Flyspray:: http://www.rockbox.org/tracker/ Flyspray:: Rockbox: Recently opened tasks 2013-06-18T09:02:41Z FS#12874: [Fuze+] Music playback stopped if touchpad pressed more than 6 second while locked http://www.rockbox.org/tracker/task/12874 2013-06-18T09:02:41Z Jean-Louis Biasini I've been having this bug for over mouths and mouths without being able to square it clearly (i.e. even before the new radio implementation). Way to reproduce: - Lock the device in WPS while playing music - put your finger on the touchpad and let it there for more than 6 sec. - the music stop playing, player seems to be frozen. - press the lock key: player won't unlock but will let playback continue and will then act as normal (i.e. second press unlock the device) Way to reproduce:
- Lock the device in WPS while playing music
- put your finger on the touchpad and let it there for more than 6 sec.
- the music stop playing, player seems to be frozen.
- press the lock key: player won't unlock but will let playback continue and will then act as normal (i.e. second press unlock the device)]]> FS#12873: Album art does not parse in ID3v2 tag http://www.rockbox.org/tracker/task/12873 2013-06-15T22:42:45Z MichaelGiacomelli http://web.mit.edu/mgg6/www/Game_Scoop_Presents__Predicting_Nintendos_2013_E3.mp3 The ID3v2 tag in this file contains a fair amount of random metadata which seems to confuse Rockbox's parser such that album art is not displayed. Extracting the album art using foobar2000, deleting the existing tag and then reattaching it fixes display of the embedded JPEG so the jpeg itself is fine. Tested with today's build (7c7fa36) on device (e200v1) and in sim. http://web.mit.edu/mgg6/www/Game_Scoop_Presents__Predicting_Nintendos_2013_E3.mp3

The ID3v2 tag in this file contains a fair amount of random metadata which seems to confuse Rockbox's parser such that album art is not displayed. Extracting the album art using foobar2000, deleting the existing tag and then reattaching it fixes display of the embedded JPEG so the jpeg itself is fine.

Tested with today's build (7c7fa36) on device (e200v1) and in sim.]]> FS#12872: Rock box can only support max 2 channels ogg_vorbis data http://www.rockbox.org/tracker/task/12872 2013-06-13T17:57:21Z ligd There are lots of vorbis files which channels more than 2 channels, we don't support? FS#12871: rockbox play ogg_vorbis data has noise when enable the macro CPU_ARM http://www.rockbox.org/tracker/task/12871 2013-06-13T11:30:19Z ligd in the file fft-ffmpeg_arm.h in the function XNPROD31_R, use the t5 and t6 as input data while ouput data, there maybe has a error. static inline FFTComplex* TRANSFORM_W01( FFTComplex* z, int n, const FFTSample* w ) { register FFTSample t1,t2 asm("r5"),t5 asm("r6"),t6 asm("r7"),r_re asm("r8"),r_im asm("r9"); /* load wre,wim into t5,t6 */ asm volatile( "ldmia %[w], {%[wre], %[wim]}\n\t":[wre] "=r" (t5), [wim] "=r" (t6):[w] "r" (w)); z += n*2; /* z[o2] -- 2n * 2 since complex numbers */ asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z)); XPROD31_R(r_re, r_im, t5 /*wre*/, t6 /*wim*/, t1,t2); z += n; /* z[o3] */ asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z)); XNPROD31_R(r_re, r_im, t5 /*wre*/, t6 /*wim*/, t5,t6); } I suggest use like this, maybe a butter way, but I am not sure: static inline FFTComplex* TRANSFORM_W01( FFTComplex* z, int n, const FFTSample* w ) { register FFTSample t1, t3, t4, t2 asm("r5"),t5 asm("r6"),t6 asm("r7"),r_re asm("r8"),r_im asm("r9"); /* load wre,wim into t5,t6 */ asm volatile( "ldmia %[w], {%[wre], %[wim]}\n\t":[wre] "=r" (t3), [wim] "=r" (t4):[w] "r" (w)); z += n*2; /* z[o2] -- 2n * 2 since complex numbers */ asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z)); XPROD31_R(r_re, r_im, t3 /*wre*/, t4 /*wim*/, t1,t2); z += n; /* z[o3] */ asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z)); XNPROD31_R(r_re, r_im, t3 /*wre*/, t4 /*wim*/, t5,t6); } in the function XNPROD31_R, use the t5 and t6 as input data while ouput data,
there maybe has a error.

