AgeCommit message (Collapse)AuthorFilesLines
2014-03-20OS X: change include paths to use actual paths.Dominik Riebeling1-2/+2
The old include path is actually a symlink for compatibility. Newer SDK versions remove that symlink to the include breaks. Change-Id: I74317557840f1886cebc1ce099e103dc415eabc0
2014-03-18some optimization for iriver h100 and h300 seriesDmitry Gamza1-2/+2
Turn on 4x oversampling inside UDA1380 codec. This allow will move the quantization noise to the higher frequencies, that slightly improve the final sound quality. Change-Id: Idf79cd877c526d8937a47c34e9c180d837a15db1 Reviewed-on: Reviewed-by: Michael Giacomelli <> Reviewed-by: Björn Stenberg <>
2014-03-18Fix build reds caused by typos.Thomas Martitz1-2/+2
Change-Id: I71cdca3b6e73e29d487e0c6d02a69a31dcf3e6bf
2014-03-18usb screen: Stop scrolling the entire display.Thomas Martitz1-1/+5
It's not easy for the originating screen to perform cleanup in case of USB because the usb screen is invoked indirectly via default_event_handler(). To avoid stale scrolling lines perform the cleanup in the usb screen itself. Side effect should be no worse than scrolling lines to restart in SBSes that show during USB. Change-Id: I9aa491b3d4b649c0b3be70048ebcb9d817c0356c
2014-03-18quickscreen: return true on USB connection.Thomas Martitz1-7/+13
This is consistent with other screens (pitch, id3) on the WPS and in other places. Change-Id: Iafae52507867ff9815a60cf44faaf09a29c55091
2014-03-18Add Dmitry Gamza and Sebastian Leonhardt to CREDITS.Thomas Martitz1-0/+2
Change-Id: I10c3c8963277905f3c74247fc16acc247323e30f
2014-03-18fix quick screen write to configDmitry Gamza1-1/+1
Change-Id: Id55f81011a9ce47589d879203ca4f8fd4c54f593 Reviewed-on: Reviewed-by: Thomas Martitz <>
2014-03-18Fix (unintentional) linkage of HAVE_BUTTON_DATA with HAVE_TOUCHSCREENSebastian Leonhardt2-2/+6
Defining HAVE_BUTTON_DATA without simultaneously defining HAVE_TOUCHSCREEN caused compile errors. (I need them separated for a scrollstrip driver.) Change-Id: I945d3437d840035ccc0c147f8155029b403c6ec2 Reviewed-on: Reviewed-by: Amaury Pouly <> Reviewed-by: Thomas Martitz <>
2014-03-15Fix stupid typo in 470989b that triggered panics easily.Thomas Martitz1-2/+2
Change-Id: Ic998a0ea93141a5399e28dc286eea4e3d40663d6
2014-03-14Fix reds.Thomas Martitz2-2/+4
Change-Id: I4f065a5b9f96df4d70ee8bb2836effc9b97b01fa
2014-03-14events: Rework event subsystem (add_event, send_event) to be more versatile.Thomas Martitz25-149/+247
add_event_ex is added that takes an extra user_data pointer. This pointer is passed to the callback (add_event and add_event_ex have slightly different callbacks types). All callbacks also get the event id passed. Events added with add_event_ex must be removed with remove_event_ex because the user_data pointer must match in addition to the callback pointer. On the other add_event is simplified to omit the oneshort parameter which was almost always false (still there with add_event_ex). As a side effect the ata_idle_notify callbacks are changed as well, they do not take a data parameter anymore which was always NULL anyway. This commit also adds some documentation to events.h Change-Id: I13e29a0f88ef908f175b376d83550f9e0231f772
2014-03-10ab_repeat: Move GUI bits to the skin engine. This allows ab_repeat to be ↵Thomas Martitz3-62/+79
compiled headless. Change-Id: I4fa3b8ef7139313891ca70df11f7f17c5df38cb7
2014-03-10android: Fix script.Thomas Martitz1-6/+6
Change-Id: I6c22043956cea5c5aec3e411045082e1a9b69442
2014-03-10Implements starting playback from a cuesheet.Michael Sevakis2-26/+80
