tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

not-SCHED_M2 livelock

On Sat, Jan 05, 2008 at 09:56:04AM +1100, Daniel Carosone wrote:
> Very possibly - gmake is certainly behaving anti-socially, whether the
> root cause is purely internal or made worse with some outside
> assistance from (say) poll timeouts not running to term or something.

The plot thickens:

 a) the gmake behaviour while building mencoder is readily reproducible 
 b) I have observed the X server getting into (and then, just as
    mysteriously, out of) a similar state, where it's spinning taking all
    the cpu it can

> My concern at this stage was to highlight the scheduler issue that
> allows 4 such miscreants to effectively kill the system -- even "pkill
> -STOP gmake" regularly took 5-10 mins to get enough time in to take
> effect. 

 c) the almost-complete starvation of every other process that results
    happens regardless of SCHED_M2 or SCHED_4BSD.


Attachment: pgpY9H8i8gWHc.pgp
Description: PGP signature

Home | Main Index | Thread Index | Old Index