[9fans] common number of interrups/sec

[9fans] common number of interrups/sec

[Rudolf Sykora]

Hello,

my system is doing practically nothing. All 'stats' graphs are low
with the exception of 'i'. I'm having like 1600 interrups per second
(so I see a number instead of just the graph). Is this normal? How can
I tell what those interrupts are?

Thanks
Ruda

Re: [9fans] common number of interrups/sec

[erik quanstrom]

> my system is doing practically nothing. All 'stats' graphs are low
> with the exception of 'i'. I'm having like 1600 interrups per second
> (so I see a number instead of just the graph). Is this normal? How can
> I tell what those interrupts are?

if you are running 9atom, /dev/irqalloc 3rd column is the number
of interrupts for that interrupt entry.  note that if two irq handlers
are chained off the same vector, an interrupt for either will increment
both of their total counts.  this is just a hardware limitation.

- erik

Re: [9fans] common number of interrups/sec

[Rudolf Sykora]

On 8 March 2010 17:21, erik quanstrom <quanstro@labs.coraid.com> wrote:
>> my system is doing practically nothing. All 'stats' graphs are low
>> with the exception of 'i'. I'm having like 1600 interrups per second
>> (so I see a number instead of just the graph). Is this normal? How can
>> I tell what those interrupts are?
>
> if you are running 9atom, /dev/irqalloc 3rd column is the number
> of interrupts for that interrupt entry.  note that if two irq handlers
> are chained off the same vector, an interrupt for either will increment
> both of their total counts.  this is just a hardware limitation.
>
> - erik

Ok. I tried to cat /dev/irqalloc and by far (2 orders of magnitude)
the highest number is

32  0  45425578 clock

What is the meaning?
Is this Ok?

Thanks
Ruda

Re: [9fans] common number of interrups/sec

[erik quanstrom]

> Ok. I tried to cat /dev/irqalloc and by far (2 orders of magnitude)
> the highest number is
>
> 32  0  45425578 clock
>
> What is the meaning?
> Is this Ok?

yes.  it should be fine.  9atom cranks up HZ to 1000.
this allows accurate sleeping delays down to 1ms.
this shouldn't be a problem, even for an older system
such as yours.  it just looks funny on stats.

- erik

Re: [9fans] common number of interrups/sec

[Rudolf Sykora]

On 8 March 2010 17:49, erik quanstrom <quanstro@labs.coraid.com> wrote:
>> Ok. I tried to cat /dev/irqalloc and by far (2 orders of magnitude)
>> the highest number is
>>
>> 32  0  45425578 clock
>>
>> What is the meaning?
>> Is this Ok?
>
> yes.  it should be fine.  9atom cranks up HZ to 1000.
> this allows accurate sleeping delays down to 1ms.
> this shouldn't be a problem, even for an older system
> such as yours.  it just looks funny on stats.
>
> - erik
>

Fine to know it is as should be.
Thanks!
Ruda

Re: [9fans] common number of interrups/sec

[ron minnich]

On Mon, Mar 8, 2010 at 8:04 AM, Rudolf Sykora <rudolf.sykora@gmail.com> wrote:
> Hello,
>
> my system is doing practically nothing. All 'stats' graphs are low
> with the exception of 'i'. I'm having like 1600 interrups per second
> (so I see a number instead of just the graph). Is this normal? How can
> I tell what those interrupts are?

This sounds like what I had with venti.

What does venti print when it starts up?

ron