On Tue, 29 Feb 2000, William Montgomery wrote:
> > William, can you explain please this effect more precisely ?
> > What does mean "the other task will have to wait" ?
> > That the other task will not be scheduled until the inode freeing is finished,
> > while the other runs fine ?
>
> Yes. Andrea has suggested that the reschedule should only be allowed once
> during the free_inode loop.
Understood,
but is this waiting for the 2nd task a real problem in practice ?
Are you able to estimate this time ?
Is this time in the 1msec range or much bigger ?
Andrea: If yes, is there any solution in sight , which allows to cut down these
times ?
>
> > How does the old 2.2.10-lowlatencyN6B kernel react to these 2 SCHED_FIFO
> > tasks ?
> It reacts in the desired manner, however, the 2.2.10-lowlatencyN6B patch
> has a bug that Andrea pointed out. If an inode is freed during a
> schedule() then the patch does not properly handle that.
>
> Wm
Benno.
-
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/
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 21:00:22 EST