Re: [PATCH v25 0/4] MR3504: ntdll: An implementation of SRWLOCK that closer matches Windows'.

[Zebediah Figura (＠zfigura)]

Zebediah Figura (@zfigura) commented about dlls/ntdll/sync.c:
> -            {
> -                /* Not locked exclusive, and no exclusive waiters.
> -                 * We can try to grab it. */
> -                ++new.s.owners;
> -                wait = FALSE;
> -            }
> -            else
> -            {
> -                wait = TRUE;
> -            }
> -        } while (InterlockedCompareExchange( u.l, new.l, old.l ) != old.l);
> +    if (tag & SRWLOCK_TAG_LIST_LOCKED) {
> +        /* We only ever blocking wait on &lock->Ptr for the LIST_LOCKED bit. */
> +        RtlWaitOnAddress(&lock->Ptr, &expected.Ptr, sizeof(expected.Ptr), NULL);
> +        return FALSE;
> +    }
> Does the futex wait/wake really help here? I'm curious what the effects are of removing it on your various benchmarks.

(Side note, this patch has some inconsistent style in a lot of places wrt. braces placement or spacing; it'd probably be nice to make that consistent.)

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/3504#note_46112