Rockbox

The interface of my Sansa Fuze and the file transfer freeze randomly when transferring large amounts of data. If the player runs in Bootloader USB mode everything is fine.

I guess this could break the vfat file system, making this a nasty bug. Disabling Key and Mouseemulation did not help. As long as this isn't fixed, a solution would be to start the player in Bootloader USB mode by default. I don't know if this is possible though, since I don't know the the technical details of rockbox very well.

Beside that everything works just great, keep up the good word!

PS.: I'm using Debian with different kernels. I couldn't see any differences between different kernel versions regarding this issue. Intels Linux USB drivers and the Linux USB stack in general usually work excellent as far as I know, therefore I would blame it on rockbox (no offense :)).

Sansa Clip v1. Stable version 3.10 [Tracker didn’t let me choose it?]
After playing the item for some time [varies] Rockbox crash and give this error:
Data abort
at 3001913E
FSR 0×8 (domain 0, fault 8)
address 0x7261702D

Using DFKT Minimum Clip theme.

In Rockbox for Android it is not possible to highlight an entry in a list without selecting (entering) it.

For example, if I am in the playlist viewer showing the list of just playing songs with my ipod (running rockbox) I was able to scroll through the list which moved the higligting bar through the list and then I could select the desired entry by pressing enter. With the touchscreen I found no way only to highlight a certain entry.

This is important if all entries are very long and have the same starting text. For example "Ludwig von Beethoven, Symphony No. 9 - 1. Adagio" cannot be distinguished from "Ludwig von Beethoven, Symphony No. 9 - 2. Andante" in the playlist viewer since the distinguishing part is cut off. With my ipod I could highlight the entries with the scroll wheel and the highlighted line started horizontally scrolling so I could see the complete text. With the Android port I can only touch the line, but touching the line does not only highlight it but also selects it and so starts playing.

This problem also exists in other list with long entries and happens with all version I compiled until now (latest was a12b2a5).

Rockbox Info: Version: 3.12

Device: iRiver H140, 80GB Toshiba MK8009GAH 8mm 1.8" IDE HDD 8mm

BUG: Repeatable, Consistent.

Additional Information:
1. To verify that I had a good installation of the 3.12 version of the Rockbox firmware working, I have reformatted, and re-flashed rockbox 3.12 two seperate times after initially encountering this, duplicating this behavior. So it doesn't seem to be a bad firmware flash.

Expected Behavior:

1. Player should Power On and boot properly every single time the bootup sequence is initiated correctly (Play Button pressed when Player is in a Powered Off state).

Buggy Behavior:

1. Player Powers Up, then locks up (i.e. "crashes") during bootloader sequence, requiring a hard reset, and upon second boot, finally Powers On and boots properly, bringing up a functional session of the Main User Interface. This is repeatable, and consistent.

Bug Symptoms & How To Repeat:

1. Consistently the Player will lock up during first "boot" sequence. The bootloader will usually reach the text stage, and sometimes crash here. Sometimes the bootloader will get through the text stage and reach the Rockbox Logo and then crash here. The Main User Interface is never reached upon first "boot" sequence.

2. The result of the first "boot" lockup (i.e. "crash") requires that I reset the Player by inserting a paperclip into the Reset hole (on the bottom plate of the Player) and triggering the button hidden in there. This Powers Down the Player, and Resets the Firmware, leaving the Player ready to be Powered On again.

3. Upon the second "boot", the Player will "boot" and the bootloader will make it past the text stage, the Rockbox Logo stage, and then spin up the HDD properly; the bootloader will not lock up the firmware, and behaves 'normally', finally resulting in a completed boot up sequence with the Main User Interface coming up (as a user would expect). The Player is now ready to use.

Playback halts.
No error screen appears – screen seems to stay where it is when freeze occurs, i.e. WPS or back light off.

Tested multiple audio formats.

Does not occur if no head phones are plugged in.

If you play a song for a bit at low volume (tested -14dB) before raising the volume, it is okay.

