Messages in this thread | | | Date | Tue, 22 Dec 1998 21:45:53 +0100 (CET) | From | Rik van Riel <> | Subject | Re: NEEDED FEATURES |
| |
On Tue, 22 Dec 1998, PIGS RUS wrote:
> We are the saviours of Linux.
[snip]
> o Crashes. People have come to expect them.
Hey, I can do this...
Just yesterday evening, when I was debugging smp_send_message() in arch/i386/kernel/smp.c I decided to uncomment the debugging statement.
Now once in a while the debugging statement should be printed and handed to syslogd -- but low and behold, passing it to syslogd can trigger the next smp_send_message(), which will fail because the CPU is already busy sending a message...
This, in turn, will trigger an error message that keeps the APIC so busy that even <sysrq> can't get through. Out of the 15 crashes I made that way, only one allowed <sysrq> and even then it only was a partial one :))
To Linus, DaveM and Ingo: I didn't complete debugging the SMP message passing stuff, but I guess you've already gathered that by now :)
cheers,
Rik -- the flu hits, the flu hits, the flu hits -- MORE +-------------------------------------------------------------------+ | Linux memory management tour guide. H.H.vanRiel@phys.uu.nl | | Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ | +-------------------------------------------------------------------+
- 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/
| |