can't have it both ways blocking directories from filename
makes shortcut folders browse into the folder but blocks
the folder when the file browser loads it causing you
to be in the wrong folder upon resume
I wanted a way for the user to be able to leave the slash off
so ensure shortcut folders have a trailing slash (on load)
Change-Id: I9ea173a90a8c12291a159e7d30ee1076e0ca5d3e
The main usecase of this is so the morse cheat sheet does not look broken (with cut characters) when we are using a different/larger sysfont. All is adjusted automatically thanks to the SYSFONT_WIDTH and SYSFONT_HEIGHT constants.
Long pushs are now represented as vertical long bars, while short pushs are represented into short bars (centered in the middle). The long pushs bars are not larger anymore, so spacing is consistent everytime for all long+short order different combos thus looking cleaner and more readable.
The code was also a bit refactored to be more easily understandable in the future by using clear variable names to describe the magic values.
- Tested with mini2g: no regression, can still show all characters no problem within the limited screen space that is available on this device
- Tested with ipod6g by configuring the same default font as erosq (14-Rockboxmix) which is larger than the default sysfont: looks really good ! Way more readable than before
- Tested with 4gmono: no regression
Change-Id: Ida8b37ac7bd0bb5f0d314b5dea7657aa41bf2650
Accidentally put one too many zeroes in the address, placing it into
(real) IRAM instead of just past the edge of DRAM.
This didn't affect the s3c2440 and imx31, as they didn't have anything
in that typo'd location.
Change-Id: I2a2e3e3e0d1222ccd04901f8a85790cf21f0aac9
This puts the entire ~26K decoder state structure into IRAM.
(was lost as part of 14c6bb798d, Nov 2019)
Change-Id: Idbf53657e7c0581b4e47247fc5550b59842b23f1
optimize the index search loop using mod()
instead of checking index and then scanning we can start at index
and roll over saving a compare in the process
Change-Id: Ie54fbf1ca0131db8914985b654248eeb8d725a82
Wrong title is due to the VIRT_PTR being 0 so when a NULL
string is passed through P2STR() you get langid(0)
that was 'Yes' now it will be 'Rockbox'
the crashing menu was due to a call to retrieve entries that wasn't guarded
against getting called on a menu with no entries
g#5911 introduced this path to call custom_action for shuffled track adding
Change-Id: I3918debbe2e79dfc9595df56b05b1ae59ba94317
Only affected status display, charging is handled entirely by hardware.
Introduced in f3026cd0 (2024-11-02)
Change-Id: I08c7c442fb3bddf18e5a0d33dac963c24d3c9182
while its not a particularly hot path it still gets called upwards
of 1000x on startup
with a little pre processing we can make it faster than even the
prior version
check first letter before calling function
move strlen check to after the case match
Change-Id: I0108aaf9828501b57b0950fbc24cd478a9eeacc1
The "charging" status is apparently "charging needed" as it
is asserted even when power is not being supplied. So first check
to see if USB is connected, and if so, then check the "charging" status.
Change-Id: I3050f187e0b6c9d97d25d80015b413cd02e5c3b2
On PATA, we'd cap our transers at UDMA2 if the device reported that an
80 pin cable wasn't detected, but SATA devices do not perform this test.
So alter the check to only apply on PATA devices, so that SATA devices
can run at full UDMA speeds.
Change-Id: Id7aa25f2a702c0af73d707395439d69da1e04719
this is not a hot path only runs at start-up and tagnav reload
so speed is not so much of a concern
we can split the data into two separate tables to
remove the padding and allow symbols to be stored in uint8
(highest symbol <128) and do strlen at TOU
use uint16_t for get_clause()
Change-Id: I0f8663d0de9cac102f1d58b43bf200272754c466
- The "Special character" entry will show all special characters (non numerical + non letters)
- The numerical entry was bug, and could show some special characters from the ASCII table. It is now fixed.
Change-Id: I001fb322fab81918996e15e4d0ca6b7c9e5160af
Add a new entry, it is useful to go this way for large libraries rather than scrolling manually through the whole list
Change-Id: Ia54b613b4a5fa40863ebb176a549287d3a765a56
* All tagnavi.config entries pulled into language files
* Database viewer looks up LANG_IDs from the entries
* If we find a match, we can translate and voice these entries
* Add (disabled) mechanism to allow voicing database metadata
Original patch by Paul Sauro
Modifications by William Wingus
Further modifications by Solomon Peachy
Todo/Problems:
* Current Tagnavi headers are rather awkward in English
* Can't voice the "By first letter" alphabetic entries
* No mechanism for generating talk clips for DB metadata
Change-Id: Ic276ccda1bd8aae550d38be852bae4c6f697cd47
callgrind identified check_viewport and scan_int as pretty hot code paths on startup
and playback
check_viewport uses strlen to check the string has at least 4 characters
we can just check if str[3] != '\0' without walking potentially much further
and no function call..
scan_int was building a buffer for atoi to parse when we can just do it in the loop
directly
Change-Id: Ie028980333cbed4c066d8ea547a89cf4fad76808
On these older iPods, power was not being shut down completely, which led to a backfeed situation leading to decreased battery life and some stability issues.
This was particuarly apparent when using SD card adapters that do not
respect the ATA power management commands (ie all of them), as they never enter a low-power state on their own.
With this change, there are reports of battery life exceeding 20 hours of continuous playback (~30% increase with CF cards, 3x improvement with SD cards) and appears to resolve intermittent wakeup stability issues with SD adapters.
Change-Id: I46cfff7a59bb18a448989812303f30869df24d2d
in testing of three ways of doing this
current: ((_ctype_+1)[(unsigned char)(c)]&_N)
alt1(this patch): (((unsigned int) (c) - '0') < 10)
alt2: ((unsigned int)(c ^ 0x30) < 10)
alt1 and alt2 are the same in terms of speed and instructions (on arm v7) but alt2 has one more
instruction on mips
(across several archs in godbolt mips, armv7v8, x86) and on ARM7 (clipzip) device about 9% faster
less false positives for both alt1 and 2 when you start supplying more than 8bits
not sure if that matters in practice though
I tried similar with isxdigit but could only get to within 1 instruction of the ctype implementation
although it negated the array lookup I saw no discernable speed difference on device
https://godbolt.org/z/qGvh4hqnG
Change-Id: I5c9e8fd3915709853e0e33427038e20a068058b6
Responding to the bug report posted by iPodVT:
https://forums.rockbox.org/index.php/topic,55180.msg255292
- An active sleep timer that runs out will now cause the
player to shut down regardless of playback state
- When playback state is paused or stopped, once the
idle timer has run out, player will shut down, regardless
of any running sleep timer
Change-Id: I33de682a63ed1db76174eb2394ef5568f37cc677
characters less than the first diacritic in the symbol table (0x300)
return false after checking the MRU table
we gain some performance by eliding the function call all together if less than first diacritic
Change-Id: I02c14e350eb168eca808523affad443cd43888b4
Adds 'Add Current to Shortcuts' for settings alowing you to save
the current value and restore it with a shortcut
Change-Id: I0d5c5da9553248a70fd46d5553486ebf7865a20a
maxlen is set to MAX_FILENAME by all callers so lets just make that part of the deal
check the file exists before we set it
Change-Id: I3074f3164fcd4b8873b69612d5c1a51a39de4baf
the fmr file wasn't being detected on multivolume targets
the global preset list couldn't be cleared once set
empty global preset file wasn't detected properly either
Change-Id: I9c4b40ed0b6f3dbb0d38eb668fc74a512ea34062