static inline FFTComplex* TRANSFORM_W01( FFTComplex* z, int n, const FFTSample* w )
{
register FFTSample t1,t2 asm("r5"),t5 asm("r6"),t6 asm("r7"),r_re asm("r8"),r_im asm("r9");

/* load wre,wim into t5,t6 */
asm volatile( "ldmia %[w], {%[wre], %[wim]}\n\t":[wre] "=r" (t5), [wim] "=r" (t6):[w] "r" (w));
z += n*2; /* z[o2] -- 2n * 2 since complex numbers */
asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z));
XPROD31_R(r_re, r_im, t5 /*wre*/, t6 /*wim*/, t1,t2);

z += n; /* z[o3] */
asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z));
XNPROD31_R(r_re, r_im, t5 /*wre*/, t6 /*wim*/, t5,t6);
}

I suggest use like this, maybe a butter way, but I am not sure:
static inline FFTComplex* TRANSFORM_W01( FFTComplex* z, int n, const FFTSample* w )
{
register FFTSample t1, t3, t4, t2 asm("r5"),t5 asm("r6"),t6 asm("r7"),r_re asm("r8"),r_im asm("r9");

/* load wre,wim into t5,t6 */
asm volatile( "ldmia %[w], {%[wre], %[wim]}\n\t":[wre] "=r" (t3), [wim] "=r" (t4):[w] "r" (w));
z += n*2; /* z[o2] -- 2n * 2 since complex numbers */
asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z));
XPROD31_R(r_re, r_im, t3 /*wre*/, t4 /*wim*/, t1,t2);

z += n; /* z[o3] */
asm volatile( "ldmia %[my_z], {%[r_re],%[r_im]}\n\t":[r_re] "=r" (r_re), [r_im] "=r" (r_im):[my_z] "r" (z));
XNPROD31_R(r_re, r_im, t3 /*wre*/, t4 /*wim*/, t5,t6);
}]]> FS#12870: If More than One JPEG Embedded Into Tag Neither Is Displayed http://www.rockbox.org/tracker/task/12870 2013-06-16T17:48:35Z Shawn Recupido I think the summary is pretty self explanatory. I use a program called gpodder that adds a cover automatically from a file post download of a podcast. If the author though has their own cover embedded into the file then rockbox will display neither. Is there any way this can be fixed? FS#12869: ClipV1 opus file playback: no metadata, and some (low bitrate) files crash the player http://www.rockbox.org/tracker/task/12869 2013-06-06T02:46:57Z Ivan Privaci First noticed the issue in the previous dev build, and just confirmed in 15fa7f8. When playing .opus files, no metadata is displayed. Some files (92kbit stereo music files so far) do play fine otherwise. Others induce crashes upon attempting to play. An example is attached. First test, I played a couple of music files successfully, then tried to play the attached file. System crashed immediately showing the following: "Undefined instruction pc:30000002 sp:300632[display cuts off here] bt end" Powered off, restarted, and tried to play the attached file immediately. After a second or so of audio, player crashed again: "Undefined instruction pc:08180002 sp:302029[display cuts off here] bt end" On the previous dev build, I was getting similar errors if I tried to play the attached file (or several other similar ones) after successfully playing other files, but would get a black-screen crash if I tried to play it immediately on powerup (this is not what is happening on 15fa7f8). I am NOT experiencing these problems on another AMSv1 player (Sansa c240 ["c200v1"]) with the same dev build, so this may be specific to the Clipv1. On the c240 with both the previous build and 15fa7f8, metadata displays fine and the attached file plays without any apparent error.
When playing .opus files, no metadata is displayed. Some files (92kbit stereo music files so far) do play fine otherwise. Others induce crashes upon attempting to play. An example is attached.

