Re: [PATCH v4 0/3] Optimize CAS [BZ #28537]

["H.J. Lu" <hjl.tools@gmail.com>]

On Wed, Nov 10, 2021 at 7:52 AM Florian Weimer <fweimer@redhat.com> wrote:
>
> * Paul A. Clarke:
>
> > On Tue, Nov 09, 2021 at 04:16:11PM -0800, H.J. Lu via Libc-alpha wrote:
> >> CAS instruction is expensive.  From the x86 CPU's point of view, getting
> >> a cache line for writing is more expensive than reading.  See Appendix
> >> A.2 Spinlock in:
> >>
> >> https://www.intel.com/content/dam/www/public/us/en/documents/white-papers/xeon-lock-scaling-analysis-paper.pdf
> >>
> >> The full compare and swap will grab the cache line exclusive and cause
> >> excessive cache line bouncing.
> >>
> >> Optimize CAS in low level locks and pthread_mutex_lock.c:
> >>
> >> 1. Do an atomic load and skip CAS if compare may fail to reduce cache
> >> line bouncing on contended locks.
> >> 2. Replace atomic_compare_and_exchange_bool_acq with
> >> atomic_compare_and_exchange_val_acq to avoid the extra load.
> >> 3. Drop __glibc_unlikely in __lll_trylock and lll_cond_trylock since we
> >> don't know if it's actually rare; in the contended case it is clearly not
> >> rare.
> >
> > I see build errors:
> >
> > In file included from pthread_mutex_cond_lock.c:23:
> > ../nptl/pthread_mutex_lock.c: In function ‘__pthread_mutex_cond_lock_full’:
> > ../nptl/pthread_mutex_lock.c:442:6: error: a label can only be part of a statement and a declaration is not a statement
> >       int private = (robust
> >       ^~~
>
> The patch has:
>
> + locked_mutex:
>             /* The mutex is locked.  The kernel will now take care of
>                everything.  */
>             int private = (robust
>
> This is only supported in recent C versions, I think the workaround is
> to add an empty statement with a semicolon, like this:
>
> + locked_mutex:;
>             /* The mutex is locked.  The kernel will now take care of
>                everything.  */
>             int private = (robust
>
> Thanks,
> Florian
>

Can you try users/hjl/x86/atomic-nptl branch:

https://gitlab.com/x86-glibc/glibc/-/commits/users/hjl/x86/atomic-nptl

I fixed a couple issues and added more CAS optimizations.

-- 
H.J.