Re: [PATCH 2/2] locking/qrwlock: Reduce cacheline contention for rwlocks used in interrupt context

[Waiman Long]

On 5/11/22 03:20, Arnd Bergmann wrote:
> On Tue, May 10, 2022 at 9:21 PM Waiman Long <longman@xxxxxxxxxx> wrote:
> > Even though qrwlock is supposed to be a fair lock, it does allow readers
> > from interrupt context to spin on the lock until it can acquire it making
> > it not as fair. This exception was added due to the requirement to allow
> > recursive read lock in interrupt context. This can also be achieved by
> > just ignoring the writer waiting bit without spinning on the lock.
> >
> > By making this change, we make qrwlock a bit more fair and eliminating
> > the problem of cacheline bouncing for rwlocks that are used heavily in
> > interrupt context, like the networking stack. This should also reduce
> > the chance of lock starvation for those interrupt context rwlocks.
> >
> > Signed-off-by: Waiman Long <longman@xxxxxxxxxx>
> > ---
> >   include/asm-generic/qrwlock.h |  6 +++---
> >   kernel/locking/qrwlock.c      | 17 ++++++-----------
> >   2 files changed, 9 insertions(+), 14 deletions(-)
> I have no opinion on the change itself, but I made sure this does
> not conflict with the generic ticket spinlock changes that I merged
> in the asm-generic tree, since those also touch the comments in
> qrwlock.h [1]
>
> I assume you are merging both patches through the tip tree, so

Yes, it is based on the latest tip tree.

Cheers,
Longman