<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel.git/drivers/infiniband/hw, branch linux-3.12.y</title>
<subtitle>Hosts the 0x221E linux distro kernel.
</subtitle>
<id>https://git.0xinfinity.dev/distro/kernel.git/atom?h=linux-3.12.y</id>
<link rel='self' href='https://git.0xinfinity.dev/distro/kernel.git/atom?h=linux-3.12.y'/>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/'/>
<updated>2017-04-07T07:17:56Z</updated>
<entry>
<title>net/mlx5: Increase number of max QPs in default profile</title>
<updated>2017-04-07T07:17:56Z</updated>
<author>
<name>Maor Gottlieb</name>
</author>
<published>2017-03-21T13:59:17Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=7d409c047a3b7a40dbd0ac8b26ec29ef62a088c8'/>
<id>urn:sha1:7d409c047a3b7a40dbd0ac8b26ec29ef62a088c8</id>
<content type='text'>
[ Upstream commit 5f40b4ed975c26016cf41953b7510fe90718e21c ]

With ConnectX-4 sharing SRQs from the same space as QPs, we hit a
limit preventing some applications to allocate needed QPs amount.
Double the size to 256K.

[js] this is in another file in 3.12

Fixes: e126ba97dba9e ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Maor Gottlieb &lt;maorg@mellanox.com&gt;
Signed-off-by: Saeed Mahameed &lt;saeedm@mellanox.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>IB/mlx4: Fix port query for 56Gb Ethernet links</title>
<updated>2017-01-26T16:40:57Z</updated>
<author>
<name>Saeed Mahameed</name>
</author>
<published>2016-11-10T09:30:59Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=b1a317d4dc67b758131638649565be133d8de0c5'/>
<id>urn:sha1:b1a317d4dc67b758131638649565be133d8de0c5</id>
<content type='text'>
commit 6fa26208206c406fa529cd73f7ae6bf4181e270b upstream.

Report the correct speed in the port attributes when using a 56Gbps
ethernet link.  Without this change the field is incorrectly set to 10.

Fixes: a9c766bb75ee ('IB/mlx4: Fix info returned when querying IBoE ports')
Fixes: 2e96691c31ec ('IB: Use central enum for speed instead of hard-coded values')
Signed-off-by: Saeed Mahameed &lt;saeedm@mellanox.com&gt;
Signed-off-by: Yishai Hadas &lt;yishaih@mellanox.com&gt;
Signed-off-by: Daniel Jurgens &lt;danielj@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx4: Set traffic class in AH</title>
<updated>2017-01-26T16:40:57Z</updated>
<author>
<name>Maor Gottlieb</name>
</author>
<published>2016-11-10T09:30:53Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=d1e2a609ac560496e87cef720b6e08b55cd097aa'/>
<id>urn:sha1:d1e2a609ac560496e87cef720b6e08b55cd097aa</id>
<content type='text'>
commit af4295c117b82a521b05d0daf39ce879d26e6cb1 upstream.

Set traffic class within sl_tclass_flowlabel when create iboe AH.
Without this the TOS value will be empty when running VLAN tagged
traffic, because the TOS value is taken from the traffic class in the
address handle attributes.

Fixes: 9106c4106974 ('IB/mlx4: Fix SL to 802.1Q priority-bits mapping for IBoE')
Signed-off-by: Maor Gottlieb &lt;maorg@mellanox.com&gt;
Signed-off-by: Daniel Jurgens &lt;danielj@mellanox.com&gt;
Reviewed-by: Mark Bloch &lt;markb@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx5: Fix fatal error dispatching</title>
<updated>2016-11-24T15:23:58Z</updated>
<author>
<name>Eli Cohen</name>
</author>
<published>2016-10-27T13:36:44Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=a6f56e55d4e94024ea8275fe50f54d667f100a2f'/>
<id>urn:sha1:a6f56e55d4e94024ea8275fe50f54d667f100a2f</id>
<content type='text'>
commit dbaaff2a2caa03d472b5cc53a3fbfd415c97dc26 upstream.

When an internal error condition is detected, make sure to set the
device inactive after dispatching the event so ULPs can get a
notification of this event.

Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Eli Cohen &lt;eli@mellanox.com&gt;
Signed-off-by: Maor Gottlieb &lt;maorg@mellanox.com&gt;
Reviewed-by: Mohamad Haj Yahia &lt;mohamad@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx5: Use cache line size to select CQE stride</title>
<updated>2016-11-24T15:23:58Z</updated>
<author>
<name>Daniel Jurgens</name>
</author>
<published>2016-10-27T13:36:41Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=577f7b9e99d1a86284a02792e038511293e56710'/>
<id>urn:sha1:577f7b9e99d1a86284a02792e038511293e56710</id>
<content type='text'>
commit 16b0e0695a73b68d8ca40288c8f9614ef208917b upstream.

When creating kernel CQs use 128B CQE stride if the
cache line size is 128B, 64B otherwise.  This prevents
multiple CQEs from residing in a 128B cache line,
which can cause retries when there are concurrent
read and writes in one cache line.

