linux-kernel - [PATCH 13/20] drbd: improve resync request throttling due to sendbuf size

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

Message-Id: <1404231410-29852-14-git-send-email-philipp.reisner@linbit.com>
Date:	Tue,  1 Jul 2014 18:16:43 +0200
From:	Philipp Reisner <philipp.reisner@...bit.com>
To:	linux-kernel@...r.kernel.org, Jens Axboe <axboe@...nel.dk>
Cc:	drbd-dev@...ts.linbit.com
Subject: [PATCH 13/20] drbd: improve resync request throttling due to sendbuf size

From: Lars Ellenberg <lars.ellenberg@...bit.com>

If we throttle resync because the socket sendbuffer is filling up,
tell TCP about it, so it may expand the sendbuffer for us.

Signed-off-by: Philipp Reisner <philipp.reisner@...bit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@...bit.com>
---
 drivers/block/drbd/drbd_worker.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 6532a69..0b5e429 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -592,7 +592,7 @@ static int make_resync_request(struct drbd_device *const device, int cancel)
 	const sector_t capacity = drbd_get_capacity(device->this_bdev);
 	int max_bio_size;
 	int number, rollback_i, size;
-	int align, queued, sndbuf;
+	int align, requeue = 0;
 	int i = 0;
 
 	if (unlikely(cancel))
@@ -619,17 +619,22 @@ static int make_resync_request(struct drbd_device *const device, int cancel)
 		goto requeue;
 
 	for (i = 0; i < number; i++) {
-		/* Stop generating RS requests, when half of the send buffer is filled */
+		/* Stop generating RS requests when half of the send buffer is filled,
+		 * but notify TCP that we'd like to have more space. */
 		mutex_lock(&connection->data.mutex);
 		if (connection->data.socket) {
-			queued = connection->data.socket->sk->sk_wmem_queued;
-			sndbuf = connection->data.socket->sk->sk_sndbuf;
-		} else {
-			queued = 1;
-			sndbuf = 0;
-		}
+			struct sock *sk = connection->data.socket->sk;
+			int queued = sk->sk_wmem_queued;
+			int sndbuf = sk->sk_sndbuf;
+			if (queued > sndbuf / 2) {
+				requeue = 1;
+				if (sk->sk_socket)
+					set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+			}
+		} else
+			requeue = 1;
 		mutex_unlock(&connection->data.mutex);
-		if (queued > sndbuf / 2)
+		if (requeue)
 			goto requeue;
 
 next_sector:
-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/