To reproduce:
Find a loud, bass-heavy part of a song, and halt playback right there.
Set volume to 6dB.
Set bass to 14.
Resume playback.

Usually freezes in less than a second when I do that.

RB version:
33f3af2-130523

Hello

I run rockbox daily (last update 27 Feb 2014) and found an issue that I like to report. I have my sandisk clip zip connected to an external USB hub, which is connected to my notebook. Today I left the sandisk clip zip connected and turned the notebook off. I expected that the player will disconnect from the USB hub → blanks the screen after 20-30 seconds (depending on the setup) and then turns off after a while (depending on the setup). Unfortunately the last step won't happen.

The sandisk clip zip really disconnected from the USB connection (the usb picture disappears from the display) with the notebook going back into the rockbox menu (Play, Radio, Settings, Recording etc.) and after a while also blanks the screen as expected.

Unfortunately it does NOT turn itself off as it would if you leave it entirely unconnected (unwired) with the USB. The sandisk clip zip starts to turn the screen on again after a while, turns off again, turns on again after a while, turns off etc.

I had to plug off the mini USB connection (the wiring between the player and the usb hub) so it turns off and shuts down after a while.

What I believe is, that the USB subroutines might not know the difference between powered on notebook and powered off notebook. It only knows that it's wired and thus tries to turn the player on again after it blanks. It should really be possible to keep the player connected (and wired) when the notebook shuts down and have the player shut down as well (after blanking and then turning itself off).

Could this be reviewed and maybe worked on ?

Thanks so far.

Hi,
I've been using the development version of Rockbox in my ipod classic 6th generation for about two years I think, but from some time, like a month or two, I'm experiencing playback issues as the track I'm listening suddenly skip forward or backward, or even jump for a few seconds on another mp3 file inside the ipod I was just listening before. I don't think is a ipod issue because I can listen the file inside the ipod with my computer player without problems.

There some way I can check older development revisions for the classic? It appears that it doesn't show in the dev build link on the site because is not fully supported. Thanks

I am trying to build Rockbox for Sony NWZ-380 from Git repo, commit ce0b31d87db3c4c1c1bfb535c50770d33e9c4aaf.
arm-elf-eabi-gcc version 5.4.0, ARM binutils 2.30.
Everything goes smooth until just the latest step where this happens:

… LD duke3d.ovl
LD bmp.ovl
LD jpeg.ovl
LD png.ovl
LD ppm.ovl
LD gif.ovl
LD rockbox.elf
/usr/libexec/gcc/arm-elf-eabi/ld: section .ARM.exidx VMA [00000000600a9548,00000000600a954f] overlaps section .bss VMA [00000000600a7f20,00000000600ff773]

I'm using Rockbox on my NWZ-384 for a couple of years now and there were no problems building it in the past. I also was very glad that NWZ-380 port reached stable status in 3.14,
but now something seems to be broken (not terrible it seems, just requires some adjustments to the linker scripts but still).

Hi there,

i am currently trying to learn the morse code. I thougt it would be fun to have rockbox as an aid, so i started to write a plugin for that.

This plugin currently has a limited functionality, but the current things are implemented:

1. Morsing of custom texts, typed from the rockbox keyboard
2. Morsing of random chars, till the battery drains (or less ;))
3. Input of own Call and 'morsing' the same
4. selection of speeds from 1-45 WPM in steps of 3
5. selection of chars from 5 mixable groups
   1. letters
   2. numbers
   3. punctuation marks
   4. special chars (KA,BT,AR,VE)
   5. pauses

It would be very nice if someone tested these on some soft-codec targets. I included the source tarball (please add plugin to SOURCES and CATEGORIES if using this), a diff on r17715 and the binary .rock for the sansa c250.

Idd be very happy if, with a little help, this could move upstream. I am new to the plugin api and need some help especially in using the definitions to make this work on different DAPs (avoid building this on targets that don't support some features and so on). Correct me if i am wrong, but his should from my point of view work with all SWCODEC devices with LCD. I tried to write at least readable code and added some comments for the curious.

This builds at least on the c250 and works for me.

