From: Pavel Begunkov <[email protected]> To: Jens Axboe <[email protected]>, [email protected] Subject: [PATCH -v2 2/2] io_uring: cancel more aggressively in exit_work Date: Mon, 4 Jan 2021 20:39:08 +0000 [thread overview] Message-ID: <f241b61d3d04b668f186af0cd465b04699274696.1609792653.git.asml.silence@gmail.com> (raw) In-Reply-To: <[email protected]> While io_ring_exit_work() is running new requests of all sorts may be issued, so it should do a bit more to cancel them, otherwise they may just get stuck. e.g. in io-wq, in poll lists, etc. Signed-off-by: Pavel Begunkov <[email protected]> --- fs/io_uring.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 85de42c42433..5bccb235271f 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -992,6 +992,9 @@ enum io_mem_account { ACCT_PINNED, }; +static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, + struct task_struct *task); + static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node); static struct fixed_file_ref_node *alloc_fixed_file_ref_node( struct io_ring_ctx *ctx); @@ -8675,7 +8678,7 @@ static void io_ring_exit_work(struct work_struct *work) * as nobody else will be looking for them. */ do { - io_iopoll_try_reap_events(ctx); + __io_uring_cancel_task_requests(ctx, NULL); } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)); io_ring_ctx_free(ctx); } @@ -8830,9 +8833,11 @@ static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, enum io_wq_cancel cret; bool ret = false; - cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, &cancel, true); - if (cret != IO_WQ_CANCEL_NOTFOUND) - ret = true; + if (ctx->io_wq) { + cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, + &cancel, true); + ret |= (cret != IO_WQ_CANCEL_NOTFOUND); + } /* SQPOLL thread does its own polling */ if (!(ctx->flags & IORING_SETUP_SQPOLL)) { -- 2.24.0
WARNING: multiple messages have this Message-ID (diff)
From: Pavel Begunkov <[email protected]> To: Jens Axboe <[email protected]>, [email protected] Subject: [PATCH -v2 2/2] io_uring: cancel more aggressively in exit_work Date: Mon, 4 Jan 2021 20:43:30 +0000 [thread overview] Message-ID: <f241b61d3d04b668f186af0cd465b04699274696.1609792653.git.asml.silence@gmail.com> (raw) Message-ID: <20210104204330.QQgfels17s9PUACtMa7zjAn02115khtDwvP-ps8zhUc@z> (raw) In-Reply-To: <[email protected]> While io_ring_exit_work() is running new requests of all sorts may be issued, so it should do a bit more to cancel them, otherwise they may just get stuck. e.g. in io-wq, in poll lists, etc. Signed-off-by: Pavel Begunkov <[email protected]> --- fs/io_uring.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 85de42c42433..5bccb235271f 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -992,6 +992,9 @@ enum io_mem_account { ACCT_PINNED, }; +static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, + struct task_struct *task); + static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node); static struct fixed_file_ref_node *alloc_fixed_file_ref_node( struct io_ring_ctx *ctx); @@ -8675,7 +8678,7 @@ static void io_ring_exit_work(struct work_struct *work) * as nobody else will be looking for them. */ do { - io_iopoll_try_reap_events(ctx); + __io_uring_cancel_task_requests(ctx, NULL); } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)); io_ring_ctx_free(ctx); } @@ -8830,9 +8833,11 @@ static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, enum io_wq_cancel cret; bool ret = false; - cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, &cancel, true); - if (cret != IO_WQ_CANCEL_NOTFOUND) - ret = true; + if (ctx->io_wq) { + cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, + &cancel, true); + ret |= (cret != IO_WQ_CANCEL_NOTFOUND); + } /* SQPOLL thread does its own polling */ if (!(ctx->flags & IORING_SETUP_SQPOLL)) { -- 2.24.0
next prev parent reply other threads:[~2021-01-04 23:32 UTC|newest] Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-01-04 20:39 [PATCH 0/2] cancellation fixes Pavel Begunkov 2021-01-04 20:39 ` Pavel Begunkov [this message] 2021-01-04 20:43 ` [PATCH -v2 2/2] io_uring: cancel more aggressively in exit_work Pavel Begunkov 2021-01-04 20:43 ` [PATCH 0/2] cancellation fixes Pavel Begunkov 2021-01-04 20:43 ` [PATCH v2 1/2] io_uring: drop file refs after task cancel Pavel Begunkov 2021-01-04 20:39 ` Pavel Begunkov 2021-01-04 22:23 ` [PATCH 0/2] cancellation fixes Jens Axboe
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=f241b61d3d04b668f186af0cd465b04699274696.1609792653.git.asml.silence@gmail.com \ [email protected] \ [email protected] \ [email protected] \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox