If Im interpreting this correctly the clockrate (and the counter) is fixed even if the freq-scaling driver adjusts the freq .
Mario
Mario
----- Reply message -----
From: "Marc Zyngier" <marc.zyngier@xxxxxxx>
To: "Mario Smarduch" <Mario.Smarduch@xxxxxxxxxx>
Cc: "kvmarm@xxxxxxxxxxxxxxxxxxxxx" <kvmarm@xxxxxxxxxxxxxxxxxxxxx>
Subject: ARM ARCH TImers and High Res Timers
Date: Wed, Jan 9, 2013 12:41
From: "Marc Zyngier" <marc.zyngier@xxxxxxx>
To: "Mario Smarduch" <Mario.Smarduch@xxxxxxxxxx>
Cc: "kvmarm@xxxxxxxxxxxxxxxxxxxxx" <kvmarm@xxxxxxxxxxxxxxxxxxxxx>
Subject: ARM ARCH TImers and High Res Timers
Date: Wed, Jan 9, 2013 12:41
On 09/01/13 11:30, Mario Smarduch wrote:
> Follow up question, if a frequency driver is employed for PM does
> arch timer tick at same rate for the clock source and timer?
I'm not sure I get your question.
The clock source and the timer are basically the same thing (the first
is a counter, and the second is a comparator associated with the
counter). The arch timer must have a *fixed* frequency though.
If you're using a separate timer for PM purpose, the kernel will take
care of it. This is little to do with KVM though.
M.
--
Jazz is not dead. It just smells funny...
> Follow up question, if a frequency driver is employed for PM does
> arch timer tick at same rate for the clock source and timer?
I'm not sure I get your question.
The clock source and the timer are basically the same thing (the first
is a counter, and the second is a comparator associated with the
counter). The arch timer must have a *fixed* frequency though.
If you're using a separate timer for PM purpose, the kernel will take
care of it. This is little to do with KVM though.
M.
--
Jazz is not dead. It just smells funny...
_______________________________________________ kvmarm mailing list kvmarm@xxxxxxxxxxxxxxxxxxxxx https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm
