kio_audiocd eats up CPU instead of displaying CD contents

Bug #48731 reported by Dukai Gábor
14
Affects Status Importance Assigned to Milestone
KDE Multimedia
Fix Released
Unknown
kdemultimedia (Ubuntu)
Fix Released
Medium
Rich Johnson

Bug Description

Binary package hint: kdemultimedia-kio-plugins

Either using directly the audiocd:// ioslave or starting it from kaudiocreator.
Dmesg shows this kind of errors:

sg_write: data in/out 12/12 bytes for SCSI command 0x43--guessing data in;
program kio_audiocd not setting count and/or reply_len properly

cdparanoia is able to rip the CD.
I have a sata dvdrw drive.

Revision history for this message
In , Graham Knap (graham-knap-rogers) wrote : This printk was introduced in Linux 2.6.10

It looks like the kernel code change that made this problem visible
occurred in 2.6.10. I've attached part of a diff between "sg.c" of
Linux 2.6.9 versus 2.6.10.

Interestingly, it appears that the diagnostic message is the only thing
that was added. You can see that the code does not alter any variables,
etc.

So the question remains: is this printk erroneous... or does it
correctly identify a bug in cdparanoia?

-- graham

Revision history for this message
In , Afayolle-nerim (afayolle-nerim) wrote : merging 305684 298648

merge 305684 298648

Revision history for this message
In , Daniel Baumann (daniel-baumann) wrote : [cdparanoia] Closing bug: cdparanoia generates syslog messages

Hi,

I'm closing this bug. You reported, that the extraction part using
sound-juicer took so long. This may be true, especially if you use a
USB-attached cdrom which the kernel often has some problems with cdda
extraction. It is not related to the fact, that something was written
into syslog (which are, btw, kernel-messages, and not cdparanoia messages).

Regards,
Daniel

--
Address: Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email: <email address hidden>
Internet: http://people.panthera-systems.net/~daniel-baumann/

Revision history for this message
In , Graham Knap (graham-knap) wrote : Re: Bug#298648 acknowledged by developer ([cdparanoia] Closing bug: cdparanoia generates syslog messages)

Hi Daniel,

I don't think this bug is resolved. I haven't seen a problem with
extraction speed, but there is a ridiculous amount of kernel log
output being generated while cdparanoia is running.

My guess is that this is indeed a kernel bug.

So could you reopen this bug and reassign it to linux-2.6?

thanks

-- graham

On 2/23/06, Debian Bug Tracking System <email address hidden> wrote:
> This is an automatic notification regarding your Bug report
> #298648: cdparanoia generates a lot of warnings from the kernel,
> which was filed against the cdparanoia package.
>
> It has been closed by one of the developers, namely
> Daniel Baumann <email address hidden> (reply to <email address hidden>).
>
> Their explanation is attached below. If this explanation is
> unsatisfactory and you have not received a better one in a separate
> message then please contact the developer, by replying to this email.
>
> Debian bug tracking system administrator
> (administrator, Debian Bugs database)
>
>
>
>
> ---------- Forwarded message ----------
> From: Daniel Baumann <email address hidden>
> To: <email address hidden>
> Date: Thu, 23 Feb 2006 23:18:16 +0100
> Subject: [cdparanoia] Closing bug: cdparanoia generates syslog messages
> Hi,
>
> I'm closing this bug. You reported, that the extraction part using
> sound-juicer took so long. This may be true, especially if you use a
> USB-attached cdrom which the kernel often has some problems with cdda
> extraction. It is not related to the fact, that something was written
> into syslog (which are, btw, kernel-messages, and not cdparanoia messages).
>
> Regards,
> Daniel
>
> --
> Address: Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
> Email: <email address hidden>
> Internet: http://people.panthera-systems.net/~daniel-baumann/

Revision history for this message
In , Daniel Baumann (daniel-baumann) wrote : Reassign

reopen 298648
reassign 298648 linux-2.6
thanks

--
Address: Daniel Baumann, Burgunderstrasse 3, CH-4562 Biberist
Email: <email address hidden>
Internet: http://people.panthera-systems.net/~daniel-baumann/

Revision history for this message
In , Jurij Smakov (jurij) wrote : Re: cdparanoia generates a lot of warnings from the kernel

Hi Graham,

I was able to find this rather old posting on the topic:

http://thread.gmane.org/gmane.comp.audio.cd-paranoia.general/611

Bill Davidsen (second message) mentions some fixes which were supposed to
improve the situation, but it looks like they never made it to mainline
kernels. Can you please confirm that with recent versions of cdparanoia
and kernel you still get messages in the logs? If you do, I might try
pinging the people involved to revive the discussion.

