[<prev] [next>] [day] [month] [year] [list]
Message-Id: <20191227174352.6264-55-sashal@kernel.org>
Date: Fri, 27 Dec 2019 12:43:23 -0500
From: Sasha Levin <sashal@...nel.org>
To: linux-kernel@...r.kernel.org, stable@...r.kernel.org
Cc: Vishal Kulkarni <vishal@...lsio.com>,
Herat Ramani <herat@...lsio.com>,
Jakub Kicinski <jakub.kicinski@...ronome.com>,
Sasha Levin <sashal@...nel.org>, netdev@...r.kernel.org
Subject: [PATCH AUTOSEL 4.19 55/84] cxgb4: Fix kernel panic while accessing sge_info
From: Vishal Kulkarni <vishal@...lsio.com>
[ Upstream commit 479a0d1376f6d97c60871442911f1394d4446a25 ]
The sge_info debugfs collects offload queue info even when offload
capability is disabled and leads to panic.
[ 144.139871] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 144.139874] CR2: 0000000000000000 CR3: 000000082d456005 CR4: 00000000001606e0
[ 144.139876] Call Trace:
[ 144.139887] sge_queue_start+0x12/0x30 [cxgb4]
[ 144.139897] seq_read+0x1d4/0x3d0
[ 144.139906] full_proxy_read+0x50/0x70
[ 144.139913] vfs_read+0x89/0x140
[ 144.139916] ksys_read+0x55/0xd0
[ 144.139924] do_syscall_64+0x5b/0x1d0
[ 144.139933] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 144.139936] RIP: 0033:0x7f4b01493990
Fix this crash by skipping the offload queue access in sge_qinfo when
offload capability is disabled
Signed-off-by: Herat Ramani <herat@...lsio.com>
Signed-off-by: Vishal Kulkarni <vishal@...lsio.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@...ronome.com>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index b429b726b987..d320e9afab88 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -3035,6 +3035,9 @@ static int sge_queue_entries(const struct adapter *adap)
int tot_uld_entries = 0;
int i;
+ if (!is_uld(adap))
+ goto lld_only;
+
mutex_lock(&uld_mutex);
for (i = 0; i < CXGB4_TX_MAX; i++)
tot_uld_entries += sge_qinfo_uld_txq_entries(adap, i);
@@ -3045,6 +3048,7 @@ static int sge_queue_entries(const struct adapter *adap)
}
mutex_unlock(&uld_mutex);
+lld_only:
return DIV_ROUND_UP(adap->sge.ethqsets, 4) +
tot_uld_entries +
DIV_ROUND_UP(MAX_CTRL_QUEUES, 4) + 1;
--
2.20.1
Powered by blists - more mailing lists