First test, I played a couple of music files successfully, then tried to play the attached file. System crashed immediately showing the following:
"Undefined instruction
pc:30000002 sp:300632[display cuts off here]
bt end"

Powered off, restarted, and tried to play the attached file immediately. After a second or so of audio, player crashed again:
"Undefined instruction
pc:08180002 sp:302029[display cuts off here]
bt end"

On the previous dev build, I was getting similar errors if I tried to play the attached file (or several other similar ones) after successfully playing other files, but would get a black-screen crash if I tried to play it immediately on powerup (this is not what is happening on 15fa7f8).

I am NOT experiencing these problems on another AMSv1 player (Sansa c240 ["c200v1"]) with the same dev build, so this may be specific to the Clipv1. On the c240 with both the previous build and 15fa7f8, metadata displays fine and the attached file plays without any apparent error.]]> FS#12868: keylock doesn't work in fm receiver mode http://www.rockbox.org/tracker/task/12868 2013-05-25T16:55:50Z a wulf enjoyinf cartography :P Hi, I noticed that you cannot enter keylock mode in FM radio. I run on a SanDisk Clip Zip 3.13 and press Select+Home which works fine in the player mode, but it doens't work on the radio. Not sure if this is platform specific bug or a general one. I run on a SanDisk Clip Zip 3.13 and press Select+Home which works fine in the player mode, but it doens't work on the radio. Not sure if this is platform specific bug or a general one.]]> FS#12867: normalise artist names http://www.rockbox.org/tracker/task/12867 2013-05-25T23:56:00Z a wulf enjoyinf cartography :P Currently a single artist is displayed multiple times in the database view. Steps to reproduce: -Add multiple albums from the same artist with slightly different ID3 tags -Browse via database-artists Current behaviour: Artists appear multiple times looking each the same. This confuses the enduser as he doesn't know which one is associated with which albums/tracks etc... Expected behaviour: The player returns a normalized and flattened list that internally clusters all name variations but presents them under the same label. Possible sollutions: On building the database, the scanner should create a normlized_artist_key that: -is lowercase -doesn't include leading or tailing spaces -doesn't include special fromating carracters -triggers on special substrings as. "feat." and splits up both artist names
Steps to reproduce:
-Add multiple albums from the same artist with slightly different ID3 tags
-Browse via database-artists

Current behaviour:
Artists appear multiple times looking each the same. This confuses the enduser as he doesn't know which one is associated with which albums/tracks etc...

Expected behaviour:
The player returns a normalized and flattened list that internally clusters all name variations but presents them under the same label.

Possible sollutions:
On building the database, the scanner should create a normlized_artist_key that:
-is lowercase
-doesn't include leading or tailing spaces
-doesn't include special fromating carracters
-triggers on special substrings as. "feat." and splits up both artist names]]> FS#12866: Radio preset with RDS tag http://www.rockbox.org/tracker/task/12866 2013-05-25T23:54:44Z a wulf enjoyinf cartography :P Can you please add the radio station name from RDS as default value for the preset name? The name is easier to identify/remember than the frequency. Thus a autoscan should create a list of friendly readable presets and the user will find the desired station very quickly. I use RockBox 3.12. at a Samsung Clip Zip The name is easier to identify/remember than the frequency. Thus a autoscan should create a list of friendly readable presets and the user will find the desired station very quickly.

