netdev - [PATCH net] net: fix hangup on napi

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-Id: <996c4bb33166b5cf8d881871ea8b61e54ad4da24.1617230551.git.pabeni@redhat.com>
Date:   Thu,  1 Apr 2021 00:46:18 +0200
From:   Paolo Abeni <pabeni@...hat.com>
To:     netdev@...r.kernel.org
Cc:     Eric Dumazet <edumazet@...gle.com>, Wei Wang <weiwan@...gle.com>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>
Subject: [PATCH net] net: fix hangup on napi_disable for threaded napi

I hit an hangup on napi_disable(), when the threaded
mode is enabled and the napi is under heavy traffic.

If the relevant napi has been scheduled and the napi_disable()
kicks in before the next napi_threaded_wait() completes - so
that the latter quits due to the napi_disable_pending() condition,
the existing code leaves the NAPI_STATE_SCHED bit set and the
napi_disable() loop waiting for such bit will hang.

Address the issue explicitly clearing the SCHED_BIT on napi_thread
termination, if the thread is owns the napi.

Fixes: 29863d41bb6e ("net: implement threaded-able napi poll loop support")
Signed-off-by: Paolo Abeni <pabeni@...hat.com>
---
 net/core/dev.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/net/core/dev.c b/net/core/dev.c
index b4c67a5be606d..e2e716ba027b8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7059,6 +7059,14 @@ static int napi_thread_wait(struct napi_struct *napi)
 		set_current_state(TASK_INTERRUPTIBLE);
 	}
 	__set_current_state(TASK_RUNNING);
+
+	/* if the thread owns this napi, and the napi itself has been disabled
+	 * in-between napi_schedule() and the above napi_disable_pending()
+	 * check, we need to clear the SCHED bit here, or napi_disable
+	 * will hang waiting for such bit being cleared
+	 */
+	if (test_bit(NAPI_STATE_SCHED_THREADED, &napi->state) || woken)
+		clear_bit(NAPI_STATE_SCHED, &napi->state);
 	return -1;
 }
 
-- 
2.26.2