<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/drivers/infiniband, branch v3.4.33</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/drivers/infiniband?h=v3.4.33</id>
<link rel='self' href='https://git.amat.us/linux/atom/drivers/infiniband?h=v3.4.33'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2013-01-17T16:50:51Z</updated>
<entry>
<title>RDMA/nes: Fix for terminate timer crash</title>
<updated>2013-01-17T16:50:51Z</updated>
<author>
<name>Tatyana Nikolova</name>
<email>Tatyana.E.Nikolova@intel.com</email>
</author>
<published>2012-12-06T19:58:27Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=0d38a8789caf511e5b997caf76eb32fddbff051c'/>
<id>urn:sha1:0d38a8789caf511e5b997caf76eb32fddbff051c</id>
<content type='text'>
commit 7bfcfa51c35cdd2d37e0d70fc11790642dd11fb3 upstream.

The terminate timer needs to be initialized just once.

Signed-off-by: Tatyana Nikolova &lt;Tatyana.E.Nikolova@intel.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: CAI Qian &lt;caiqian@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>RDMA/nes: Fix for crash when registering zero length MR for CQ</title>
<updated>2013-01-17T16:50:51Z</updated>
<author>
<name>Tatyana Nikolova</name>
<email>Tatyana.E.Nikolova@intel.com</email>
</author>
<published>2012-12-06T20:05:02Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=e5ca2896f817ef43802387ae2f41a9c70e30ee86'/>
<id>urn:sha1:e5ca2896f817ef43802387ae2f41a9c70e30ee86</id>
<content type='text'>
commit 7d9c199a55200c9b9fcad08e150470d02fb385be upstream.

Signed-off-by: Tatyana Nikolova &lt;Tatyana.E.Nikolova@intel.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: CAI Qian &lt;caiqian@redhat.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>infiniband: pass rdma_cm module to netlink_dump_start</title>
<updated>2012-10-28T17:14:15Z</updated>
<author>
<name>Gao feng</name>
<email>gaofeng@cn.fujitsu.com</email>
</author>
<published>2012-10-04T20:15:49Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=6114941a295ff186d29ab7462cce6a41a089c354'/>
<id>urn:sha1:6114941a295ff186d29ab7462cce6a41a089c354</id>
<content type='text'>
[ Upstream commit 809d5fc9bf6589276a12bd4fd611e4c7ff9940c3 ]

set netlink_dump_control.module to avoid panic.

Signed-off-by: Gao feng &lt;gaofeng@cn.fujitsu.com&gt;
Cc: Roland Dreier &lt;roland@kernel.org&gt;
Cc: Sean Hefty &lt;sean.hefty@intel.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>IB/srp: Avoid having aborted requests hang</title>
<updated>2012-10-07T15:32:29Z</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2012-08-24T10:29:11Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=22fb582405002812d8fb89d0ed1264e97d3d25ad'/>
<id>urn:sha1:22fb582405002812d8fb89d0ed1264e97d3d25ad</id>
<content type='text'>
commit d8536670916a685df116b5c2cb256573fd25e4e3 upstream.

We need to call scsi_done() for commands after we abort them.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Acked-by: David Dillow &lt;dillowda@ornl.gov&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>IB/srp: Fix use-after-free in srp_reset_req()</title>
<updated>2012-10-07T15:32:29Z</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2012-08-24T10:27:54Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=0e4765fd9c0f8df577e0b8cbea927c2051710f27'/>
<id>urn:sha1:0e4765fd9c0f8df577e0b8cbea927c2051710f27</id>
<content type='text'>
commit 9b796d06d5d1b1e85ae2316a283ea11dd739ef96 upstream.

srp_free_req() uses the scsi_cmnd structure contents to unmap
buffers, so we must invoke srp_free_req() before we release
ownership of that structure.

Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Acked-by: David Dillow &lt;dillowda@ornl.gov&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>IPoIB: Fix use-after-free of multicast object</title>
<updated>2012-10-07T15:32:28Z</updated>
<author>
<name>Patrick McHardy</name>
<email>kaber@trash.net</email>
</author>
<published>2012-08-30T07:01:30Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=350f3edb0a05c340853c75f4c5007a72ebd9e32f'/>
<id>urn:sha1:350f3edb0a05c340853c75f4c5007a72ebd9e32f</id>
<content type='text'>
commit bea1e22df494a729978e7f2c54f7bda328f74bc3 upstream.