Best regards,

Jurij Smakov <email address hidden>
Key: http://www.wooyd.org/pgpkey/ KeyID: C99E03CC

Revision history for this message
In , Graham Knap (graham-knap) wrote :

Hello Jurij,

Thank you for looking into this problem.

I am ripping a disc now, with cdparanoia 3a9.8-13, and a Linux 2.6.16
kernel that I built April 9 from the Debian-packaged source. I'll
attach the configuration in case anyone is interested.

The messages are still occurring.

I haven't noticed a performance issue. possibly because I have a
reasonably fast PC (AthlonXP 2500+).

-- graham

On 5/6/06, Jurij Smakov <email address hidden> wrote:
> Hi Graham,
>
> I was able to find this rather old posting on the topic:
>
> http://thread.gmane.org/gmane.comp.audio.cd-paranoia.general/611
>
> Bill Davidsen (second message) mentions some fixes which were supposed to
> improve the situation, but it looks like they never made it to mainline
> kernels. Can you please confirm that with recent versions of cdparanoia
> and kernel you still get messages in the logs? If you do, I might try
> pinging the people involved to revive the discussion.
>
> Best regards,
>
> Jurij Smakov <email address hidden>
> Key: http://www.wooyd.org/pgpkey/ KeyID: C99E03CC

Revision history for this message
In , Jurij Smakov (jurij) wrote :

Hello gentlemen,

Some time ago you've participated in a thread regarding the problems
with cdparanoia filling up the kernel logs with lots of messages [0].
You've mentioned that some fixes for this issue were supposed to be pushed
upstream. Unfortunately, they've never made it. We still have a Debian bug
report open (for a while now) [1], and submitter just confirmed that he is
still experiencing this behavior with 2.6.16 kernel. I wonder if anything
can be done about this old bug, which still remains a nuisance.

[0] http://thread.gmane.org/gmane.comp.audio.cd-paranoia.general/611
[1] http://bugs.debian.org/298648

Best regards,

Jurij Smakov <email address hidden>
Debian kernel team
Key: http://www.wooyd.org/pgpkey/ KeyID: C99E03CC

Revision history for this message
In , Graham Knap (graham-knap) wrote : New email address

submitter 315701 !
submitter 264186 !
submitter 264190 !
submitter 298648 !
submitter 315497 !
submitter 322729 !
submitter 227313 !
submitter 338089 !
thanks

Revision history for this message
Dukai Gábor (gdukai) wrote :

Binary package hint: kdemultimedia-kio-plugins

Either using directly the audiocd:// ioslave or starting it from kaudiocreator.
Dmesg shows this kind of errors:

sg_write: data in/out 12/12 bytes for SCSI command 0x43--guessing data in;
program kio_audiocd not setting count and/or reply_len properly

cdparanoia is able to rip the CD.
I have a sata dvdrw drive.

Revision history for this message
Nicolas JEUDY (njeudy) wrote :

I had the same probleme, but since I upgrade with the stable dapper release, it works.
I had sometimes problemes with DRM audio CD ...

Nicolas JEUDY

Revision history for this message
Dukai Gábor (gdukai) wrote :

On some lists I read that enabling storage.media_check_enabled in HAL would fix this but it is already enabled in my system.

Revision history for this message
Bart Cerneels (stecchino) wrote :

I can confirm this bug, I also have a SATA dvd-writer in my asus W3000A laptop. I get the same error messages as Dukai Gábor.

Revision history for this message
Bart Cerneels (stecchino) wrote :

According to this bug http://bugs.kde.org/show_bug.cgi?id=103157
it is a problem with cdparanoia which audiocd and kaudiocreator use for ripping. But it's an old bug and ripping seemed to work for them.
Hope it helps.

