[net-next PATCH] ixgbe: Fix a set_num_queues() bug that can result in num_(r|t)x_queues = 0

Subsystems: networking drivers, the rest

STALE6463d

2 messages, 2 authors, 2009-02-07 · open the first message on its own page

[net-next PATCH] ixgbe: Fix a set_num_queues() bug that can result in num_(r|t)x_queues = 0

From: Jeff Kirsher <hidden>
Date: 2009-02-07 08:33:23

From: PJ Waskiewicz <redacted>

Now that our set_num_queues() routines for each feature are re-entrant, and
can be called at any point, they shouldn't zero out the feature's indices
or mask bits.  Subsequent calls into those routines for those features can
result in zero Rx and Tx queues being assigned, causing a panic later in
driver reinitialization.

Signed-off-by: Peter P Waskiewicz Jr <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---

 drivers/net/ixgbe/ixgbe_main.c |    4 ----
 1 files changed, 0 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 8e270b6..a3572d1 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2350,8 +2350,6 @@ static inline bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter)
 		                      adapter->ring_feature[RING_F_DCB].indices;
 		ret = true;
 	} else {
-		adapter->ring_feature[RING_F_DCB].mask = 0;
-		adapter->ring_feature[RING_F_DCB].indices = 0;
 		ret = false;
 	}
 
@@ -2371,8 +2369,6 @@ static inline bool ixgbe_set_rss_queues(struct ixgbe_adapter *adapter)
 		                      adapter->ring_feature[RING_F_RSS].indices;
 		ret = true;
 	} else {
-		adapter->ring_feature[RING_F_RSS].mask = 0;
-		adapter->ring_feature[RING_F_RSS].indices = 0;
 		ret = false;
 	}
 

Re: [net-next PATCH] ixgbe: Fix a set_num_queues() bug that can result in num_(r|t)x_queues = 0

From: David Miller <davem@davemloft.net>
Date: 2009-02-07 10:17:13

From: Jeff Kirsher <redacted>
Date: Sat, 07 Feb 2009 00:33:02 -0800
Now that our set_num_queues() routines for each feature are re-entrant, and
can be called at any point, they shouldn't zero out the feature's indices
or mask bits.  Subsequent calls into those routines for those features can
result in zero Rx and Tx queues being assigned, causing a panic later in
driver reinitialization.

Signed-off-by: Peter P Waskiewicz Jr <redacted>
Signed-off-by: Jeff Kirsher <redacted>
Applied.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help