Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-27 Thread Prarit Bhargava
Jeremy Fitzhardinge wrote: Prarit Bhargava wrote: Jeremy Fitzhardinge wrote: Prarit Bhargava wrote: I'd like to see this patch implement/fix touch_cpu_softlockup_watchdog and touch_softlockup_watchdog to mimic touch_nmi_watchdog's behaviour. Wh

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-27 Thread Jeremy Fitzhardinge
Prarit Bhargava wrote: > Jeremy Fitzhardinge wrote: > >> Prarit Bhargava wrote: >> >> >>> I'd like to see this patch implement/fix touch_cpu_softlockup_watchdog >>> and touch_softlockup_watchdog to mimic touch_nmi_watchdog's behaviour. >>> >>> >> Why? Is that more correct?

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-27 Thread Prarit Bhargava
Jeremy Fitzhardinge wrote: Prarit Bhargava wrote: I'd like to see this patch implement/fix touch_cpu_softlockup_watchdog and touch_softlockup_watchdog to mimic touch_nmi_watchdog's behaviour. Why? Is that more correct? It seems to me that you're interested in whether a specific CPU

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-27 Thread Jeremy Fitzhardinge
Prarit Bhargava wrote: > I'd like to see this patch implement/fix touch_cpu_softlockup_watchdog > and touch_softlockup_watchdog to mimic touch_nmi_watchdog's behaviour. Why? Is that more correct? It seems to me that you're interested in whether a specific CPU has gone and locked up. If touching

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-27 Thread Prarit Bhargava
Jeremy Fitzhardinge wrote: --- kernel/softlockup.c | 28 +++- 1 file changed, 19 insertions(+), 9 deletions(-) === --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -17,8 +17,8 @@ static DEFINE_

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-26 Thread Eric Dumazet
On Tue, 27 Mar 2007 00:12:53 -0700 Jeremy Fitzhardinge <[EMAIL PROTECTED]> wrote: > Eric Dumazet wrote: > > Jeremy Fitzhardinge a écrit : > > > >> +static DEFINE_PER_CPU(unsigned long long, touch_timestamp); > > > > ... > > > >> void touch_softlockup_watchdog(void) > >> { > >> -__raw_get_cpu

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-26 Thread Jeremy Fitzhardinge
Eric Dumazet wrote: > Jeremy Fitzhardinge a écrit : > >> +static DEFINE_PER_CPU(unsigned long long, touch_timestamp); > > ... > >> void touch_softlockup_watchdog(void) >> { >> -__raw_get_cpu_var(touch_timestamp) = jiffies; >> +__raw_get_cpu_var(touch_timestamp) = sched_clock(); >> } > >

Re: [patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-26 Thread Eric Dumazet
Jeremy Fitzhardinge a écrit : +static DEFINE_PER_CPU(unsigned long long, touch_timestamp); ... void touch_softlockup_watchdog(void) { - __raw_get_cpu_var(touch_timestamp) = jiffies; + __raw_get_cpu_var(touch_timestamp) = sched_clock(); } Not very clear if this is safe on 32b

[patch 1/2] Ignore stolen time in the softlockup watchdog

2007-03-26 Thread Jeremy Fitzhardinge
The softlockup watchdog is currently a nuisance in a virtual machine, since the whole system could have the CPU stolen from it for a long period of time. While it would be unlikely for a guest domain to be denied timer interrupts for over 10s, it could happen and any softlockup message would be co