Fix a crash in ipoib_mcast_join_task().  (with help from Or Gerlitz)

Commit c8c2afe360b7 ("IPoIB: Use rtnl lock/unlock when changing device
flags") added a call to rtnl_lock() in ipoib_mcast_join_task(), which
is run from the ipoib_workqueue, and hence the workqueue can't be
flushed from the context of ipoib_stop().

In the current code, ipoib_stop() (which doesn't flush the workqueue)
calls ipoib_mcast_dev_flush(), which goes and deletes all the
multicast entries.  This takes place without any synchronization with
a possible running instance of ipoib_mcast_join_task() for the same
ipoib device, leading to a crash due to NULL pointer dereference.

Fix this by making sure that the workqueue is flushed before
ipoib_mcast_dev_flush() is called.  To make that possible, we move the
RTNL-lock wrapped code to ipoib_mcast_join_finish().

Signed-off-by: Patrick McHardy &lt;kaber@trash.net&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>IB/srp: Fix a race condition</title>
<updated>2012-08-26T22:00:44Z</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2012-08-14T13:18:53Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=ddffad3427e2e80fff78d2fc77f21afbba3d11b0'/>
<id>urn:sha1:ddffad3427e2e80fff78d2fc77f21afbba3d11b0</id>
<content type='text'>
commit 220329916c72ee3d54ae7262b215a050f04a18fc upstream.

Avoid a crash caused by the scmnd-&gt;scsi_done(scmnd) call in
srp_process_rsp() being invoked with scsi_done == NULL.  This can
happen if a reply is received during or after a command abort.

Reported-by: Joseph Glanville &lt;joseph.glanville@orionvm.com.au&gt;
Reference: http://marc.info/?l=linux-rdma&amp;m=134314367801595
Acked-by: David Dillow &lt;dillowda@ornl.gov&gt;
Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>IB/core: Fix mismatch between locked and pinned pages</title>
<updated>2012-06-01T07:18:17Z</updated>
<author>
<name>Yishai Hadas</name>
<email>yishaih@mellanox.com</email>
</author>
<published>2012-05-10T20:28:05Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=b10dfc7a2b20aaf6fad56f7b0725c4aaa9438ad1'/>
<id>urn:sha1:b10dfc7a2b20aaf6fad56f7b0725c4aaa9438ad1</id>
<content type='text'>
commit c4870eb874ac16dccef40e1bc7a002c7e9156adc upstream.

Commit bc3e53f682d9 ("mm: distinguish between mlocked and pinned
pages") introduced a separate counter for pinned pages and used it in
the IB stack.  However, in ib_umem_get() the pinned counter is
incremented, but ib_umem_release() wrongly decrements the locked
counter.  Fix this.

Signed-off-by: Yishai Hadas &lt;yishaih@mellanox.com&gt;
Reviewed-by: Christoph Lameter &lt;cl@linux.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>RDMA/cxgb4: Drop peer_abort when no endpoint found</title>
<updated>2012-06-01T07:18:16Z</updated>
<author>
<name>Steve Wise</name>
<email>swise@opengridcomputing.com</email>
</author>
<published>2012-04-30T20:31:29Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=ece17ce979baa62e074ba2d617e3a3ffc77ada84'/>
<id>urn:sha1:ece17ce979baa62e074ba2d617e3a3ffc77ada84</id>
<content type='text'>
commit 14b9222808bb8bfefc71f72bc0dbdcf3b2f0140f upstream.

Log a warning and drop the abort message.  Otherwise we will do a
bogus wake_up() and crash.

Signed-off-by: Steve Wise &lt;swise@opengridcomputing.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
<entry>
<title>RDMA/cxgb4: Use dst parameter in import_ep()</title>
<updated>2012-06-01T07:18:16Z</updated>
<author>
<name>Steve Wise</name>
<email>swise@opengridcomputing.com</email>
</author>
<published>2012-04-27T15:24:33Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=466dab407e47968777912aa23e5c66d1bd715405'/>
<id>urn:sha1:466dab407e47968777912aa23e5c66d1bd715405</id>
<content type='text'>
commit bd61baaf59669accae2720799394a51fecabe5d9 upstream.

Function import_ep() is incorrectly using ep-&gt;dst instead of the dst
ptr passed in.  This causes a crash when accepting new rdma connections
becase ep-&gt;dst is not initialized yet.

Signed-off-by: Steve Wise &lt;swise@opengridcomputing.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
Signed-off-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

</content>
</entry>
</feed>
