it appears that the same NFS-related soft-lockup is still in 2.3.6pre1.
I can trigger it now using the command
find . -type f -print|xargs -n 1000 md5sum
inside an NFS-mounted directory. Precondition is the mount option
rsize=4096,wsize=4096, I cannot trigger the bug using rsize=1024,wsize=1024.
I managed to capture some EIP values:
c01d600e stext_lock+0x22D2
c01d6007 stext_lock+0x22CB
c010ae07 __global_cli+0xCB
c01d2391 __delay+0x21
c01d2390 __delay+0x20
c01d2380 __delay+0x10
c01d2370 __delay
c01d23a6 __udelay+0x12
After some time hitting Sysreq-P I got the message:
wait_on_irq, CPU 1
irq: 0 [0 0]
bh: 1 [1 0]
<[c010adfd]> <[c011615c]> <[c01538e5]> <[c0128f21]> <[c0109168]>
This translates to:
c010adfd __global_cli+0xC1
c011615c do_syslog+0x2DC
c01538e5 kmsg_read+0x11
c0128f21 sys_read+0xC5
c0109168 system_call+0x34
The system is based on Dual-PentiumII, 233MHz,
Gigabyte GA-686LX2 motherboard. The kernel is compiled with
egcs 1.1.2.
The remote NFS-server is Linux-2.2.7, Universal NFS Server 2.2beta29.
Find the hardware and kernel configuration used at
http://www.snafu.de/~zahn/autoconf.html
Complete boot messages are at http://www.snafu.de/~zahn/bootmsg.html
Steffen
-- home email: user@domain where domain=berlin.snafu.de, user=zahn Use of my address for unsolicited commercial advertising is forbidden. 2^3021377 - 1 | "Where do you want to crash today?"- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/