Re: [patch] recover losed timer interrupt using the TSC [Re: [patch] kstat change to see how much Li
Andi Kleen (ak@muc.de)
Sun, 14 Mar 1999 12:20:03 +0100
On Sun, Mar 14, 1999 at 11:51:55AM +0100, Ingo Molnar wrote:
>
> On Sun, 14 Mar 1999, Andi Kleen wrote:
>
> > Perhaps it is a good idea to only print the warning a few times, to
> > avoid flooding the logs.
>
> yep. OTOH, note that the printout itself is rate-limited. (it detects
> 1.5*10msec delays, which period restarts after the message is started).
> For the already known case (SCSI resets) we already have a few messages
> per 'event' anyway. Also, it should not be possible to trigger such huge
> delays from user-space, if it's possible then we are in trouble anyway.
I meant mainly to keep the machine still usable in case of bugs, not to
prevent any DoS. It is only intended as a warning, not as a panic.
Also assume the case: block driver disables irqs for a long
time, message is printed, syslog is called, syslog does fsync, block driver
disables irqs again etc.
-Andi
--
This is like TV. I don't like TV.
-
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/