[PATCH 5.15] io_uring: apply worker limits to previous users

[PATCH 5.15] io_uring: apply worker limits to previous users

[Pavel Begunkov]

Another change to the API io-wq worker limitation API added in 5.15,
apply the limit to all prior users that already registered a tctx. It
may be confusing as it's now, in particular the change covers the
following 2 cases:

TASK1                   | TASK2
_________________________________________________
ring = create()         |
                        | limit_iowq_workers()
*not limited*           |

TASK1                   | TASK2
_________________________________________________
ring = create()         |
                        | issue_requests()
limit_iowq_workers()    |
                        | *not limited*

A note on locking, it's safe to traverse ->tctx_list as we hold
->uring_lock, but do that after dropping sqd->lock to avoid possible
problems. It's also safe to access tctx->io_wq there because tasks
kill it only after removing themselves from tctx_list, see
io_uring_cancel_generic() -> io_uring_clean_tctx()

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 fs/io_uring.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index d5cc103224f1..bc18af5e0a93 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -10649,7 +10649,9 @@ static int io_unregister_iowq_aff(struct io_ring_ctx *ctx)
 
 static int io_register_iowq_max_workers(struct io_ring_ctx *ctx,
 					void __user *arg)
+	__must_hold(&ctx->uring_lock)
 {
+	struct io_tctx_node *node;
 	struct io_uring_task *tctx = NULL;
 	struct io_sq_data *sqd = NULL;
 	__u32 new_count[2];
@@ -10702,6 +10704,22 @@ static int io_register_iowq_max_workers(struct io_ring_ctx *ctx,
 	if (copy_to_user(arg, new_count, sizeof(new_count)))
 		return -EFAULT;
 
+	/* that's it for SQPOLL, only the SQPOLL task creates requests */
+	if (sqd)
+		return 0;
+
+	/* now propagate the restriction to all registered users */
+	list_for_each_entry(node, &ctx->tctx_list, ctx_node) {
+		struct io_uring_task *tctx = node->task->io_uring;
+
+		if (WARN_ON_ONCE(!tctx->io_wq))
+			continue;
+
+		for (i = 0; i < ARRAY_SIZE(new_count); i++)
+			new_count[i] = ctx->iowq_limits[i];
+		/* ignore errors, it always returns zero anyway */
+		(void)io_wq_max_workers(tctx->io_wq, new_count);
+	}
 	return 0;
 err:
 	if (sqd) {
-- 
2.33.1

Re: [PATCH 5.15] io_uring: apply worker limits to previous users

[Jens Axboe]

On Thu, 21 Oct 2021 13:20:29 +0100, Pavel Begunkov wrote:
> Another change to the API io-wq worker limitation API added in 5.15,
> apply the limit to all prior users that already registered a tctx. It
> may be confusing as it's now, in particular the change covers the
> following 2 cases:
> 
> TASK1                   | TASK2
> _________________________________________________
> ring = create()         |
>                         | limit_iowq_workers()
> *not limited*           |
> 
> [...]

Applied, thanks!

[1/1] io_uring: apply worker limits to previous users
      commit: 7c810c192d6c18213db7cafc647a46398519094a

Best regards,
-- 
Jens Axboe

Re: [PATCH 5.15] io_uring: apply worker limits to previous users

[Hao Xu]

在 2021/10/21 下午8:20, Pavel Begunkov 写道:
> Another change to the API io-wq worker limitation API added in 5.15,
> apply the limit to all prior users that already registered a tctx. It
> may be confusing as it's now, in particular the change covers the
> following 2 cases:
> 
> TASK1                   | TASK2
> _________________________________________________
> ring = create()         |
>                          | limit_iowq_workers()
> *not limited*           |
> 
> TASK1                   | TASK2
> _________________________________________________
> ring = create()         |
>                          | issue_requests()
> limit_iowq_workers()    |
>                          | *not limited*
> 
> A note on locking, it's safe to traverse ->tctx_list as we hold
> ->uring_lock, but do that after dropping sqd->lock to avoid possible
> problems. It's also safe to access tctx->io_wq there because tasks
> kill it only after removing themselves from tctx_list, see
> io_uring_cancel_generic() -> io_uring_clean_tctx()
> 
> Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
> ---
>   fs/io_uring.c | 18 ++++++++++++++++++
>   1 file changed, 18 insertions(+)
> 
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index d5cc103224f1..bc18af5e0a93 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -10649,7 +10649,9 @@ static int io_unregister_iowq_aff(struct io_ring_ctx *ctx)
>   
>   static int io_register_iowq_max_workers(struct io_ring_ctx *ctx,
>   					void __user *arg)
> +	__must_hold(&ctx->uring_lock)
>   {
> +	struct io_tctx_node *node;
>   	struct io_uring_task *tctx = NULL;
>   	struct io_sq_data *sqd = NULL;
>   	__u32 new_count[2];
> @@ -10702,6 +10704,22 @@ static int io_register_iowq_max_workers(struct io_ring_ctx *ctx,
>   	if (copy_to_user(arg, new_count, sizeof(new_count)))
>   		return -EFAULT;
>   
> +	/* that's it for SQPOLL, only the SQPOLL task creates requests */
> +	if (sqd)
> +		return 0;
> +
> +	/* now propagate the restriction to all registered users */
> +	list_for_each_entry(node, &ctx->tctx_list, ctx_node) {
> +		struct io_uring_task *tctx = node->task->io_uring;
> +
> +		if (WARN_ON_ONCE(!tctx->io_wq))
> +			continue;
> +
> +		for (i = 0; i < ARRAY_SIZE(new_count); i++)
> +			new_count[i] = ctx->iowq_limits[i];
> +		/* ignore errors, it always returns zero anyway */
> +		(void)io_wq_max_workers(tctx->io_wq, new_count);
> +	}
>   	return 0;
>   err:
>   	if (sqd) {
> 
Reviewed-by: Hao Xu <haoxu@linux.alibaba.com>