Bug 75133

Summary: playing music on smp results in hard freeze
Product: [Retired] Red Hat Linux Reporter: Gene Czarcinski <gczarcinski>
Component: kernelAssignee: Arjan van de Ven <arjanv>
Status: CLOSED NOTABUG QA Contact: Brian Brock <bbrock>
Severity: high Docs Contact:
Priority: high    
Version: 8.0   
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2002-10-07 14:30:58 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
/proc/cpuinfo
none
/proc/devices
none
/proc/driver/emu10k1 info
none
/proc/interrupts
none
lspci
none
lspci -vv
none
lsmod
none
/proc/meminfo none

Description Gene Czarcinski 2002-10-04 18:42:15 UTC
Description of Problem:
The system is a dual processor PIII and I have had the problem with an Ensoniq
sound card (ES1370 drive) and the SB Live sound card (emu10k1 driver).

Using various sound applications such as realplay, xmms, and mplayer, sometimes
the system will hang in a hard freeze (and sound loop) -- requires a hardware
reset or power cycle.  In addition, I also cannot ssh into the system from
another system.  Furthermore, I have this system on a Belkin KVM box and when it
hangs, I cannot use the keyboard or the Belkin switch to change to another
system ... hardware reset is the only way to get out.

When the system hangs, the sound "loops" with a usually short segment which will
repeat until the hardware reset.

As far as showing the problem, I have found that using xmms to play various ogg
files rip'ed from cds will demonstrate the problem.

Note that there is nothing in any of the logs to show the problem.

I suspect that the kernel is going into some kind of loop with everything disabled.

Note that I do not see the problem if I reboot and run the non-smp kernel.

Version-Release number of selected component (if applicable):
8.0, kernel-smp-2.4.18-14

How Reproducible:
yes but not predictable.  Somtimes it will run for hours before a freeze and
others only a couple of minutes.  Processor activity can be light or heavy.

Steps to Reproduce:
1. boot smp kernel
2. run some applications which will play lots of music.
3. 

Additional Information:
	
I do not know if just playing cd's will cause the problem but I doubt it.

There is really no way around this other than "don't do that" which just will
not work for me.

Comment 1 Gene Czarcinski 2002-10-04 20:40:25 UTC
The motherboard is a ABit VP6 with no on-board sound card ... the SB Live is a
pci card.  Other info attached.

Comment 2 Gene Czarcinski 2002-10-04 20:41:10 UTC
Created attachment 78724 [details]
/proc/cpuinfo

Comment 3 Gene Czarcinski 2002-10-04 20:41:59 UTC
Created attachment 78725 [details]
/proc/devices

Comment 4 Gene Czarcinski 2002-10-04 20:42:38 UTC
Created attachment 78726 [details]
/proc/driver/emu10k1 info

Comment 5 Gene Czarcinski 2002-10-04 20:43:15 UTC
Created attachment 78727 [details]
/proc/interrupts

Comment 6 Gene Czarcinski 2002-10-04 20:43:59 UTC
Created attachment 78728 [details]
lspci

Comment 7 Gene Czarcinski 2002-10-04 20:44:36 UTC
Created attachment 78729 [details]
lspci -vv

Comment 8 Gene Czarcinski 2002-10-04 20:45:50 UTC
Created attachment 78730 [details]
lsmod

Comment 9 Gene Czarcinski 2002-10-04 20:46:25 UTC
Created attachment 78732 [details]
/proc/meminfo

Comment 10 Gene Czarcinski 2002-10-05 23:20:44 UTC
Oh, not good, not good.  I put a fresh install of 7,3 on this system and applied
all errata including kernel-smp-2.4.18-10.

The problem occurs!  This looks bad!

Comment 11 Gene Czarcinski 2002-10-07 14:30:52 UTC
OK, I am going to leave this open for a while but I believe the problem is the
POS HPT370 controller that I am using.

Because the new DVD-RW drive I installed would not blank cdrw's, I installed a
fifth ide device on the builtin HPT370 controller.  While I was getting some
error message in /var/log/messages about the device, it did seem to operate
properly.  However, when I ran nmi_watchdog=1 and got the OOPS, it seemed to
point to this controllers driver.  Therefore, I have pulled the device and will
continue testing to see the freeze occurs again.  If it does not appear in a
week, I will close this report.

Comment 12 Gene Czarcinski 2002-10-10 09:43:05 UTC
no re-occurance ... closing