Before time-based resume this was impossible since playback could not be started at a specified elapsed time, only seeked with playback already running. Right now the "FILE" field is used, if present, to do the lookup from from the .cue to the audio file when it is separate from the audio file. If no path is specified, the .cue and audio file must be in the same directory. When the cuesheet is embedded, the containing file is used and the FILE field is ignored. Supports starting playback and seeking to cue points from the cuesheet browser even without Cuesheet Support turned on. Change-Id: Ib5b534c406f179a7f8c7042a31572b24a62c0731 Reviewed-on: Reviewed-by: Michael Sevakis <> Tested: Michael Sevakis <>
2014-03-10Implement time-based resume and playback start.Michael Sevakis65-302/+729
This complements offset-based resume and playback start funcionality. The implementation is global on both HWCODEC and SWCODEC. Basically, if either the specified elapsed or offset are non-zero, it indicates a mid-track resume. To resume by time only, set elapsed to nonzero and offset to zero. To resume by offset only, set offset to nonzero and elapsed to zero. Which one the codec uses and which has priority is up to the codec; however, using an elapsed time covers more cases: * Codecs not able to use an offset such as VGM or other atomic formats * Starting playback at a nonzero elapsed time from a source that contains no offset, such as a cuesheet The change re-versions pretty much everything from tagcache to nvram. Change-Id: Ic7aebb24e99a03ae99585c5e236eba960d163f38 Reviewed-on: Reviewed-by: Michael Sevakis <> Tested: Michael Sevakis <>
2014-03-09ypr1: Add some commands to which are needed for playbackThomas Martitz1-0/+13
Change-Id: I2772e218174b5d1fbc7c066fee41d6fb06c01900
2014-03-09Fix ipod6g build.Thomas Martitz1-0/+2
Change-Id: I218e1721b8ddbb067581f4bc686a20f4c88694ae
2014-03-09android: Get the port up and running againThomas Martitz7-53/+61
The build system needed fixes because the tools paths changed and one tool that we used (apkbuilder) was removed entirely. Recent NDKs don't ship gcc 4.4.3 anymore, therefore switch to 4.6. The code itself needed a fix for a jni reference bug that was uncovered by KitKat. The port now builds with latest sdk (r22) and ndk (r9d). Change-Id: Id74fa54ba93bbb0ee30373fbe79e92c5ff03201d
2014-03-09zenxfi: enable speakerAmaury Pouly2-1/+4
Change-Id: I603112d2a348abf92d5c8975ea76d3a57fda7cfc
2014-03-08zen(xfi): enable frequency scalingAmaury Pouly2-2/+2
With the lcd underflow code at work, it results in small flicker when switching frequency but it should be fine because it doesn't happen very often and I don't think there is a way to avoid it without disabling EMI frequency scaling. Change-Id: I458732d52bbf27cda2510c3836539db1fb24ea15
2014-03-08imx233: rewrite frequency scaling to use the new functionAmaury Pouly1-20/+4
Change-Id: I0f9968de76ce17710d31f7bc609440654e68b6f1
2014-03-08imx233: add function to change cpu/hbus frequency safelyAmaury Pouly2-0/+34
Change-Id: I88e9ad54ba65846ae4d94ae03009b3656f2489f2
2014-03-08imx233: prepare frequency scaling for stmp3700Amaury Pouly2-8/+13
Add entry for stmp3700 maximum frequency which is 320MHz. Change-Id: I6db4aad4efa0a7c1347a1ceb262a0295f63057ae
2014-03-08imx233: fix regulator code on stmp3700Amaury Pouly1-0/+17
For some reason the power subsystem needs to know the relationship between the VDD{D,A,IO} and uses a weird register to do so. Change-Id: I7fcc75f6cc0460b4997914986deda7ca544a4940
