue with the configuration screens
Change-Id: I897f01b7210cbbec32665c2bc67920c965ea0bec
- upon a failed smalloc(), the audio buffer will be used for further allocations
- should fix things on low-memory targets (c100 and c200v2), but breaks playback
- playback should still be intact on other targets
Change-Id: Ic239f1316efadc957050afacf5c614dbbca3f805
Some changes in behavior were made with filesystem code commit
for the sake of compatibility that changed expected behavior.
* Restore substitution of drive spec in fully-qualified DOS paths
with the playlists's volume spec (or root on univolume targets).
Drive-relative paths of the form "c:foo" (no separator after
':') will be treated as purely relative.
* Restore old behavior of preserving leading whitespace in the
source path and trimming only trailing tabs and spaces.
* Multivolume: Volume substition on fully-qualified UNIX/RB paths
has NOT been reintroduced (and perhaps wasn't intended in the
first place). They will not be modified because there is no
ambiguity to resolve. Doing so would prevent a playlist on
external storage from referencing a file on main storage without
qualifying it with "/<0>...".
* Plain relative paths are and always have been interpreted as
relative to the location of the playlist.
Change-Id: Ic0800cea79c59563b7bac20f8b08abb5051906c7