Thanks, Joel

Remind me, if i forgot something.

The Rockbox firmware and cross compilers build by using MinGW.

It is necessary to use MSYS’s make (MinGW’s make (mingw32-make) does not used.).

The cross_compilers.zip is a patch file necessary to make cross compilers using by rockboxdev.sh.

I confirmed the Rockbox firmware, codecs and plugins was built by using the compiler on MinGW.
But I can not confirm whether the Rockbox executes excluding iPod video.

Could you confirm operation to various players?

This patch adds a changable wheel scroll speed option on ipod video.

Hello,

This plugin is a game to be played in two.
I like to play go and pente and so is my wife (at least she plays pente) so I wrote this plugin.

I think the code is clear enough. I also used all the #ifdef that I knew.. so, I believe it wouldn’t break any compilation.

The plugin was tested on IPOD 5G (VIDEO) and works just fine. I also compiled it to a few more players without any problem.
I did not try to compile it for players that don’t have the lcd_update() function. It won’t work (at least this first version).

The board size in the LCD is based on WIDTH, since Sansa has width smaller than height you will lose part of the board (horizontally).
I just tested the code in sansa using the Simulator…

It is easy to fix and I will, but I’ll be traveling for a week (that’s why I wrote it, so I could play it while I wait in the airport/plane…) When I came back I will fix it.

Meanwhile, you guys can patch it or use it.
Feedbacks are also welcome.

Current Features:
- Save / Load games
- Pente / Go (Board 9×9, 13×13 and 19×19)
- Remove groups / stones automatically when they are dead

I attached the code and the “executable” for IPOD 5G (Video)

Enjoy!
Alex

I think this is good since now the plugins in the list appear in the order of their occurrence in viewers.config which is rather random.

Hi there,

first of all: this is my first contribution to rockbox and I tried my best to make myself familiar with the way things work there. I am sure, not everything is where it belongs. Please help me to get things in order :).

second: thanks to the author of Patch  FS#1803  that made this possible.

The purpose of this patch is to watch the tuner reported field strength (estimate) and mute, when it drops below a customizable threshold. Wether muting or not as well as the threshold are customizable in the fm tuner menu. It should not interfere with manual muting and should not affect recording (i didn’t test that yet). THe threshold is specified in pseudo-dBuV.
It also adjusts frequency display to display as MMM.KKK instead of MMM.KK. This is for people who adjust region settings and channel widths.

My testing conditions were limited to the sansa c250.  FS#1803  reports to work on the tuner chips LV24020LP (e.g. sansa c250) and TEA5767. I only tested the first. Please report back possible field strength values you see on devices using these chips.

This code should not break anything. I did a testbuild for some iPods (without tuner) that compiled cleanly. Simulator also works.

Currently only english and german translation are done.

User guide:
This patch conditionally adds 1 or 2 lines to the tuner main window that look like this:

Sig: [-M] -235dBuV
Automute below -555dBuV

The second line is only displayed if automute is enabled. The first line shows the status of the tuner. The 2 characters in the braces exlain as [ab]:

a: either 'O' or '-' 
   'O': current field strength is above threshold
   '-': current field strength is equal to or below threshold
b: either 'M' or '-'
   'M': will mute on or below threshold
   '-': automute disabled.
(both are only updated on field strength change)

The second line should be selfexplaining, same for the menu entries.

Request for comments!

The attached proof-of-concept patch highlights (places the cursor at) each album's last-played track in database views. The intention is to allow easy resuming of multi-track audiobooks or podcasts where they were last left off. This works best when autoresume has also been enabled.

The patch relies on runtime statistics gathering being enabled (and the DB needs to be enables as well, of course).

The current patch has some caveats that I plan to address in future revisions:

Right now, it's unconditionally enabled for all database-based track views. We might consider allowing to turn this off for regular music albums.