2014-03-08zen: handle lcd underflow by handAmaury Pouly1-4/+17
Contrary to the imx233, the stmp37xx lcdif doesn't know how to properly recover from underflow and things are worse because of the errata which makes the lcdif not clear the fifo. Workaround this by detecting underflow and taking action: stop dotclk mode (will clear fifo) and schedule next frame. The dma transfers now write the ctrl register as part of the PIO writes, making the code simpler. Change-Id: I15abc24567f322cd03bf2ef7903094f7f0178427
2014-03-08imx233: add lcdif underflow irq supportAmaury Pouly2-0/+23
Change-Id: Ie3fe223ff40abff38bb9b09f398eb5411fa0be4c
2014-03-08Greatly reduce volume-change zipper artifacts with SW volume.Michael Sevakis3-28/+108
Uses a cosine factor to smoothly shift the PCM level from the old level to the new one over the length of a frame. Implements indirect calls to PCM scaling function instead of testing conditions on every callback, cleanly assigning a different call to do the volume transition. The volume change call then assigns the final scaling function. Change-Id: If1004b92a91c5ca766dd0e4014ec274636e8ed26 Reviewed-on: Reviewed-by: Michael Sevakis <> Tested: Michael Sevakis <>
2014-03-06Oscilloscope: Fix loading of the speed setting for waveform viewMichael Sevakis1-1/+1
Somehow, a long long time ago, I put MAX*/MAX* in the config data range fields instead of MIN*/MAX* which forced the settings loader to acknowledge only 100 as a valid setting other than the default. Change-Id: Ifbfcafad27e239942449848d6a09426173bbde54
2014-03-06Gigabeat S: Improve headphone thread implementation.Michael Sevakis1-81/+54
Implement scanning as binary tree in array. Make the ADC calls fewer without compromising read quality. Declare the thread function as 'noreturn' to save some stack. Reduce stack size (regardless, % use is now a bit lower). Change-Id: I239792fd2a0a2c019d1ec4af1d6d4b466cdf0ef5
2014-03-04atj213x: various fixes in sample code snippetsMarcin Bukat4-22/+25
Change-Id: I78781e1a56cb6705d011ee2296f1789b497a566a
2014-03-04ingenic: crt0.S fixes #3Marcin Bukat1-6/+5
Hopefyly this time is the final one. Change-Id: I386964f407ca98499e9a482507e4d5dfa02ee891
2014-03-04ingenic: fix bugs introduced by 5900bf7Marcin Bukat1-6/+6
I must have been blind and copied some intermediate file instead of final version. No functional change, only fixes. Change-Id: Icaa868225517662c88f8492b1f04808407853361
2014-03-03Fix last warnings.Thomas Martitz2-2/+0
Change-Id: If0a7e11c3e107cb7756c0d70ed5b17fda92b166e
2014-03-03Fix more reds.Thomas Martitz11-13/+22
Change-Id: I4b58dda0953b7f9799238c32b78037b0a5403c04
2014-03-03Fix various reds. Some includes needed fixup.Thomas Martitz15-4/+20
Change-Id: I4327740bae17054131feb917abdd58846c451988
2014-03-03zen/zenxfi: correctly implement lcdif errataAmaury Pouly1-1/+3
Change-Id: Ib2a6000035d70d687025a78bbea416d77af562fe
2014-03-03kernel: Break out kernel primitives into separate files and move to separate ↵Thomas Martitz30-756/+1564
dir. No code changed, just shuffling stuff around. This should make it easier to build only select parts kernel and use different implementations. Change-Id: Ie1f00f93008833ce38419d760afd70062c5e22b5
2014-03-03Iaudio m3: implement lcd_get_dpi() (aka fix red).Thomas Martitz1-1/+3
Change-Id: I66dbbb4d8795cf1274fc721bf0e8769b8f757c31
2014-03-03Implement lcd_get_dpi() for X-Fi Style (aka fix red).Thomas Martitz1-2/+2
Change-Id: I9cab3a08c9e28558db2c16e8df30afcef6e11004
2014-03-03Fix greyscale and mono builds.Thomas Martitz2-3/+6
Change-Id: I511376eb42d1109dbe10d4fc5aa849b21d9f7582
