On Tue, May 19, 2020 at 10:19:08PM +0200, Sebastian Andrzej Siewior wrote:
> diff --git a/mm/swap.c b/mm/swap.c
> index bf9a79fed62d7..03c97d15fcd69 100644
> --- a/mm/swap.c
> +++ b/mm/swap.c
> @@ -44,8 +44,14 @@
> /* How many pages do we try to swap or page in/out together? */
> int page_cluster
On Tue, May 19, 2020 at 04:58:37PM -0700, Andrew Morton wrote:
> On Tue, 19 May 2020 22:19:08 +0200 Sebastian Andrzej Siewior
> wrote:
> > + local_lock(swapvec_lock);
>
> It's quite peculiar that these operations appear to be pass-by-value.
> All other locking operatio
On Tue, 19 May 2020 22:19:08 +0200 Sebastian Andrzej Siewior
wrote:
> From: Ingo Molnar
>
> The various struct pagevec per CPU variables are protected by disabling
> either preemption or interrupts across the critical sections. Inside
> these sections spinlocks have to be acquired.
>
> These
From: Ingo Molnar
The various struct pagevec per CPU variables are protected by disabling
either preemption or interrupts across the critical sections. Inside
these sections spinlocks have to be acquired.
These spinlocks are regular spinlock_t types which are converted to
"sleeping" spinlocks on
4 matches
Mail list logo