When the last-played track has been completed (played through to its end), we should actually highlight the next track (or the first track if the last-played track was the final track of the album). [EDIT: fixed – see latest patch in comments section]

It would be great if a similar feature was available for the file browser. I'd find it acceptable if that would require the DB as well (as that's the place where the last-played info is stored). It could benefit from an add-to-DB-as-files-are-played feature (aka FS#12073).

Comments?

[EDIT: Clarified summary. Fixed a rather severe bug in the patch to make it actually work. :) ]
[ EDIT 2: Linked to FS#12073. Addressed one caveat. ]

This patch makes some changes to codec_api to help remove dependencies for my project.

The unused strcasestr callback is removed.

The global_settings pointer was being used by several codecs to check whether REPEAT_ONE mode is on; this has been replaced with a callback "should_loop()".

The MOD codec has been modified not to restart the file when it reaches the end. This does not affect MOD files that use "position jump" to loop part of the song. Ideally the MOD codec would only follow such loops if should_loop() returns 1, but detecting loops is nontrivial.

The attached patch slightly modifies the keymap of the H300 non-LCD remote to allow access to the browser from the WPS (by use of LONG PLAY).

I know that it works ok from an H300 with a non-LCD remote, but I don’t know if I’ve broken the LCD remotes completely.

Please could users of both the H100 and H300 LCD remotes try this patch and tell me that the WPS context keymap is unaffected. For reference, you can see the intended mapping here: http://www.rockbox.org/twiki/bin/view/Main/KeymapIriverHSeries

It would be useful to change the configure and voice.pl files in order to allow the user to choose also the Loquendo and ViaVoice TTS

I’d like to suggest the following new WPS feature(s):

1. A viewport can be assigned an image as a “backdrop”.

 a. A viewport may be assigned an image dynamically via the WPS tag "%VX<identifier>"
   (i.e. %VXa for image a, %VXZ for image Z, etc.)
  b. As a viewport is being processed and drawn, its backdrop would be drawn "first", followed by any other normal images/characters.

2. Viewport may define FF00FF as a background color, which would then be “transparent”. This would allow overlapping viewports.

3. Fonts drawn in a viewport with a backdrop would automatically be drawn with DRMODE_FG.

Notes on the attached patch file:
+ It was created from the SVN revision 18997 on Nov 3, 2008.
+ Although I stubbed in the new functions for all the LCD drivers, I’ve only flushed out the 16 bit driver so far.

Cheers!

This plugin searches in the .rockbox/rocks/apps/drumkit directory for .raw sound files, loads them into memory, and assigns the keys to play them, creating a drumkit /sample trigger / instrument out of the player. Currently only designed for the Sansa e200.

Room for improvement:

- Add keymaps for all
Allow multiple samples to be played at the same time (by using
Allow for a variety of sound file types to be
Find a better name

This patch adds a sector-based read/write cache between the storage and fat drivers.
It has a static pool of 64 sector buffers which can be locked by the file handling code to replace the sector buffers in the file and dir structs, and serve as cache as long as they aren’t used. It can also be used to enforce exclusive access to a sector.

If requested using the STORAGE_NEEDS_ALIGN define, all storage accesses will be aligned to cacheline boundaries.

This is a new plugin that mimics the open source PC program mnemosyne (http://www.mnemosyne-proj.org/), a program that assists with memorizing flash cards. The plugin is fully functional, but it still needs a menu and assumes that the user has setup the folders it needs for storing cards before running (if you don’t create them it will crash). I’ll fix these issues in the near future, but in the mean time any feedback is appreciated :)

This patch attempts to permanently fix the inconsistent mixture of “Title Case” and “Sentence case” in english.lang by applying “Sentence case” to the entire file.

I suggested we do this in a post to the rockbox-dev mailing list last year, but that just ended in a very split discussion with no conclusion.

I’m not sure if this patch wil result in anything different, but I made it for my own use, and thought I would post it, as there were a fair number of posters to that discussion that also prefered sentence case.

