FS#325 - CPUAdrEr at 0f00079a during playback

Attached to Project: Rockbox
Opened by Anonymous Submitter - Monday, 30 September 2002, 18:58 GMT
Last edited by Daniel Stenberg (bagder) - Friday, 11 October 2002, 06:26 GMT
Task Type Bugs
Category Music playback
Status Closed
Assigned To Linus Nielsen Feltzing (linusnielsen)
Operating System
Severity Low
Priority Normal
Reported Version Release 1.3
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No


Archos Jukebox Recorder 10
Built-in firmware 1.27b
Rockbox 1.3

Selected a playlist, used >> to skip the first albums
worth of MP3's, and then allowed it to play. On about
the 4th track, I received an:
at 0f00079A
error. I shutdown and restarted, attempted the same
sequence, but it worked fine the second time.

This task depends upon

Closed by  Daniel Stenberg (bagder)
Friday, 11 October 2002, 06:26 GMT
Reason for closing:  Fixed
Comment by Anonymous Submitter - Wednesday, 02 October 2002, 23:56 GMT

Well, this is getting downright annoying. This error appears
every 15-60 minutes while playing on my Recorder. Is
anyone else having this problem?


Comment by David Ashman (ashmand) - Thursday, 03 October 2002, 04:30 GMT

I've actually seen this on my Recorder 15 a couple times. I
only got it to do it twice, both times in the same folder of
MP3s. Hasn't happened since, though.

Comment by Björn Stenberg (zagor) - Thursday, 03 October 2002, 08:17 GMT

Does this happen with the daily builds too, or just 1.3?
Comment by David Ashman (ashmand) - Thursday, 03 October 2002, 14:00 GMT

It happened those two times on 1.3. I did get it to do it once
on the daily build (10/1 I believe), but I'm using 10/1 now and
it's been fine.

Comment by David Ashman (ashmand) - Monday, 07 October 2002, 14:30 GMT

I just got it again. This time the error looked like:

at 0F00084A

This is on a recorder 15 with 9/30 version.

Comment by Linus Nielsen Feltzing (linusnielsen) - Tuesday, 08 October 2002, 06:28 GMT

Just double checking, are you absolutely sure that you typed the error
message correctly? And are you absolutely sure that you ran the 9/30
daily build?

What did you do when that last crash happened?
Battery status?

The latest crash is not the same that you had in
the original report. This looks like a serious bug...
Comment by Linus Nielsen Feltzing (linusnielsen) - Tuesday, 08 October 2002, 08:16 GMT

I have found and fixed a bug in the mpeg code (the bitswapping) that could
cause CPUAddr errors. Please try the latest daily build and see if the
crashes are gone.
Comment by David Ashman (ashmand) - Tuesday, 08 October 2002, 13:03 GMT

I double checked the version, and it was definitely the 9/30
build. I also made sure to get the memory address correct in
the error. I'm not sure about the battery status, but I did have
the adapter plugged in and have the "drain and charge" option
turned on, so the battery could have been low at the time. If I
see it again, I'll check that. Is there any other data I should
collect to help you debug?
Comment by Daniel Stenberg (bagder) - Tuesday, 08 October 2002, 15:08 GMT

Yes, you should try the daily build that arrives at october
9, or one of the current bleeding edge builds. This problem
is believed to have been fixed in CVS.
Comment by Anonymous Submitter - Wednesday, 09 October 2002, 19:16 GMT

The October 9 build seems to have solved this problem for

Thanks Linus.