2014-03-03Touchscreen: Show a line separator in lists.Thomas Martitz10-14/+116
This patch adds a configurable line separator between list items, very similar to lists in Android. Additionally, below the list item there is a thicker line. It can be disabled in the settings. Its color can be configured as well. Remote and monochrome displays are explicitly unsupported. If there is desire this can be changed but it doesn't seem useful to me. Change-Id: I005313b0d8f5ecd15864bf20e66ea4e3390d8b7d
2014-03-03wpsbuild: The setting strings can now contain an additional feature constraint.Thomas Martitz1-24/+50
The setting strings are now of the form setting[.resolution[&feature]] (resolution can be a regex, .+ to match all resultions). will check against features.txt to see if the target meets it. This can be used, for example, to add/override default settings for touchscreen devices. Change-Id: I2eafa02f10b362a8e7de8e1f3a53115e70d28084
2014-03-03Implement lcd_get_dpi() for all bitmap targets.Thomas Martitz85-8/+171
DPI values are autogenerated from the resolution and collected display size values. These values are inserted as comments as well. Change-Id: Id03aedf9af18348f773dfce002805e480f6751e8
2014-03-03Fix regression introduced by eec89a9.Thomas Martitz1-3/+3
The icon pixel position must be calculated before y is modified. Change-Id: I16947dd8862683530dd7c5f00cfa7cf8d9636e61
2014-03-03ingenic: Tweak a few details in crt0.SMarcin Bukat1-13/+14
1) Avoid load/store delay slot by reorganizing instructions in copy loops 2) Fix off-by-one error in cache initialization code. This was harmless as it simply set line0 twice, now it sets every cacheline only once. 3) Fix off-by-word error in .bss clearing loop. The addiu in branch delay slot even if calculated is not seen by the branch instruction itself, so the code did one word too much in clearing. 4) Fix off-by-word error in deadbeefing stack. See above. Change-Id: Iabb09a55979de7aa2e2b9234273683fc7e9762c5
2014-03-03Fix FS#12951: The icons could be a pixel to far down.Thomas Martitz1-1/+5
Since eec89a9 icons have been centered using same calculation as for fonts. In edge cases this is visually different from before and didn't align well to the font's baseline. Revert to the old calculation just for centering icons to fix. A proper aligorithm would take the baseline into account but this has worked sufficiently well for us (fix this if needed) Change-Id: I86593529b16cd28ae4552641e216e73795f2450c
2014-03-03Fix hostfs_init() return value.Thomas Martitz5-5/+9
Change-Id: Ic8048e3fa0075de234e8879ba9faad101168bf09
2014-02-28application(ypr0): Fix mount option selected by the current codepage.Thomas Martitz1-16/+16
A missing comma messed the lookup table up :'( Be aware that the kernel on the ypr0 doesn't support many codepages in which case its default (iso8859) will be used. Change-Id: Iaa4a4b8338fb047d29f6a4e2a80eba5c296d27ce
2014-02-26application: fix opendir() for "/" directory.Thomas Martitz1-1/+1
"/" couldn't be opened. This regression was introduced by 731072d. Trailing slash should not be stripped for the root directory. Change-Id: Idd2d9b1f33e7be789bcf432e586e6715eb4152f4
2014-02-25test_disk: Add an additional dirscan test that calls dir_get_info() also.Thomas Martitz1-0/+25
Especially on hosted dir_get_info() can be relatively. It is commonly called after every readdir() for further information about the dir entry. So it's peformance compared to readdir()-only is interesting. Change-Id: I03ab41fb190acf738e04a5d3b2a4fe29276094f6