Messages in this thread | | | Date | Mon, 25 Oct 1999 11:23:22 -0700 (PDT) | From | Robert Dinse <> | Subject | 2.2.13 Sparc SMP-32 |
| |
The patch that Dave Miller posted here a bit ago for Sparc SMP 32 bit hardware didn't help on my systems. It changed the nature of the crash so instead of getting spin_lock(... blah blah blah, I'd get spin_lock_irq(... blah blah blah, but the actual frequency of crashes seemed to get worse instead of better.
Tonight I installed 2.2.13 on my web server and news server, both SS-10's with quad RTK-625 CPU's. The web server is still running, the news server crashed after about an hour with:
spin_lock_irqsave(f015c4bc) CPU#0 stuck at f0113fe8, owner PC(f00fcfc4):CPU(0)
This is the first time I've ever seen the stuck CPU be the same as the owner of the lock.
This may be hardware related, mainboard on the news machine is dying. I am getting ethernet memory errors and all sorts of SCSI flakiness (and SCSI controller and ethernet are built in on these). I have a new mainboard coming so will replace it before posting any further reports of problems with this particular machine.
The web server continues to run good now about six hours on 2.2.13 so hopefully new hardware will make the other machine happier.
- 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/
| |