Bug 128606

Summary: DMA timeout with PDC202XX reset
Product: [Fedora] Fedora Reporter: Joerg Fechner <java-joerch>
Component: kernelAssignee: Dave Jones <davej>
Status: CLOSED NEXTRELEASE QA Contact:
Severity: high Docs Contact:
Priority: medium    
Version: 2CC: pfrields, wtogami
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-04-16 05:31:35 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/ide/pdc202xx of the WORKING configuration none

Description Joerg Fechner 2004-07-26 23:48:27 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040510

Description of problem:
Please excusemy bad english, but I try my very best!

I have a HD connected with a Promise Ultra100 to my System. When I
copy large files via samba or from another HD to this 250GB disk the
System-Log shows the following:

Jul 25 01:46:03 familix kernel: hdf: dma_timer_expiry: dma status == 0x40
Jul 25 01:46:03 familix kernel: hdf: DMA timeout retry
Jul 25 01:46:03 familix kernel: PDC202XX: Primary channel reset.
Jul 25 01:46:03 familix kernel: PDC202XX: Secondary channel reset.
Jul 25 01:46:03 familix kernel: hdf: timeout waiting for DMA
Jul 25 01:46:23 familix kernel: hdf: dma_timer_expiry: dma status == 0x41
Jul 25 02:58:12 familix kernel: hdf: DMA timeout error
Jul 25 02:58:12 familix kernel: hdf: dma timeout error: status=0x58 {
DriveReady SeekComplete DataRequest }
Jul 25 02:58:12 familix kernel: 
Jul 25 02:58:12 familix kernel: hdf: status timeout: status=0xd0 { Busy }
Jul 25 02:58:12 familix kernel: 
Jul 25 02:58:12 familix kernel: PDC202XX: Primary channel reset.
Jul 25 02:58:12 familix kernel: PDC202XX: Secondary channel reset.
Jul 25 02:58:12 familix kernel: hdf: drive not ready for command
Jul 25 02:58:12 familix kernel: ide2: reset: master: error (0x00?)

Sometimes, the hd turns into WRITE-PROTECTED. The only thing is to
reboot the computer.

Version-Release number of selected component (if applicable):
any of 2.6.x released by Fedora

How reproducible:
Always

Steps to Reproduce:
1. copy a large file (>700MB)
2.
3.
    

Actual Results:  The problem as described

Expected Results:  Copying the file very fast without an error.

Additional info:

On Fedora-Core 1 with kernel 2.4.whatever, the system worked properly.
After updating the system, I have this above mentioned behavior.

Comment 1 Bernhard Weisshuhn 2004-10-17 13:42:25 UTC
Same here.

After adding a new disk (200GB Seagate Barracuda ST3200822A) replacing
 and older IBM DTLA, dms timeouts occur under heavy load.

With USE_DMA and -X69 in /etc/sysconfig/harddisks, the controller
seems to go into readonly mode as soon as it happens. -X68 doesn't
seem to be as bad, it made these log entries survive:

kernel: hdf: DMA timeout error
kernel: hdf: dma timeout error: status=0xd0 { Busy }
kernel:
kernel: hde: DMA disabled
kernel: hdf: DMA disabled
kernel: PDC202XX: Primary channel reset.
kernel: PDC202XX: Secondary channel reset.
kernel: ide2: reset: master: error (0x00?)
kernel: hdf: dma_timer_expiry: dma status == 0x41
kernel: hdf: DMA timeout error
kernel: hdf: dma timeout error: status=0xd0 { Busy }
kernel:
kernel: hdf: DMA disabled
kernel: PDC202XX: Primary channel reset.
kernel: PDC202XX: Secondary channel reset.
kernel: ide2: reset: master: error (0x00?)
kernel: end_request: I/O error, dev hdf, sector 272714424
kernel: Buffer I/O error on device dm-1, logical block 34089303

It didn't happen again after I removed USE_DMA and EXTRA_PARAMS=-X69
from /etc/sysconfig/harddisks although hdparm still reports the disk
to be using dma and udma5.

It might me unrelated to the drive though, I was also upgrading to
kernel 2.6.8-1.521 and using xfs with evms at the same time.

Comment 2 Bernhard Weisshuhn 2004-10-17 13:52:34 UTC
Created attachment 105350 [details]
/proc/ide/pdc202xx of the WORKING configuration

This is the output of `cat /proc/ide/pdc202xx` in the configuration where the
lockup does NOT seem to happen.

Comment 3 Dave Jones 2005-04-16 05:31:35 UTC
Fedora Core 2 has now reached end of life, and no further updates will be
provided by Red Hat.  The Fedora legacy project will be producing further kernel
updates for security problems only.

If this bug has not been fixed in the latest Fedora Core 2 update kernel, please
try to reproduce it under Fedora Core 3, and reopen if necessary, changing the
product version accordingly.

Thank you.