I use RockBox 3.12. at a Samsung Clip Zip]]> FS#12865: 5G iPod USB connection errors and full speed mode if hard drive is spinning when connecting http://www.rockbox.org/tracker/task/12865 2013-05-24T19:44:38Z Boris Gjenero With my 5th generation iPod, if I connect USB while the hard drive is spinning, I often get full speed mode instead of high speed mode. This testing is with 00b8563, but I've observed the same problem in the past. Here's the relevant part of Linux 3.8.0-21-generic dmesg output: [26644.712035] usb 2-2: new high-speed USB device number 18 using ehci-pci [26644.904471] usb 2-2: device descriptor read/all, error -71 [26644.965905] ehci-pci 0000:00:1d.7: port 2 reset error -110 [26644.967780] ehci-pci 0000:00:1d.7: port 2 reset error -110 [26644.968004] ehci-pci 0000:00:1d.7: port 2 reset error -110 [26644.971577] ehci-pci 0000:00:1d.7: port 2 reset error -110 [26644.973440] ehci-pci 0000:00:1d.7: port 2 reset error -110 [26644.973459] hub 2-0:1.0: hub_port_status failed (err = -32) [26645.176048] hub 2-0:1.0: unable to enumerate USB device on port 2 [26645.428038] usb 6-2: new full-speed USB device number 8 using uhci_hcd [26645.979070] usb 6-2: not running at top speed; connect to a high speed hub [26646.009069] usb 6-2: New USB device found, idVendor=05ac, idProduct=1209 [26646.009076] usb 6-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [26646.009080] usb 6-2: Product: Rockbox media player [26646.009083] usb 6-2: Manufacturer: Rockbox.org [26646.009087] usb 6-2: SerialNumber: 100000000000A270014B32BEB [26646.024242] scsi21 : usb-storage 6-2:1.0 [26647.038109] scsi 21:0:0:0: Direct-Access TOSHIBA MK3008GAL BU11 PQ: 0 ANSI: 4 (I've removed the remaining USB storage messages because they don't seem relevant) If the hard drive is not spinning, I get high speed mode. However, there is a disconnect and the device is detected twice: [27356.784126] usb 6-2: USB disconnect, device number 9 [27371.208044] usb 2-2: new high-speed USB device number 24 using ehci-pci [27371.347053] usb 2-2: New USB device found, idVendor=05ac, idProduct=1209 [27371.347058] usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [27371.347062] usb 2-2: Product: Rockbox media player [27371.347065] usb 2-2: Manufacturer: Rockbox.org [27371.347069] usb 2-2: SerialNumber: 100000000000A270014B32BEB [27373.079038] scsi25 : usb-storage 2-2:1.0 [27373.337244] usb 2-2: USB disconnect, device number 24 [27373.640029] usb 2-2: new high-speed USB device number 25 using ehci-pci [27373.779220] usb 2-2: New USB device found, idVendor=05ac, idProduct=1209 [27373.779226] usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [27373.779229] usb 2-2: Product: Rockbox media player [27373.779233] usb 2-2: Manufacturer: Rockbox.org [27373.779236] usb 2-2: SerialNumber: 100000000000A270014B32BEB [27373.820382] scsi26 : usb-storage 2-2:1.0 [27374.823266] scsi 26:0:0:0: Direct-Access TOSHIBA MK3008GAL BU11 PQ: 0 ANSI: 4 (Again, I've removed the remaining USB storage messages because they don't seem relevant) I get the same behaviour regarding high speed vs. full speed mode in Windows 7, but I'm not aware of Windows reporting other USB errors anywhere. I was sometimes having USB difficulties that seemed to be due to contact oxidization or dust from not using the dock connector in a long time, but I don't think that's the cause here. I'm attaching USB traces from WireShark using Linux usbmon. In both cases I removed all the USB mass storage stuff after get max lun because it seems irrelevant and increases file size a lot. When the result is full speed, something goes wrong after the set address response. When the result is high speed, something goes wrong after the set configuration request. In both cases the first sign of trouble is an interrupt from the USB hub. This is on a Gigabyte GA-P35-DS3R rev V1.0 motherboard with Intel P35 + ICH9R chipset.
[26644.712035] usb 2-2: new high-speed USB device number 18 using ehci-pci
[26644.904471] usb 2-2: device descriptor read/all, error -71
[26644.965905] ehci-pci 0000:00:1d.7: port 2 reset error -110
[26644.967780] ehci-pci 0000:00:1d.7: port 2 reset error -110
[26644.968004] ehci-pci 0000:00:1d.7: port 2 reset error -110
[26644.971577] ehci-pci 0000:00:1d.7: port 2 reset error -110
[26644.973440] ehci-pci 0000:00:1d.7: port 2 reset error -110
[26644.973459] hub 2-0:1.0: hub_port_status failed (err = -32)
[26645.176048] hub 2-0:1.0: unable to enumerate USB device on port 2
[26645.428038] usb 6-2: new full-speed USB device number 8 using uhci_hcd
[26645.979070] usb 6-2: not running at top speed; connect to a high speed hub
[26646.009069] usb 6-2: New USB device found, idVendor=05ac, idProduct=1209
[26646.009076] usb 6-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[26646.009080] usb 6-2: Product: Rockbox media player
[26646.009083] usb 6-2: Manufacturer: Rockbox.org
[26646.009087] usb 6-2: SerialNumber: 100000000000A270014B32BEB
[26646.024242] scsi21 : usb-storage 6-2:1.0
[26647.038109] scsi 21:0:0:0: Direct-Access TOSHIBA MK3008GAL BU11 PQ: 0 ANSI: 4
(I've removed the remaining USB storage messages because they don't seem relevant)