The old discussion:

http://www.rockbox.org/mail/archive/rockbox-dev-archive-2008-08/0034.shtml

EDIT: I forgot to say this patch also changes a couple of uses of “center” to “centre”…

EQ settings changed via Touchscreen WPS action setting_inc & setting_dec are not reflected in the sound output of the device even though they display having been changed in the menus. The changes are also not written to the main config file on disk so they do not "stick" after Rockbox has been rebooted.

example syntax to toggle EQ on/off (display DOES get toggled):
%T(148,520,54,46,setting_dec,eq enabled)
%T(148,520,54,46,&setting_inc,eq enabled)
%xl(E,eq.bmp,0,0)
%V(148,520,54,46,-)
%?St(eq enabled)<%xd(E)|>

I'm not sure if this problem is specific to Android or if the problem applies to all targets.
APK Build: Feb 4, 2011

I didn't like that the "Backlight on first keypress" function ignores the first keypress no matter which key was pressed. For example, I want the volume keys to work normally even when the backlight is off (that is, they should turn on the backlight *and* change the volume).

This patch restricts "Backlight on first keypress" to certain buttons only. in other words, these buttons work normally only when the backlight is on; otherwise, they just turn on the backlight. As a proof of concept, it changes the Clip and Clip+ behavior to filter out only the Home key when the backlight is off, allowing the other keys to function normally.

It works by allowing button-target.h to define the buttons that are filtered by the "Backlight on first keypress" function. If left undefined, the function filters all buttons when the backlight is off, as in the status quo.

[EDIT: Clarified that there is no overlap with  FS#9305  - Context sensitive backlight on key press]

r29445 introduced problems with screendump on sansa e200 and e200v2 (and possibly more). Screendump works properly with r29444.

Steps to reproduce:

1. delete or backup existing .rockbox directory
2. install r29445 (or later)
3. reboot
4. navigate to System → Debug (Keep Out!) → Screendump and press select to enable
5. attach USB cable to trigger screendump

On e200, this results in:
*PANIC*
Stkov usb (0)

on e200v2, this results in a deadlock - no keypad response - backlight & button light never timeout. Upon reboot, multiple dump files exist.

On e200v2 only, when dircache is disabled, screendump will function properly.

The player (Sansa Clip Zip) appears to refuse or unable to play some of my WMA tracks; as far as I can tell, the problem is specific to my WMA Pro files only as it plays none of them (but works with WMA2 successfully). I have attached one of the examples.

I understand that a similar issue was reported recently (in  FS#12619 ) although, in my case, the player doesn’t start playing and skips to the next immediately, rather than playing part of the track.

For the last couple of weeks, I can’t get Rockbox to run anymore on my phone. I use Rasher’s build, which used to work fine. I even built one on my own, but it still doesn’t work. I tried running it on the emulator and it worked fine. I am not sure what is wrong, reading the logcat gets me nowhere because I don’t really understand most of it. I have attached the logcat here. I do not know if I’m the only one having this problem.

I tried various method to get it working again on my phone, I even tried resetting my phone in case if I messed up somewhere. I also noticed this when I was trying out another ROM for my phone, which runs on Gingerbread 2.3.7. I thought it was the ROM’s problem, so I changed back to the ROM that I frequently use, which used to had no problems at all running Rockbox. The last working build was on 01 March 2014, after that, nothing works anymore on my phone.

The only possibility that I can think of is that my phone is outdated and the kernel or the ROM is missing something crucial to run the latest Rockbox builds. Other than that, I have no idea. I hope the logcat attached could help you identify this problem. Thank you in advance.

Build version: 915bc09-140318
Last working build (on phone): unknown revision, dated 01 March 2014
Phone: Sony Ericsson Xperia X10
Android: 4.1.2 Jelly Bean (CyanogenMod 10)
Emulator: 4.1.2 (API 16) with same screen dimension.

Pressing the remote volume UP key on my iriver 140 in wps sometimes results in directory skip.
This is not easy to reproduce, but I was able to repeat once that problem in conditions where I'm sure that the directory skip key wasn't pressed.

I don't know if it's a hardware related problem, but I've read somewhere about another user having the same problem.

