diff options
| author | Michael S. Tsirkin <mst@mellanox.co.il> | 2006-01-30 16:22:29 -0800 |
|---|---|---|
| committer | Roland Dreier <rolandd@cisco.com> | 2006-01-30 16:22:29 -0800 |
| commit | e3aa31c517cb6fd0a3d8b23e6a7e71a6aafc2393 (patch) | |
| tree | 97c1ca504dc60a7b380be402b91af59a4a8f8e04 /drivers/infiniband/hw/mthca/mthca_mcg.c | |
| parent | 8e9e5f4f5eb1d44ddabfd1ddea4ca4e4244a9ffb (diff) | |
| download | blackbird-obmc-linux-e3aa31c517cb6fd0a3d8b23e6a7e71a6aafc2393.tar.gz blackbird-obmc-linux-e3aa31c517cb6fd0a3d8b23e6a7e71a6aafc2393.zip | |
IB/mthca: Don't cancel commands on a signal
We have run into the following problem: if a task receives a signal
while in the process of e.g. destroying a resource (which could be
because the relevant file was closed) mthca could bail out from trying
to take a command interface semaphore without performing the
appropriate command to tell hardware that the resource is being
destroyed.
As a result we see messages like
ib_mthca 0000:04:00.0: HW2SW_CQ failed (-4)
In this case, hardware could access the resource after the memory has
been freed, possibly causing memory corruption.
A simple solution is to replace down_interruptible() by down() in
command interface activation.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
[ It's also not safe to bail out on multicast table operations, since
they may be invoked on the cleanup path too. So use down() for
mcg_table.sem too. ]
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Diffstat (limited to 'drivers/infiniband/hw/mthca/mthca_mcg.c')
| -rw-r--r-- | drivers/infiniband/hw/mthca/mthca_mcg.c | 14 |
1 files changed, 4 insertions, 10 deletions
diff --git a/drivers/infiniband/hw/mthca/mthca_mcg.c b/drivers/infiniband/hw/mthca/mthca_mcg.c index 77bc6c746f43..55ff5e5bcfef 100644 --- a/drivers/infiniband/hw/mthca/mthca_mcg.c +++ b/drivers/infiniband/hw/mthca/mthca_mcg.c @@ -154,10 +154,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) return PTR_ERR(mailbox); mgm = mailbox->buf; - if (down_interruptible(&dev->mcg_table.sem)) { - err = -EINTR; - goto err_sem; - } + down(&dev->mcg_table.sem); err = find_mgm(dev, gid->raw, mailbox, &hash, &prev, &index); if (err) @@ -242,7 +239,7 @@ int mthca_multicast_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) mthca_free(&dev->mcg_table.alloc, index); } up(&dev->mcg_table.sem); - err_sem: + mthca_free_mailbox(dev, mailbox); return err; } @@ -263,10 +260,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) return PTR_ERR(mailbox); mgm = mailbox->buf; - if (down_interruptible(&dev->mcg_table.sem)) { - err = -EINTR; - goto err_sem; - } + down(&dev->mcg_table.sem); err = find_mgm(dev, gid->raw, mailbox, &hash, &prev, &index); if (err) @@ -372,7 +366,7 @@ int mthca_multicast_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) out: up(&dev->mcg_table.sem); - err_sem: + mthca_free_mailbox(dev, mailbox); return err; } |
