linux-kernel - [PATCH 5.13 145/175] io-wq: fix race between worker exiting and activating free worker

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20210810173005.743233009@linuxfoundation.org>
Date:   Tue, 10 Aug 2021 19:30:53 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Nadav Amit <nadav.amit@...il.com>,
        Jens Axboe <axboe@...nel.dk>
Subject: [PATCH 5.13 145/175] io-wq: fix race between worker exiting and activating free worker

From: Jens Axboe <axboe@...nel.dk>

commit 83d6c39310b6d11199179f6384c2b0a415389597 upstream.

Nadav correctly reports that we have a race between a worker exiting,
and new work being queued. This can lead to work being queued behind
an existing worker that could be sleeping on an event before it can
run to completion, and hence introducing potential big latency gaps
if we hit this race condition:

cpu0					cpu1
----					----
					io_wqe_worker()
					schedule_timeout()
					 // timed out
io_wqe_enqueue()
io_wqe_wake_worker()
// work_flags & IO_WQ_WORK_CONCURRENT
io_wqe_activate_free_worker()
					 io_worker_exit()

Fix this by having the exiting worker go through the normal decrement
of a running worker, which will spawn a new one if needed.

The free worker activation is modified to only return success if we
were able to find a sleeping worker - if not, we keep looking through
the list. If we fail, we create a new worker as per usual.

Cc: stable@...r.kernel.org
Link: https://lore.kernel.org/io-uring/BFF746C0-FEDE-4646-A253-3021C57C26C9@gmail.com/
Reported-by: Nadav Amit <nadav.amit@...il.com>
Tested-by: Nadav Amit <nadav.amit@...il.com>
Signed-off-by: Jens Axboe <axboe@...nel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
 fs/io-wq.c |   38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)

--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -131,6 +131,7 @@ struct io_cb_cancel_data {
 };
 
 static void create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index);
+static void io_wqe_dec_running(struct io_worker *worker);
 
 static bool io_worker_get(struct io_worker *worker)
 {
@@ -169,26 +170,21 @@ static void io_worker_exit(struct io_wor
 {
 	struct io_wqe *wqe = worker->wqe;
 	struct io_wqe_acct *acct = io_wqe_get_acct(worker);
-	unsigned flags;
 
 	if (refcount_dec_and_test(&worker->ref))
 		complete(&worker->ref_done);
 	wait_for_completion(&worker->ref_done);
 
-	preempt_disable();
-	current->flags &= ~PF_IO_WORKER;
-	flags = worker->flags;
-	worker->flags = 0;
-	if (flags & IO_WORKER_F_RUNNING)
-		atomic_dec(&acct->nr_running);
-	worker->flags = 0;
-	preempt_enable();
-
 	raw_spin_lock_irq(&wqe->lock);
-	if (flags & IO_WORKER_F_FREE)
+	if (worker->flags & IO_WORKER_F_FREE)
 		hlist_nulls_del_rcu(&worker->nulls_node);
 	list_del_rcu(&worker->all_list);
 	acct->nr_workers--;
+	preempt_disable();
+	io_wqe_dec_running(worker);
+	worker->flags = 0;
+	current->flags &= ~PF_IO_WORKER;
+	preempt_enable();
 	raw_spin_unlock_irq(&wqe->lock);
 
 	kfree_rcu(worker, rcu);
@@ -215,15 +211,19 @@ static bool io_wqe_activate_free_worker(
 	struct hlist_nulls_node *n;
 	struct io_worker *worker;
 
-	n = rcu_dereference(hlist_nulls_first_rcu(&wqe->free_list));
-	if (is_a_nulls(n))
-		return false;
-
-	worker = hlist_nulls_entry(n, struct io_worker, nulls_node);
-	if (io_worker_get(worker)) {
-		wake_up_process(worker->task);
+	/*
+	 * Iterate free_list and see if we can find an idle worker to
+	 * activate. If a given worker is on the free_list but in the process
+	 * of exiting, keep trying.
+	 */
+	hlist_nulls_for_each_entry_rcu(worker, n, &wqe->free_list, nulls_node) {
+		if (!io_worker_get(worker))
+			continue;
+		if (wake_up_process(worker->task)) {
+			io_worker_release(worker);
+			return true;
+		}
 		io_worker_release(worker);
-		return true;
 	}
 
 	return false;