Tested with IPoIB on PPC64, saw ~5% throughput
improvement.

Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Daniel Jurgens &lt;danielj@mellanox.com&gt;
Signed-off-by: Maor Gottlieb &lt;maorg@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx4: Fix create CQ error flow</title>
<updated>2016-11-24T15:23:57Z</updated>
<author>
<name>Matan Barak</name>
</author>
<published>2016-11-10T09:30:55Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=3ee9c742bb021ab292e97f1e26bf2a339b20310c'/>
<id>urn:sha1:3ee9c742bb021ab292e97f1e26bf2a339b20310c</id>
<content type='text'>
commit 593ff73bcfdc79f79a8a0df55504f75ad3e5d1a9 upstream.

Currently, if ib_copy_to_udata fails, the CQ
won't be deleted from the radix tree and the HW (HW2SW).

Fixes: 225c7b1feef1 ('IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters')
Signed-off-by: Matan Barak &lt;matanb@mellanox.com&gt;
Signed-off-by: Daniel Jurgens &lt;danielj@mellanox.com&gt;
Reviewed-by: Mark Bloch &lt;markb@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx4: Fix incorrect MC join state bit-masking on SR-IOV</title>
<updated>2016-10-07T08:06:57Z</updated>
<author>
<name>Alex Vesker</name>
</author>
<published>2016-09-12T16:16:18Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=3a6d50fe6a54175cadd8e0f3881cfff3ef5ca82c'/>
<id>urn:sha1:3a6d50fe6a54175cadd8e0f3881cfff3ef5ca82c</id>
<content type='text'>
commit e5ac40cd66c2f3cd11bc5edc658f012661b16347 upstream.

Because of an incorrect bit-masking done on the join state bits, when
handling a join request we failed to detect a difference between the
group join state and the request join state when joining as send only
full member (0x8). This caused the MC join request not to be sent.
This issue is relevant only when SRIOV is enabled and SM supports
send only full member.

This fix separates scope bits and join states bits a nibble each.

Fixes: b9c5d6a64358 ('IB/mlx4: Add multicast group (MCG) paravirtualization for SR-IOV')
Signed-off-by: Alex Vesker &lt;valex@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx4: Fix the SQ size of an RC QP</title>
<updated>2016-08-19T07:51:23Z</updated>
<author>
<name>Yishai Hadas</name>
</author>
<published>2016-06-22T14:27:28Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=5f858e6227cb9d4e41bca39a8bef70f346521ab4'/>
<id>urn:sha1:5f858e6227cb9d4e41bca39a8bef70f346521ab4</id>
<content type='text'>
commit f2940e2c76bb554a7fbdd28ca5b90904117a9e96 upstream.

When calculating the required size of an RC QP send queue, leave
enough space for masked atomic operations, which require more space than
"regular" atomic operation.

Fixes: 6fa8f719844b ("IB/mlx4: Add support for masked atomic operations")
Signed-off-by: Yishai Hadas &lt;yishaih@mellanox.com&gt;
Reviewed-by: Jack Morgenstein &lt;jackm@mellanox.co.il&gt;
Reviewed-by: Eran Ben Elisha &lt;eranbe@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx5: Fix post send fence logic</title>
<updated>2016-08-19T07:51:22Z</updated>
<author>
<name>Eli Cohen</name>
</author>
<published>2016-06-22T14:27:26Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=871ac53ff20207ce73ce23ee8afe1fe4bac7aaa0'/>
<id>urn:sha1:871ac53ff20207ce73ce23ee8afe1fe4bac7aaa0</id>
<content type='text'>
commit c9b254955b9f8814966f5dabd34c39d0e0a2b437 upstream.

If the caller specified IB_SEND_FENCE in the send flags of the work
request and no previous work request stated that the successive one
should be fenced, the work request would be executed without a fence.
This could result in RDMA read or atomic operations failure due to a MR
being invalidated. Fix this by adding the mlx5 enumeration for fencing
RDMA/atomic operations and fix the logic to apply this.

Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Eli Cohen &lt;eli@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

</content>
</entry>
<entry>
<title>IB/mlx5: Fix returned values of query QP</title>
<updated>2016-08-19T07:51:21Z</updated>
<author>
<name>Noa Osherovich</name>
</author>
<published>2016-06-04T12:15:32Z</published>
<link rel='alternate' type='text/html' href='https://git.0xinfinity.dev/distro/kernel.git/commit/?id=d70382dd16741064205c869e3c9f628f8110d6ca'/>
<id>urn:sha1:d70382dd16741064205c869e3c9f628f8110d6ca</id>
<content type='text'>
commit 0540d8148d419bf769e5aa99c77027febd8922f0 upstream.

Some variables were not initialized properly: max_recv_wr,
max_recv_sge, max_send_wr, qp_context and max_inline_data.

Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB...')
Signed-off-by: Noa Osherovich &lt;noaos@mellanox.com&gt;
Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;
Reviewed-by: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Signed-off-by: Doug Ledford &lt;dledford@redhat.com&gt;
Signed-off-by: Jiri Slaby &lt;jslaby@suse.cz&gt;

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