On 2018-06-06 13:18:50 [+0200], To Peter Zijlstra wrote: > I suggest to remove the custom alpha implementation of dec_and_lock() > and if it is an issue (performance wise) we could still inline the fast > path of dec_and_lock(). For everyone who lost track, we are now down to this: |577cd5e894ec alpha: remove custom dec_and_lock() implementation |831923bd9fd3 spinlock: atomic_dec_and_lock: Add an irqsave variant |f452abc5a3c1 drivers/md/raid5: Use irqsave variant of atomic_dec_and_lock() |8e609dcd8b0d drivers/md/raid5: Do not disable irq on release_inactive_stripe_list() call |fdf84cf5bee4 bdi: use refcount_t for reference counting instead atomic_t |c3fce80b4112 userns: use refcount_t for reference counting instead atomic_t |a9595ca33f5b locking/refcount: implement refcount_dec_and_lock_irqsave() |32ded25a41a9 bdi: Use irqsave variant of refcount_dec_and_lock() |6a63e34ed625 userns: Use irqsave variant of refcount_dec_and_lock() for reference at https://git.kernel.org/pub/scm/linux/kernel/git/bigeasy/staging.git/log/?h=refcount_t_irqsave Sebastian -- To unsubscribe from this list: send the line "unsubscribe linux-alpha" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