If the hard drive is not spinning, I get high speed mode. However, there is a disconnect and the device is detected twice:

[27356.784126] usb 6-2: USB disconnect, device number 9
[27371.208044] usb 2-2: new high-speed USB device number 24 using ehci-pci
[27371.347053] usb 2-2: New USB device found, idVendor=05ac, idProduct=1209
[27371.347058] usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[27371.347062] usb 2-2: Product: Rockbox media player
[27371.347065] usb 2-2: Manufacturer: Rockbox.org
[27371.347069] usb 2-2: SerialNumber: 100000000000A270014B32BEB
[27373.079038] scsi25 : usb-storage 2-2:1.0
[27373.337244] usb 2-2: USB disconnect, device number 24
[27373.640029] usb 2-2: new high-speed USB device number 25 using ehci-pci
[27373.779220] usb 2-2: New USB device found, idVendor=05ac, idProduct=1209
[27373.779226] usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[27373.779229] usb 2-2: Product: Rockbox media player
[27373.779233] usb 2-2: Manufacturer: Rockbox.org
[27373.779236] usb 2-2: SerialNumber: 100000000000A270014B32BEB
[27373.820382] scsi26 : usb-storage 2-2:1.0
[27374.823266] scsi 26:0:0:0: Direct-Access TOSHIBA MK3008GAL BU11 PQ: 0 ANSI: 4
(Again, I've removed the remaining USB storage messages because they don't seem relevant)

I get the same behaviour regarding high speed vs. full speed mode in Windows 7, but I'm not aware of Windows reporting other USB errors anywhere.

I was sometimes having USB difficulties that seemed to be due to contact oxidization or dust from not using the dock connector in a long time, but I don't think that's the cause here.

I'm attaching USB traces from WireShark using Linux usbmon. In both cases I removed all the USB mass storage stuff after get max lun because it seems irrelevant and increases file size a lot. When the result is full speed, something goes wrong after the set address response. When the result is high speed, something goes wrong after the set configuration request. In both cases the first sign of trouble is an interrupt from the USB hub.

This is on a Gigabyte GA-P35-DS3R rev V1.0 motherboard with Intel P35 + ICH9R chipset.]]>