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]
Date:   Mon, 19 Jun 2017 13:22:23 +0200
From:   Julian Wiedmann <jwi@...ux.vnet.ibm.com>
To:     David Miller <davem@...emloft.net>
Cc:     <netdev@...r.kernel.org>, <linux-s390@...r.kernel.org>,
        Martin Schwidefsky <schwidefsky@...ibm.com>,
        Heiko Carstens <heiko.carstens@...ibm.com>,
        Stefan Raspl <raspl@...ux.vnet.ibm.com>,
        Ursula Braun <ubraun@...ux.vnet.ibm.com>,
        Julian Wiedmann <jwi@...ux.vnet.ibm.com>
Subject: [PATCH net-next 2/4] s390/qeth: fix packing buffer statistics

There's two spots in qeth_send_packet() where we don't accurately
account for transmitted packing buffers in qeth's performance
statistics:

1) when flushing the current buffer due to insufficient size,
   and the next buffer is not EMPTY, we need to account for that
   flushed buffer.
2) when synchronizing with the TX completion code, we reset
   flush_count and thus forget to account for any previously
   flushed buffers.

Reported-by: Nils Hoppmann <niho@...ibm.com>
Signed-off-by: Julian Wiedmann <jwi@...ux.vnet.ibm.com>
---
 drivers/s390/net/qeth_core_main.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 08338f27c82c..13a55f1dee7a 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -4103,7 +4103,8 @@ int qeth_do_send_packet(struct qeth_card *card, struct qeth_qdio_out_q *queue,
 							   flush_count);
 				atomic_set(&queue->state,
 						QETH_OUT_Q_UNLOCKED);
-				return -EBUSY;
+				rc = -EBUSY;
+				goto out;
 			}
 		}
 	}
@@ -4122,19 +4123,21 @@ int qeth_do_send_packet(struct qeth_card *card, struct qeth_qdio_out_q *queue,
 	 * In that case we will enter this loop
 	 */
 	while (atomic_dec_return(&queue->state)) {
-		flush_count = 0;
 		start_index = queue->next_buf_to_fill;
 		/* check if we can go back to non-packing state */
-		flush_count += qeth_switch_to_nonpacking_if_needed(queue);
+		tmp = qeth_switch_to_nonpacking_if_needed(queue);
 		/*
 		 * check if we need to flush a packing buffer to get a pci
 		 * flag out on the queue
 		 */
-		if (!flush_count && !atomic_read(&queue->set_pci_flags_count))
-			flush_count += qeth_prep_flush_pack_buffer(queue);
-		if (flush_count)
-			qeth_flush_buffers(queue, start_index, flush_count);
+		if (!tmp && !atomic_read(&queue->set_pci_flags_count))
+			tmp = qeth_prep_flush_pack_buffer(queue);
+		if (tmp) {
+			qeth_flush_buffers(queue, start_index, tmp);
+			flush_count += tmp;
+		}
 	}
+out:
 	/* at this point the queue is UNLOCKED again */
 	if (queue->card->options.performance_stats && do_pack)
 		queue->card->perf_stats.bufs_sent_pack += flush_count;
-- 
2.11.2

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