According to http://download.rockbox.org/manual/rockbox-sansac200/rockbox-buildch5.html#x8-930005.8 the power button is supposed to exit the recording menu, but when I press it nothing happens. I have to power off the device and power it on again to get back to the main menu.

- What behaviour you are seeing

Rockbox 17612-080523 on an iRiver H340 skips entire tracks, and also skips within tracks, when the battery level falls below 30%

- What behaviour you expected

No skipping.

- How to reproduce the bug

Allow the battery level to fall below 30% and start playing a directory of files. Some files will be skipped, some files will have parts skipped in the middle.

- The SVN revision you are using and if possible the particular SVN revision in which you think the bug was introduced.

17612-080523, no idea when the bug started.

If you need any further info on this bug, please ask and I’ll give more details. I have no other hardware to test whether this behaviour occurs on other Rockbox platforms.

This may represent other players as well.

I installed a new battery on my iriver, a 2100 mAh that is working perfectly. The problem however, is that when fully charged, the voltage holds up to 4.12 volts, and until it drains to somewhere under 3.9 volts, Rockbox thinks that it is powered via USB, so naturally Rockbox gets stuck in USB mode until the battery is drained enough for Rockbox to realize that it IS a battery. With this problem at hand, I have to charge my iriver to only less than seventy percent of its battery life, or else wait for an hour or so to drain it to that point. I HAVE tested it in its original firmware, like I said, works perfectly. The problem lies with Rockbox thinking that the battery is a USB charger. I am asking for a fix around this. Is there a way that I can modify Rockbox so that it may understand that it is running off of a battery that will get up to 4.12 volts and not mistake that for USB power? I just now thought to turn off the “Charge During USB Connection,” maybe that will solve the problem; I will get back when I find out whether or not that works. However, I would still like to request a work around for this problem. Any help is much appreciated. Thanks.

Tracks played via a cuesheet are not recorded in .scrobbler Last.fm log.

I am running 3.4 on a 5.5G iPod modified with a 240gb drive. Rockbox is stock 3.4 with LBA48 added in and the max sector size turned off. The database was working ok on 3.3. Only issue was it took forever to add new files. After upgrading to 3.4 I noticed every time I powered on rockbox would lock up in the menus for up to 10 minutes before functioning normally. So I completely removed the .rockbox dir (including the database files of course) and re-expanded a fresh copy. I then tried to init the db. It runs until it is around ~44k files and 99% and then just sits there. If left alone it will eventually just turn it self off. When I turn it back on it will ask me to init the db again. I have seen it try to commit once or twice but it never takes. I am not really sure how to debug this at this point. Also in the debug disk info area, it things my drive is 131071MB. Under Rockbox Info it says 223GB which is better, but these two things are the same as in 3.3. All my Oggs are converted via a script I wrote from my flacs. I do not believe my flacs have any art, I am not even sure you can store art in a flac tag and even them I don’t think the ogg encoder is transferring them. I ran a tag dump (ogginfo) against my oggs and do not see anything to do with album art. I had 4 files on the device with album art. They were all in a directory with a database.ignore file. So they should not have been noticed. Just in case I removed that directory and rebuilt. It went though 44k files, commited 1-9 and then went back to “scanning” but the disk activity had stopped. Upon reboot it asked me to init the db again.

This is not a patched build. I simply changed the config to include large drive support. This is an issue that needs to be looked at. Please check into it, I will be happy to test any ideas.

After playing a few opus files (~128 kbps) the system becomes so unresponsive that it cannot decode the file in real time, and the device cannot respond to any input within a reasonable timeframe (have to hard power off).

This does not occur when you set the cpu frequency beyond the highest setting in the debug menu!

To reproduce:
play opus files in a playlist with normal cpu scaling
on beginning of 2nd or 3rd file, system will become unresponsive.

Here’s a patch to version r29143M-110324 that adds the Zipit Z2 as a build target. It requires one new file (also attached) firmware/target/hosted/zipit/powermgmt-zipit-z2.c to monitor the battery.

