From: Davidlohr Bueso
> Sent: 20 November 2020 21:38
>
> On Fri, 20 Nov 2020, David Laight wrote:
> >I got massive performance improvements from changing a driver
> >we have to use mutex instead of the old semaphores (the driver
> >was written a long time ago).
> >
> >While these weren't 'rw' the
On 11/20/20 9:44 AM, Peter Zijlstra wrote:
On Tue, Nov 17, 2020 at 09:35:56PM -0800, Davidlohr Bueso wrote:
On Tue, 17 Nov 2020, Waiman Long wrote:
The column "CS Load" represents the number of pause instructions issued
in the locking critical section. A CS load of 1 is extremely short and
is
On Fri, 20 Nov 2020, David Laight wrote:
I got massive performance improvements from changing a driver
we have to use mutex instead of the old semaphores (the driver
was written a long time ago).
While these weren't 'rw' the same issue will apply.
The problem was that the semaphore/mutex was
From: Waiman Long
> Sent: 20 November 2020 17:04
>
> On 11/20/20 8:11 AM, David Laight wrote:
> > From: Waiman Long
> >> Sent: 19 November 2020 18:40
> > ...
> >> My own testing also show not too much performance difference when
> >> removing reader spinning except in the lightly loaded cases.
>
On 11/20/20 8:11 AM, David Laight wrote:
From: Waiman Long
Sent: 19 November 2020 18:40
...
My own testing also show not too much performance difference when
removing reader spinning except in the lightly loaded cases.
I'm confused.
I got massive performance improvements from changing a
On Tue, Nov 17, 2020 at 09:35:56PM -0800, Davidlohr Bueso wrote:
> On Tue, 17 Nov 2020, Waiman Long wrote:
>
> > The column "CS Load" represents the number of pause instructions issued
> > in the locking critical section. A CS load of 1 is extremely short and
> > is not likey in real situations.
On Tue, Nov 17, 2020 at 10:04:29PM -0500, Waiman Long wrote:
> -static inline bool osq_is_empty(struct rw_semaphore *sem)
> -{
> - return !osq_is_locked(>osq);
> -static inline bool osq_is_empty(sem)
> -{
> - return false;
> -}
Oh, it doesn't live...
From: Waiman Long
> Sent: 19 November 2020 18:40
...
> My own testing also show not too much performance difference when
> removing reader spinning except in the lightly loaded cases.
I'm confused.
I got massive performance improvements from changing a driver
we have to use mutex instead of the
On 11/18/20 12:35 AM, Davidlohr Bueso wrote:
On Tue, 17 Nov 2020, Waiman Long wrote:
The column "CS Load" represents the number of pause instructions issued
in the locking critical section. A CS load of 1 is extremely short and
is not likey in real situations. A load of 20 (moderate) and 100
On Tue, 17 Nov 2020, Waiman Long wrote:
The column "CS Load" represents the number of pause instructions issued
in the locking critical section. A CS load of 1 is extremely short and
is not likey in real situations. A load of 20 (moderate) and 100 (long)
are more realistic.
It can be seen that
Reader optimistic spinning is helpful when the reader critical section
is short and there aren't that many readers around. It also improves
the chance that a reader can get the lock as writer optimistic spinning
disproportionally favors writers much more than readers.
Since commit d3681e269fff
11 matches
Mail list logo