Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-Q5

[Mark_H_Johnson]

>regarding this particular latency, could you try the attached patch
>ontop of -Q5? It turns the ->poll() loop into separate, individually
>preemptable iterations instead of one batch of processing. In theory
>this should result in latency being lower regardless of the
>netdev_max_backlog value.

First time - stopped during init script - when trying to start a network
service (automount).
Second time - stopped during init script - did not get the "OK" for
bringing up interface eth0.
Alt-SysRq-P still does not show any data [not sure why].
Alt-SysRq-T captured data - for example, shows dhclient in sys_select ->
__pollwait -> do_select -> process_timeout -> add_wait_queue ->
schedule_timeout -> __mod_timer. Very odd, almost every other task (except
initlog) is stuck in one of:
 - generic_handle_IRQ_event
 - sub_preempt_count
 - do_irqd
 - do_hardirq
and all the tasks I can see have "S" status.
Third time - ditto - back to stopping at automount start up.
Alt-SysRq-T captured data, again everything I could look at in "S" mode.
In all three cases, another system attempting to "ping" the system under
test failed to get any responses.

In all cases, Ctrl-Alt-Del was good enough to get a clean reboot.

This looks like a bad patch; will go back to the last good kernel for
further testing.

--Mark H Johnson
  <mailto:Mark_H_Johnson@xxxxxxxxxxxx>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/