Re: 2.4.34-pre1 USB mass-storage burped...

[Willy Tarreau <w@1wt.eu>]

Hi Grant,

On Sat, Aug 19, 2006 at 06:41:50PM +1000, Grant Coady wrote:
> Hi there,
> 
> Running NFS testing (continuous kernel build + untar, diff against previous) 
> over two NFS mounts.  Also running continuous kernel build to USB HDD:
> 
> No datestamp on rebuild start, sorry -- about 6 or 7 builds from end of 
> log roughly matches logged events' datestamp, only 1 of 25 builds has 
> the error:
> 
> /bin/sh: line 1: 29947 Bus error               /mnt/hd/linux-2.4.33/scripts/mkdep -D__KERNEL__ -I/mnt/hd/linux-2.4.33/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe
> -mpreferred-stack-boundary=2 -march=athlon -nostdinc -iwithprefix include -- amdtp.c amdtp.h cmp.c cmp.h csr.c csr.h dma.c dma.h dv1394-private.h dv1394.c dv1394.h eth1394.c eth1394.h highlevel.c highlevel.h hosts.c hosts.h ieee1394-ioctl.h
> ieee1394.h ieee1394_core.c ieee1394_core.h ieee1394_hotplug.h ieee1394_transactions.c ieee1394_transactions.h ieee1394_types.h iso.c iso.h nodemgr.c nodemgr.h ohci1394.c ohci1394.h pcilynx.c pcilynx.h raw1394-private.h raw1394.c raw1394.h sbp2.c
> sbp2.h video1394.c video1394.h >.depend
> make[4]: *** [fastdep] Error 135
> make[3]: *** [_sfdep_ieee1394] Error 2
> make[2]: *** [fastdep] Error 2
> make[1]: *** [_sfdep_drivers] Error 2
> make: *** [dep-files] Error 2
> 
> /var/log/messages:
> Aug 19 14:04:26 sempro kernel: Initializing USB Mass Storage driver...
> Aug 19 14:04:26 sempro kernel: usb.c: registered new driver usb-storage
> Aug 19 14:04:26 sempro kernel: scsi2 : SCSI emulation for USB Mass Storage devices
> Aug 19 14:04:26 sempro kernel:  sdc: sdc1 sdc2 sdc3
> Aug 19 14:04:26 sempro kernel: USB Mass Storage support registered.
> Aug 19 14:09:34 sempro sshd[31910]: Accepted publickey for grant from 192.168.1.31 port 1940 ssh2
> Aug 19 14:24:23 sempro -- MARK --
> Aug 19 14:44:23 sempro -- MARK --
> Aug 19 15:04:23 sempro -- MARK --
> Aug 19 15:24:23 sempro -- MARK --
> Aug 19 15:44:23 sempro -- MARK --
> Aug 19 16:04:23 sempro -- MARK --
> Aug 19 16:24:23 sempro -- MARK --
> Aug 19 16:44:23 sempro -- MARK --
> Aug 19 17:04:23 sempro -- MARK --
> Aug 19 17:08:54 sempro kernel: usb.c: USB disconnect on device 00:10.4-5 address 4
> Aug 19 17:08:54 sempro kernel: hub.c: new USB device 00:10.4-5, assigned address 5
> Aug 19 17:08:54 sempro kernel: Product: USB TO IDE
> Aug 19 17:24:23 sempro -- MARK --
> 
> /var/log/syslog:
> Aug 19 17:08:54 sempro kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 70000
> Aug 19 17:08:54 sempro kernel:  I/O error: dev 08:22, sector 238536
> Aug 19 17:08:54 sempro kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 70000
> Aug 19 17:08:54 sempro kernel:  I/O error: dev 08:22, sector 345840
> Aug 19 17:08:54 sempro kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 70000
> Aug 19 17:08:54 sempro kernel:  I/O error: dev 08:22, sector 345848
> Aug 19 17:08:54 sempro kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 70000
> Aug 19 17:08:54 sempro kernel:  I/O error: dev 08:22, sector 345840
> Aug 19 17:08:54 sempro kernel: SCSI disk error : host 2 channel 0 id 0 lun 0 return code = 70000
> Aug 19 17:08:54 sempro kernel:  I/O error: dev 08:22, sector 345840
> 
> I'll leave the test running overnight, unless I smell smoke ;)
> 
> Test boxen dmesg + comment stripped config:
>   <http://bugsplatter.mine.nu/test/boxen/peetoo/2.4.xx/> NFS server
>   <http://bugsplatter.mine.nu/test/boxen/sempro/2.4.xx/> NFS client and build host
> 
> Both running 2.4.34-pre1 for this test.  
> 
> Recent kernel rebuild testing over NFS produced no errors:
> grant@sempro:~$ grep Error /home/public/kbuildtest-2.4.log-2.4.33-final
> grant@sempro:~$ grep Error /home/public/kbuildtest-2.4.log-2.6.17.8-tcp
> grant@sempro:~$ grep Error /home/public/kbuildtest-2.4.log-2.6.17.8-udp
> grant@sempro:~$ grep Error /home/public/kbuildtest-2.4.log  <<== current test

Have you tried building over USB HDD for another kernel (at least 2.4.33) ?
If not, could you give it a try please ? I would like to know if this problem
could have been introduced by the locking changes in 2.4.34-pre1.

Thanks in advance,
Willy

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/