diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2023-04-04 15:39:45 +0300 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2023-04-04 18:30:39 +0300 |
| commit | b8fb5b4fdd67f9d18109c5d21d44a8bd4ddb608b (patch) | |
| tree | d88a6c8049aeb16442461afa53182a2c603a5eb6 /io_uring/rsrc.c | |
| parent | e3ef728ff07b42668e7e12f49cd2f9055e064ec1 (diff) | |
| download | linux-b8fb5b4fdd67f9d18109c5d21d44a8bd4ddb608b.tar.xz | |
io_uring/rsrc: use non-pcpu refcounts for nodes
One problem with the current rsrc infra is that often updates will
generates lots of rsrc nodes, each carry pcpu refs. That takes quite a
lot of memory, especially if there is a stall, and takes lots of CPU
cycles. Only pcpu allocations takes >50 of CPU with a naive benchmark
updating files in a loop.
Replace pcpu refs with normal refcounting. There is already a hot path
avoiding atomics / refs, but following patches will further improve it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e9ed8a9457b331a26555ff9443afc64cdaab7247.1680576071.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'io_uring/rsrc.c')
| -rw-r--r-- | io_uring/rsrc.c | 15 |
1 files changed, 5 insertions, 10 deletions
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 7a43aed8e395..f2da9e251e3f 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -155,7 +155,7 @@ void io_rsrc_refs_refill(struct io_ring_ctx *ctx) __must_hold(&ctx->uring_lock) { ctx->rsrc_cached_refs += IO_RSRC_REF_BATCH; - percpu_ref_get_many(&ctx->rsrc_node->refs, IO_RSRC_REF_BATCH); + refcount_add(IO_RSRC_REF_BATCH, &ctx->rsrc_node->refs); } static void __io_rsrc_put_work(struct io_rsrc_node *ref_node) @@ -220,13 +220,11 @@ void io_wait_rsrc_data(struct io_rsrc_data *data) void io_rsrc_node_destroy(struct io_rsrc_node *ref_node) { - percpu_ref_exit(&ref_node->refs); kfree(ref_node); } -static __cold void io_rsrc_node_ref_zero(struct percpu_ref *ref) +__cold void io_rsrc_node_ref_zero(struct io_rsrc_node *node) { - struct io_rsrc_node *node = container_of(ref, struct io_rsrc_node, refs); struct io_ring_ctx *ctx = node->rsrc_data->ctx; unsigned long flags; bool first_add = false; @@ -269,11 +267,7 @@ static struct io_rsrc_node *io_rsrc_node_alloc(void) if (!ref_node) return NULL; - if (percpu_ref_init(&ref_node->refs, io_rsrc_node_ref_zero, - 0, GFP_KERNEL)) { - kfree(ref_node); - return NULL; - } + refcount_set(&ref_node->refs, 1); INIT_LIST_HEAD(&ref_node->node); INIT_LIST_HEAD(&ref_node->rsrc_list); ref_node->done = false; @@ -298,7 +292,8 @@ void io_rsrc_node_switch(struct io_ring_ctx *ctx, spin_unlock_irq(&ctx->rsrc_ref_lock); atomic_inc(&data_to_kill->refs); - percpu_ref_kill(&rsrc_node->refs); + /* put master ref */ + io_rsrc_put_node(rsrc_node, 1); ctx->rsrc_node = NULL; } |