The number of outstanding kiocbs is one of the few shared things left that has to be touched for every kiocb - it'd be nice to make it percpu. We can make it per cpu by treating it like an allocation problem: we have a maximum number of kiocbs that can be outstanding (i.e. slots) - then we just allocate and free slots, and we know how to write per cpu allocators. So as prep work for that, we convert reqs_active to reqs_available. Signed-off-by: Kent Overstreet <koverstreet@xxxxxxxxxx> --- fs/aio.c | 29 ++++++++++++++--------------- 1 file changed, 14 insertions(+), 15 deletions(-) diff --git a/fs/aio.c b/fs/aio.c index 7dee3aa..e6f29dc 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -79,7 +79,7 @@ struct kioctx { long nr_pages; struct { - atomic_t reqs_active; + atomic_t reqs_available; } ____cacheline_aligned; struct { @@ -301,17 +301,17 @@ static void free_ioctx(struct kioctx *ctx) head = ring->head; kunmap_atomic(ring); - while (atomic_read(&ctx->reqs_active) > 0) { + while (atomic_read(&ctx->reqs_available) < ctx->nr) { wait_event(ctx->wait, head != ctx->tail); avail = (head < ctx->tail ? ctx->tail : ctx->nr) - head; - atomic_sub(avail, &ctx->reqs_active); + atomic_add(avail, &ctx->reqs_available); head += avail; head %= ctx->nr; } - WARN_ON(atomic_read(&ctx->reqs_active) < 0); + WARN_ON(atomic_read(&ctx->reqs_available) > ctx->nr); aio_free_ring(ctx); @@ -368,6 +368,8 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) if (aio_setup_ring(ctx) < 0) goto out_freectx; + atomic_set(&ctx->reqs_available, ctx->nr); + /* limit the number of system wide aios */ spin_lock(&aio_nr_lock); if (aio_nr + nr_events > aio_max_nr || @@ -448,7 +450,7 @@ void exit_aio(struct mm_struct *mm) "exit_aio:ioctx still alive: %d %d %d\n", atomic_read(&ctx->users), atomic_read(&ctx->dead), - atomic_read(&ctx->reqs_active)); + atomic_read(&ctx->reqs_available)); /* * We don't need to bother with munmap() here - * exit_mmap(mm) is coming and it'll unmap everything. @@ -476,12 +478,9 @@ static inline struct kiocb *aio_get_req(struct kioctx *ctx) { struct kiocb *req; - if (atomic_read(&ctx->reqs_active) >= ctx->nr) + if (atomic_dec_if_positive(&ctx->reqs_available) <= 0) return NULL; - if (atomic_inc_return(&ctx->reqs_active) > ctx->nr) - goto out_put; - req = kmem_cache_alloc(kiocb_cachep, GFP_KERNEL|__GFP_ZERO); if (unlikely(!req)) goto out_put; @@ -491,7 +490,7 @@ static inline struct kiocb *aio_get_req(struct kioctx *ctx) return req; out_put: - atomic_dec(&ctx->reqs_active); + atomic_inc(&ctx->reqs_available); return NULL; } @@ -565,7 +564,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2) /* * Take rcu_read_lock() in case the kioctx is being destroyed, as we - * need to issue a wakeup after decrementing reqs_active. + * need to issue a wakeup after incrementing reqs_available. */ rcu_read_lock(); @@ -582,7 +581,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2) * when the event got cancelled. */ if (test_and_set_bit(KIF_CANCELLED, &iocb->ki_flags)) { - atomic_dec(&ctx->reqs_active); + atomic_inc(&ctx->reqs_available); /* Still need the wake_up in case free_ioctx is waiting */ goto put_rq; } @@ -707,7 +706,7 @@ static int aio_read_events(struct kioctx *ctx, struct io_event __user *event, ring->head = head; kunmap_atomic(ring); - atomic_sub(ret, &ctx->reqs_active); + atomic_add(ret, &ctx->reqs_available); pr_debug("%d h%u t%u\n", ret, head, ctx->tail); out: @@ -770,7 +769,7 @@ static int read_events(struct kioctx *ctx, break; /* Try to only show up in io wait if there are ops in flight */ - if (atomic_read(&ctx->reqs_active)) + if (atomic_read(&ctx->reqs_available) != ctx->nr) io_schedule(); else schedule(); @@ -1161,7 +1160,7 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, return 0; out_put_req: - atomic_dec(&ctx->reqs_active); + atomic_inc(&ctx->reqs_available); aio_put_req(req); /* drop extra ref to req */ aio_put_req(req); /* drop i/o ref to req */ return ret; -- 1.7.12 -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html