SanDisk Sansa c200v2, m200v4, clipv1, clipv2, clip+, and fuzev2

[bilditup1]

I've come across something weird - after fiddling with the Hotkey settings (new feature from 25906), when going back to the WPS, the Album Art wouldn't show. Had to go to the next track for it to show. Dunno how reproducible it is yet though. (Build: 25916)

-e

[funman]

I disabled dynamic CPU frequency in r25924

Power use will increase a bit but at least it will confirm that there are still weird crashes even with this functionality off

(well I hope it will still crash for everyone because for me it does)

[cris]

Sorry if this is a stupid question.  How can I update to a more recent bootloader.  I have a pre-patched OF provided by Funman a few weeks back.  Where can I find the updated bootloader (the website still shows 22 March - I appreciate these are the stable versions)?

Thanks

Cris

[marc2003]

(well I hope it will still crash for everyone because for me it does)

well my clip v2 has never crashed when boosted (touch wood).

i ran into this stability problem over a month ago now and i remember on ABI you telling me to boost the cpu frequency. since then, it's been rock solid. admittedly i only use it for an hour or 2 a day but i have successfully completed several battery benchmarks without issue.

unboosted, the longest i've managed is 50 minutes. and sometimes it will crash within seconds of starting playback. i've never seen any error messages though. most of the time it just goes off. one time the screen just froze and i had to hold the power button to turn it off

@cris. at the moment, i think your only option would be to read the "how to compile" guide found on the wiki pages.

[MasterFen]

I'd like to report a bug with the Clip+ port (although it probably happens in all the other Sansa AMS ports too) regarding clearing all my settings back to default.

It does what it's meant to do but for some reason whenever the line selector is in any subfolders in the "Plugins" section, it does not show; it scrolls and selects things as usual except you have no idea what you've selected. The problem seems to go away after a reboot back into Rockbox.

Can anybody see if it happens in any of the other AMS ports?

[ranma]

GPIO registers of OF vs Rockbox on C200v2.
Last time I tried reading those I used the wrong address (0xc80[bcde]0000 instead of 0xc80[bcde]03fc) for reading the data and always got 0 because of that (low address lines are ANDed with the data byte on reads and select which bits are written to on writes). RIS is kind of interesting because it shows which bits have seen a high->low transition (reset by writing to IC).

Code: [Select]

OF, USB connected, both buttonlights + display on
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 000000a7 000000b3 00000000 00000000 00000000 00000000 00000012 00000000 00000000 00000000
B 0000007e 00000070 00000000 00000000 00000001 00000001 0000007c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000009f 00000080 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

OF, USB disconnected, both buttonlights + display on
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 000000a7 000000b3 00000000 00000000 00000000 00000000 00000012 00000000 00000000 00000000
B 0000007e 00000070 00000000 00000000 00000001 00000001 0000007c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000009f 00000080 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

OF, USB disconnected, buttonlights + display off
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 00000007 000000b3 00000000 00000000 00000000 00000000 000000b2 00000000 00000000 00000000
B 0000007e 00000070 00000000 00000000 00000001 00000001 0000007c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000001f 00000080 00000000 00000000 00000000 00000000 00000080 00000000 00000000 00000000

Rockbox (locally modified r25924), USB disconnected, main buttonlight + display on
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 000000a5 000000b3 00000000 00000004 00000000 00000004 000000b2 00000000 00000000 00000000
B 0000007e 00000070 00000000 00000000 00000000 00000000 0000007c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000001f 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

Rockbox (locally modified r25924), USB disconnected, buttonlights + display off
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 00000005 000000b3 00000000 00000004 00000000 00000004 000000b2 00000000 00000000 00000000
B 0000007e 00000070 00000000 00000000 00000000 00000000 0000007c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000001f 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

Rockbox (vanilla r25924), USB disconnected, main buttonlight + display on
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 000000b6 000000a0 00000000 00000004 00000000 00000004 000000a0 00000000 00000000 00000000
B 0000007e 00000030 00000000 00000000 00000000 00000000 0000003c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000001f 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

Rockbox (vanilla r25924), USB disconnected, buttonlights + display off
  DATA     DIR      IS       IBE      IEV      IE       RIS      MIS      IC       AFSEL
A 00000016 000000a0 00000000 00000004 00000000 00000004 000000a0 00000000 00000000 00000000
B 0000007e 00000030 00000000 00000000 00000000 00000000 0000003c 00000000 00000000 0000000c
C 000000ff 00000000 00000000 00000000 00000000 00000000 000000ff 00000000 00000000 000000ff
D 0000001f 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

I'm guessing that port B0 (interrupt enabled on OF) may be used for remote control via the docking connector (http://www.amazon.com/dp/B000IBJKIE).

My locally modified version sets the LCD reset GPIO (B6) and the unknown port A GPIOs to output to match OF more closely.

[vadik]

(well I hope it will still crash for everyone because for me it does)

well my clip v2 has never crashed when boosted (touch wood).

Same here.  Clip v2, r25914 (both bootloader and rockbox).  Without boosting the CPU, crashes within two minutes (playing an MP3); boosted, survives for an hour.

Will it help if I run battery bench on my clip v2, or do we have enough statistics?

[funman]

r25914

Please don't report bugs with builds older than r25924 (and always use very recent builds because development happens fast on those)

[bilditup1]

There are some interesting uSD-power-related patches in 25936...charging up my battery for an overnight test (it probably won't finish 'till tomorrow afternoon).

[Xanikseo]

I have no idea if this is due to rockbox on the fuze, or a problem with rockbox in general, but there are problems playing OGGs with 48KHz sample frequency. There is a sharp crackling in the background, making it impossible to listen to.

http://www.mediafire.com/?myjmdtjyyt5
Screw that, just a case of clipping on RG.
It isn't RG clipping, it's just that using RG with a 48KHz sample rate causes this crackling noise.

BTW haven't had any crashes so far with cpu frequency scaling disabled.

[saratoga]

I have no idea if this is due to rockbox on the fuze, or a problem with rockbox in general, but there are problems playing OGGs with 48KHz sample frequency. There is a sharp crackling in the background, making it impossible to listen to.

Here is such a sample http://www.mediafire.com/?tdgztnm3hnm.

Try it in the sim, and then file a bug report.

[Xanikseo]

Argh simulator is crashing atm in wine whatever file I play (didn't used to before). Trying in VM will post results here.
Ok, seems to be a problem with RB in general. Filing bug report - done, link

[GodEater]

Why on earth would you run the sim in Wine? It compiles just fine for linux.

[bYOndo]

Please don't report bugs with builds older than r25924 (and always use very recent builds because development happens fast on those)

r25939-100510: my clipv2 seems much more stable now!
I tried my personal "pitch stressing test" many times and no "undefined instruction" or "data abort" has happened.
crossed fingers...
I didn't modified anything about cpu boosting, I don't even know how I can do that, so it's just the default settings.

Massimo

[bilditup1]

New battery bench is attached for build 25936.

I kept all the settings the same:
5-track jazz album, 40 minutes, ~190-210kbps mp3
Volume -20db
Flat EQ
No backlight
No RG
Headphones plugged in

Time was 15:24. This is 24 minutes shorter than build 25821. I guess this is due to the changes funman made in 25924? I don't know how much, if at all, this was mitigated by saratoga's changes in 25936 to sd-as3525v2.c.