This is mostly a generic SDL app with a few tweaks for the Zipit buttons and battery monitoring. The patch also includes a few small bug fixes for SDL app builds. I had to fix the labels in the asm section of lcd-16bit.c. There were some missing parenthesis on an #ifdef in button-application.c. And the apps/SOURCES file needs an #ifdef to handle an SDL app build without touchscreen support.

I’m using several SD-cards with my Sansa. When switching cards
and updating the database the db entries of other cards are lost
(unless saved to the database log).

In order to differentiate between files from different cards the
easiest way is to give them different path names. So instead of
mounting each card to <microSD1> this patch takes the volume label
i.e. ‘Classic1’ of the filesystem and mounts it to <Classic1>.

The database update is modified in a way that it erases only entries
for files with a filename of a mounted filesystem. So data of other
cards isn’t erased.

So far the database search isn’t modified to handle nonexisting files,
you still get albums etc. which are currently unavailable.
The patch only makes sure that nonexisting files aren’t entered into
the current playlist.

The patch is only activated for Sansa e200, as I couldn’t test it for
other players.

This patch enables you to control the simulator using a named pipe to provide simulated keypresses, rather than the keyboard.

Unfortunately there is a bug that means that not all events seem to have any effect. On the e200 sim, I can use the scroll buttons to move in the main menu, but neither RIGHT nor SELECT lets me enter any of the entries. I suspect it's related to repeat or hold/release handling, but I'm not sure how to fake that properly.

When you play back a broken flac file (in this case a file that was truncated) it can happen that he decoder outputs constant loud noise instead of just skipping to the next track.

I’m not sure if this is really a bug since you are not supposed to play back broken files.

Attached is a file where this happens.

H300 r18009 (happens in the uisim too).

Note: This is work in progress.
Status: Code implemented

      Manual not updated
      Wiki not updated (only after commit)

The idea of this patch is to have the same keyboard keys assign to the same actions in all platforms.

For example, have +/- keys (both on keyboard and on Numpad) are assigned to Volume Up and Volume Down keys of the original platform.
For platforms which do not have these keys, pressing the +/- keys in the simulator will do nothing.

This change will help developers to do manual regression testing on lots of platforms easily, as they do not have to memorize each platform’s key assignment.

Future development: Display key assignment on screen, and/or on console

I noticed that when filling the buffer my Video iPod (32MB RAM) does noticeable seeks at semi-regular intervals even if reading from a single non-fragmented file. It seems this is due to having to read the next FAT sector. In firmware/drivers/fat.c there is a FAT cache but no read-ahead of the FAT.

I created fat_preload( struct fat_file *file, long sectorcount ) in firmware/drivers/fat.c to preload the FAT cache. In firmware/common/file.c I created file_preloadfat(int fd, size_t count) because calling a fat_ function directly would be too much of a kludge. I use file_preloadfat in apps/buffering.c just before filling the buffer.

This got rid of that regular seeking and it seems to it can remove about two seconds from the time needed to fill the buffer when reading a single non-fragmented file. Currently the code is quite simple and unintelligent. Fragmentation or other activity can overwrite the pre-loaded FAT sectors in the cache, and in such cases the pre-load leads to needless extra reads. More work is probably needed before this can be included in Rockbox.

Currently this patch is for SW-codec only. I think file_preloadfat could be called from HW-codec buffering code also, though few seeks would be prevented with 2MB of RAM. This patch is also probably useless with flash storage.

The patch allows to use the persistent dircache even in targets that do not define HAVE_EEPROM_SETTINGS.

Starting trying to implement such a feature for my iPod Video, I "discovered" the functions dircache_load/save (I have a slight knowledge of the RB software). Investigating why it was not used on my target, I saw that it's conditioned to the define HAVE_EEPROM_SETTINGS without an apparently reason to limit them just to some targets.
So, I simply "detach" the functions definition and use from the define.
I'm using it since an hour, starting up and shutting down the DAP various times, jumping back and forth inside the dynamic play list, and I got no problems at all.

If I guess correctly the code, the dircache should be updated also when changing the DAP content, if the USB connection is done by RB and not the OF.