Revision history for this message
Dukai Gábor (gdukai) wrote :

Although the error messages still appear in dmesg, upgrading to KDE 3.5.3 has solved this problem.

Changed in kdemultimedia:
status: Unknown → Unconfirmed
Revision history for this message
Thilo (mib) wrote :

same here while trying to play an audio-cd with kscd. I've got an Philips DROM 6216. KDE 3.5.4

the strange thing is, it's only happend with audio-cds, but not with all. I can't replicate this error at all times.

Revision history for this message
Thilo (mib) wrote :

some additional infos to my last mail:

on my system every audio application (eg. grip, kscd) which is using the device directly needs very long to read the cd content or it fails. it is rarely possible that the cpu load is very high. (maybe this isn't the same bug? should we change the subject? or commit a new bug report?)

the following error is found in the syslog if k3b is started:
sr 1:0:0:0: SCSI error: return code = 0x8000002
sr0: Current: sense key: Medium Error
     Additional sense: L-EC uncorrectable error
 [17188027.348000] Assertion failed! qc->n_elem > 0,drivers/scsi/libata-core.c,ata_fill_sg,line=2531

this is happend with different audio-cds. I've tested the cds on players/pcs, there are ok!

Revision history for this message
David H (david-houlder) wrote :

Happening to me too on a Dell Inspiron 9400 - kio_audiocd uses close to 100% cpu and never does anything useful. cdparanoia seems to work OK. Kubuntu 6.06 with all updates to 12 Dec 2006.

While I get the sg_write kernel messages (from both cdparanoia and kio_audiocd), it looks like there is a memory leak in kio_audiocd that is the real problem. 'top' shows the %MEM constantly increasing and strace of the kio_audiocd process shows...

david@david-laptop:~$ strace -p 7037
Process 7037 attached - interrupt to quit
brk(0x126a3000) = 0x126a3000
brk(0x126c4000) = 0x126c4000
brk(0x126e5000) = 0x126e5000
brk(0x12706000) = 0x12706000
brk(0x12727000) = 0x12727000
brk(0x12748000) = 0x12748000
brk(0x12769000) = 0x12769000
brk(0x1278a000) = 0x1278a000
brk(0x127ab000) = 0x127ab000
brk(0x127cc000) = 0x127cc000
brk(0x127ed000) = 0x127ed000
brk(0x1280e000) = 0x1280e000
... and so on.

The drive is...
david@david-laptop:~$ cdparanoia -Qv

Checking /dev/cdrom for cdrom...
        Testing /dev/cdrom for cooked ioctl() interface
                /dev/scd0 is not a cooked ioctl CDROM.
        Testing /dev/cdrom for SCSI interface
                generic device: /dev/sg1
                ioctl device: /dev/scd0

Found an accessible SCSI CDROM drive.
Looking at revision of the SG interface in use...
        SG interface version 3.5.33; OK.

CDROM model sensed sensed: _NEC DVD+-RW ND-6650A 102C

Checking for SCSI emulation...
        Drive is ATAPI (using SCSI host adaptor emulation)
        Couldn't disable kernel command translation layer

Checking for MMC style command set...
        Drive is MMC style
        DMA scatter/gather table entries: 128
        table entry size: 32768 bytes
        maximum theoretical transfer: 1783 sectors
        Setting default read size to 13 sectors (30576 bytes).

Verifying CDDA command set...
        Expected command set reads OK.

Changed in kdemultimedia:
status: New → Fix Released
Revision history for this message
In , Debbugs Internal Request (owner-bugs) wrote : Internal Control

# A New Hope
# A log time ago, in a galaxy far, far away
# something happened.
#
# Magically this resulted in the following
# action being taken, but this fake control
# message doesn't tell you why it happened
#
# The action:
# Bug archived.
thanks
# This fakemail brought to you by your local debbugs
# administrator

Revision history for this message
Rich Johnson (nixternal) wrote :

Marking this as fixed release as I cannot reproduce this issue using the new 3.5.9 packages and seeing as the upstream changes from all reports have been included in our packages. Thank you.

Changed in kdemultimedia:
assignee: nobody → nixternal
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.