Bug 133750

Summary: Writing multitrack audio CD fails with OOM
Product: [Fedora] Fedora Reporter: vvs <vvs009>
Component: kernelAssignee: Dave Jones <davej>
Status: CLOSED RAWHIDE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: rawhideCC: pfrields, wtogami
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2004-10-15 05:27:03 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
output of dmesg
none
output of free
none
contents of /proc/meminfo
none
contents of /var/log/messages
none
contents of /proc/slabinfo
none
contents of /proc/vmstat none

Description vvs 2004-09-27 06:54:26 UTC
Description of problem:

When I try to write multitrack audio CD it fails with OOM. Writing one
track data CD works ok.

My configuration:
RAM 128MB
ATAPI CD-RW Teac CD-W54E (firmware 1.1Y)

Notes: It doesn't depend on whether I write it using cdrecord or
cdrdao. It happens with FC2 stock versions of cdrdao and cdrecord as
well. It doesn't matter if I write single .bin image or multiple .wav
files. It doesn't depend on swap size or if there is no swap at all.
But it doesn't happen on the box with 512MB of RAM with USB CD/DVD-RW
drive.

Version-Release number of selected component (if applicable):

kernel-2.6.8-1.521
cdrdao-1.1.9-5
cdrecord-2.01.1-2

How reproducible:

Always (but see above notes)

Steps to Reproduce:
1. Make sure you have 128MB of RAM and ATAPI CD-R/RW drive
2. Create an image of multitrack audio CD or rip it into .wav files
3. cdrdao write --device ATAPI:0,0,0 image.toc
  
Actual results:

After writing about 266-268MB and 5 tracks the buffers drop from 100%
to 81% and then OOM killer killed it.

Expected results:

It should write it to the end.

Additional info:

It worked before on 2.4.x kernel series. And there are suspicious
messages from the kernel about DMA memory.

Comment 1 vvs 2004-09-27 06:55:27 UTC
Created attachment 104349 [details]
output of dmesg

Comment 2 vvs 2004-09-27 06:56:16 UTC
Created attachment 104350 [details]
output of free

Comment 3 vvs 2004-09-27 06:56:48 UTC
Created attachment 104351 [details]
contents of /proc/meminfo

Comment 4 vvs 2004-09-27 06:57:36 UTC
Created attachment 104352 [details]
contents of /var/log/messages

Comment 5 vvs 2004-09-27 06:58:18 UTC
Created attachment 104353 [details]
contents of /proc/slabinfo

Comment 6 vvs 2004-09-27 06:59:01 UTC
Created attachment 104354 [details]
contents of /proc/vmstat

Comment 7 vvs 2004-10-08 07:16:23 UTC
Seems that it was fixed in 2.6.8.1-mm2, according to announcement.
Will it be merged with RH kernel?


Comment 8 vvs 2004-10-15 05:26:10 UTC
Well, it seems that the fix was merged at least in kernel-2.6.8-1.607