[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1458115601-5762-22-git-send-email-lizf@kernel.org>
Date: Wed, 16 Mar 2016 16:05:16 +0800
From: lizf@...nel.org
To: stable@...r.kernel.org
Cc: linux-kernel@...r.kernel.org,
Wengang Wang <wen.gang.wang@...cle.com>,
Doug Ledford <dledford@...hat.com>,
Zefan Li <lizefan@...wei.com>
Subject: [PATCH 3.4 022/107] rds: rds_ib_device.refcount overflow
From: Wengang Wang <wen.gang.wang@...cle.com>
3.4.111-rc1 review patch. If anyone has any objections, please let me know.
------------------
commit 4fabb59449aa44a585b3603ffdadd4c5f4d0c033 upstream.
Fixes: 3e0249f9c05c ("RDS/IB: add refcount tracking to struct rds_ib_device")
There lacks a dropping on rds_ib_device.refcount in case rds_ib_alloc_fmr
failed(mr pool running out). this lead to the refcount overflow.
A complain in line 117(see following) is seen. From vmcore:
s_ib_rdma_mr_pool_depleted is 2147485544 and rds_ibdev->refcount is -2147475448.
That is the evidence the mr pool is used up. so rds_ib_alloc_fmr is very likely
to return ERR_PTR(-EAGAIN).
115 void rds_ib_dev_put(struct rds_ib_device *rds_ibdev)
116 {
117 BUG_ON(atomic_read(&rds_ibdev->refcount) <= 0);
118 if (atomic_dec_and_test(&rds_ibdev->refcount))
119 queue_work(rds_wq, &rds_ibdev->free_work);
120 }
fix is to drop refcount when rds_ib_alloc_fmr failed.
Signed-off-by: Wengang Wang <wen.gang.wang@...cle.com>
Reviewed-by: Haggai Eran <haggaie@...lanox.com>
Signed-off-by: Doug Ledford <dledford@...hat.com>
Signed-off-by: Zefan Li <lizefan@...wei.com>
---
net/rds/ib_rdma.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
index e8fdb17..a985158 100644
--- a/net/rds/ib_rdma.c
+++ b/net/rds/ib_rdma.c
@@ -759,8 +759,10 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
}
ibmr = rds_ib_alloc_fmr(rds_ibdev);
- if (IS_ERR(ibmr))
+ if (IS_ERR(ibmr)) {
+ rds_ib_dev_put(rds_ibdev);
return ibmr;
+ }
ret = rds_ib_map_fmr(rds_ibdev, ibmr, sg, nents);
if (ret == 0)
--
1.9.1
Powered by blists - more mailing lists