It's applied to r29907.

First approach, to add dma read/write disk access to the sansa.

There are some restrictions on the dma functionality with this patch:
- only stabile at 80MHz ⇒ cpu boost control added to the driver
- only on 16byte aligned data ptr ⇒ all data are copied to/from aligned memory
- no interrupt support ⇒ using polling method

At 30MHz the driver is much more stable with the following PLL settings:
PLL_CONTROL = 0x8a221e0c; /* 30 MHz = (48MHz/12*30) / 4 */
CLOCK_SOURCE = 0×40007777; /* source #1, #2, #3, #4: PLL (#4 active) */

Reading seems to be in general more reliable than writing of the cards.
The data throughput is a little worse than in the current implementation.
This is partly due to the data copying.
The sporadical ‘2bytes inserted’ problem with the USB file transfer is still
there. Interestingly @30MHz with the above settings, there is another problem:
sporadically ‘2bytes missing’ and ‘2bytes inserted’. Perhaps the DMA transfer
needs some more initialization for proper synchronization.

This patch modifies the Gigabeat S keymap so that it can be used almost solely using the buttons on the face of the player.

Currently, there is an unused button in the WPS screen (the BACK button, up-left from the cross) which in my mind makes more sense as the “back to the menu” button, while the MENU button (up-right from the cross) is the logical Play/Pause button (at least to a right-handed individual). It’s unfortunate that the button with the Windows logo has been called the MENU button in the code, thus making this rather confusing to read, but if enough people agree with the feel of this change, that could be modified as well.

i asked some time ago why there is the buttontable in general settings –>bookmark. i wondered because the buttons are the same as in other lists. Marc Guay said "the delete button is unique", an i wondered why there is a dedicated delete button… in no other lists exists such a button afaik.

i changed the manual accordingly.
i also corrected some wrong (?) comments.
i hope i did everything right, at least for the e200 it compiles and works fine.

i would like to get some opinions about it.

A new background image for bubbles that looks a lot better.
Made using GIMP and images from frozen bubble.

Copy the attached file to ‘apps/plugins/bitmaps’.

This patch adds a secondary backlight timeout (similar to what the the iPod OF does).

It adds two extra settings: secondary backlight timeout and secondary brightness (I'm not sure wrt the naming of those, any suggestions?)

The code isn't very clean, but it works (except for CONFIG_BACKLIGHT_FADING == BACKLIGHT_FADING_PWM targets, for which I haven't implemented this).

Please test this.

Steps to reproduce:

1. Create a directory: /AUDIO;BOOKS
2. Place a file in it: /AUDIO;BOOKS/file.mp3
3. Start playing the file, and then create a bookmark
4. Do “List bookmarks” for the file
5. The bookmark menu lists something like this:

BOOKS/;file
0:02, 2, Shuffle

The problem seems to be that .bmark files use semicolons to separate the fields of information. If there’s a semicolon in the path, it thinks that it indicates the end of the path and the beginning of the filename.

I have RockBox version 3.9 on my Sansa Fuze v2 and so far I love it, but I have one problem. On a few tracks in my library, when I play them, the first few minutes of the song will be strangely distorted. Most of them are songs where the music starts off in just one of the speakers before moving into the other one. There will be bursts of loud, crackly noise in the other speaker during these first few seconds (that seem to be in time with the music) and then it will fix itself and the rest of the song will play perfectly. When I boot the default Sandisk firmware the songs play perfect like they should, it is only in RockBox that this glitch happens. I’ve messed with the settings a bit and even reset them completely, but nothing has worked thus far.

Attached is one of the songs that this happens with, from about 2 seconds in to about 7 seconds in. Any help would be greatly appreciated. I really want to enjoy my music without that crap in it.

The location for recordings can be set, but the only way to see the current value for the setting is to look into the config file. The patch adds a note stating that to the manual.

Also, there is a known bug ( FS#12517 ,  FS#12500 ) preventing saving the recorded files in the root directory (maybe this applies only to some targets, not all). The patch adds a warning about that.