Arjan van de Ven <[EMAIL PROTECTED]> wrote:
> On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> > Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> > (across different architectures) ? Isn't rwsem used very widely ?
>
> oh also rwsems aren't used all that much
On Wed, Mar 09, 2005 at 12:21:01PM +0100, Andi Kleen wrote:
> Arjan van de Ven <[EMAIL PROTECTED]> writes:
>
> > On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> >> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> >> (across different architectures) ? Isn't
Andrew Morton <[EMAIL PROTECTED]> wrote:
> spin_lock_irq() is OK for down_*(), since down() can call schedule() anyway.
>
> spin_lock_irqsave() should be used in up_*() and I guess down_*_trylock(),
> although the latter shouldn't need to go into the slowpath anyway.
That's what I've done. I'm
Arjan van de Ven <[EMAIL PROTECTED]> wrote:
>
> On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> > Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> > (across different architectures)
>
> on x86 it makes a difference of maybe a few cycles. At most.
> However
Arjan van de Ven <[EMAIL PROTECTED]> writes:
> On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
>> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
>> (across different architectures) ? Isn't rwsem used very widely ?
>
> oh also rwsems aren't used all that much
Arjan van de Ven <[EMAIL PROTECTED]> writes:
> On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
>> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
>> (across different architectures)
>
> on x86 it makes a difference of maybe a few cycles. At most.
> However please
Suparna Bhattacharya <[EMAIL PROTECTED]> writes:
> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> (across different architectures) ? Isn't rwsem used very widely ?
On P4s cli/sti is quite costly, let's says 100+ cycles. That is mostly
because it synchronizes the CPU partly.
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> (across different architectures) ? Isn't rwsem used very widely ?
oh also rwsems aren't used all that much simply because they are quite
more expensive than regular
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> (across different architectures)
on x86 it makes a difference of maybe a few cycles. At most.
However please consider using spin_lock_irqsave(); the _irq() variant,
Suparna Bhattacharya <[EMAIL PROTECTED]> wrote:
>
> Any sense of how costly it is to use spin_lock_irq's vs spin_lock
> (across different architectures) ? Isn't rwsem used very widely ?
It's only on the slow path, and we've already done a bunch of atomic ops
and a schedule()/wakeup() anyway.
-
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
Regards
Suparna
On Wed, Mar 09, 2005 at 10:33:58AM +, David Howells wrote:
> Andrew Morton <[EMAIL PROTECTED]> wrote:
>
> > If we want to take the spinlock
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
Regards
Suparna
On Wed, Mar 09, 2005 at 10:33:58AM +, David Howells wrote:
Andrew Morton [EMAIL PROTECTED] wrote:
If we want to take the spinlock from
Suparna Bhattacharya [EMAIL PROTECTED] wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
It's only on the slow path, and we've already done a bunch of atomic ops
and a schedule()/wakeup() anyway.
-
To
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures)
on x86 it makes a difference of maybe a few cycles. At most.
However please consider using spin_lock_irqsave(); the _irq() variant,
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
oh also rwsems aren't used all that much simply because they are quite
more expensive than regular
Suparna Bhattacharya [EMAIL PROTECTED] writes:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
On P4s cli/sti is quite costly, let's says 100+ cycles. That is mostly
because it synchronizes the CPU partly. The
Arjan van de Ven [EMAIL PROTECTED] writes:
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures)
on x86 it makes a difference of maybe a few cycles. At most.
However please consider
Arjan van de Ven [EMAIL PROTECTED] writes:
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
oh also rwsems aren't used all that much simply
Arjan van de Ven [EMAIL PROTECTED] wrote:
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures)
on x86 it makes a difference of maybe a few cycles. At most.
However please consider
Andrew Morton [EMAIL PROTECTED] wrote:
spin_lock_irq() is OK for down_*(), since down() can call schedule() anyway.
spin_lock_irqsave() should be used in up_*() and I guess down_*_trylock(),
although the latter shouldn't need to go into the slowpath anyway.
That's what I've done. I'm just
On Wed, Mar 09, 2005 at 12:21:01PM +0100, Andi Kleen wrote:
Arjan van de Ven [EMAIL PROTECTED] writes:
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very
Arjan van de Ven [EMAIL PROTECTED] wrote:
On Wed, 2005-03-09 at 16:34 +0530, Suparna Bhattacharya wrote:
Any sense of how costly it is to use spin_lock_irq's vs spin_lock
(across different architectures) ? Isn't rwsem used very widely ?
oh also rwsems aren't used all that much simply
22 matches
Mail list logo
