<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/net/tipc, branch v3.11-rc3</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/net/tipc?h=v3.11-rc3</id>
<link rel='self' href='https://git.amat.us/linux/atom/net/tipc?h=v3.11-rc3'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2013-07-12T00:03:36Z</updated>
<entry>
<title>net/tipc: use %*phC to dump small buffers in hex form</title>
<updated>2013-07-12T00:03:36Z</updated>
<author>
<name>Andy Shevchenko</name>
<email>andriy.shevchenko@linux.intel.com</email>
</author>
<published>2013-07-10T14:30:34Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d77e41e12744e53ca7f98f920350998b5f00c93a'/>
<id>urn:sha1:d77e41e12744e53ca7f98f920350998b5f00c93a</id>
<content type='text'>
Instead of passing each byte by stack let's use nice specifier for that.

Signed-off-by: Andy Shevchenko &lt;andriy.shevchenko@linux.intel.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: remove dev_base_lock use from enable_bearer</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:51Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=2537af9dcabbdd6c93c041a955d3a9ae42c0c008'/>
<id>urn:sha1:2537af9dcabbdd6c93c041a955d3a9ae42c0c008</id>
<content type='text'>
Convert enable_bearer() to RCU locking with dev_get_by_name().

Based on a similar changeset in commit 840a185d ["aoe: remove
dev_base_lock use from aoecmd_cfg_pkts()"] -- quoting that:

  "dev_base_lock is the legacy way to lock the device list,
   and is planned to disappear. (writers hold RTNL, readers
   hold RCU lock)"

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: fix wrong return value for link_send_sections_long routine</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:50Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=126c0524648631a0f6fba4d016586b236209fe6f'/>
<id>urn:sha1:126c0524648631a0f6fba4d016586b236209fe6f</id>
<content type='text'>
When skb buffer cannot be allocated in link_send_sections_long(),
-ENOMEM error code instead of -EFAULT should be returned to its
caller.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: make tipc_link_send_sections_fast exit earlier</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:49Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=7410f967ba9bdc14b1e336e5d235929ed878cbfc'/>
<id>urn:sha1:7410f967ba9bdc14b1e336e5d235929ed878cbfc</id>
<content type='text'>
Once message build request function returns invalid code, the
process of sending message cannot continue. So in case of message
build failure, tipc_link_send_sections_fast() should return
immediately.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: enhance priority of link protocol packet</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:48Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=796c75d0d3ef13cd1df00779abb8b27edb630504'/>
<id>urn:sha1:796c75d0d3ef13cd1df00779abb8b27edb630504</id>
<content type='text'>
pfifo_fast is set as default traffic class queueing discipline. This
queue has three so called "bands". Within each band, FIFO rules apply.
However, as long as there are packets waiting in band 0, band 1 won't
be processed.

Now all kind of TIPC type packet priorities are never set, that is,
their priorities are 0, so they are mapped to band 1 of pfifo_fast
qdisc. But, especially during link congestion, if link protocol packet
can be sent out as earlier as possible than other type of packets so
that protocol packet can arrive at peer endpoint in time, the peer
will timely reset its link timeout timer to keep the link alive.
So enhancing the priority of link protocol packets can meet the
specific demand to avoid unnecessary link reset due to a transient
link congestion.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: cosmetic realignment of function arguments</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Paul Gortmaker</name>
<email>paul.gortmaker@windriver.com</email>
</author>
<published>2013-06-17T14:54:47Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=ae8509c420122866344bde1241e31858d0aa2fbc'/>
<id>urn:sha1:ae8509c420122866344bde1241e31858d0aa2fbc</id>
<content type='text'>
No runtime code changes here.  Just a realign of the function
arguments to start where the 1st one was, and fit as many args
as can be put in an 80 char line.

Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: save sock structure pointer instead of void pointer to tipc_port</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:46Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=c0fee8aca7206264d5e3dcc4e60aaf86501f4ea1'/>
<id>urn:sha1:c0fee8aca7206264d5e3dcc4e60aaf86501f4ea1</id>
<content type='text'>
Directly save sock structure pointer instead of void pointer to avoid
unnecessary cast conversions.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Jon Maloy &lt;jon.maloy@ericsson.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: convert config_lock from spinlock to mutex</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:45Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=28e5297281ab85d636aa814a9b65cfb99375d092'/>
<id>urn:sha1:28e5297281ab85d636aa814a9b65cfb99375d092</id>
<content type='text'>
As the configuration server is now running under process context,
it's unnecessary for us to have a spinlock serializing the TIPC
configuration process. Instead, we replace it with a mutex lock,
which gives us more freedom. For instance, we can now call
pre-emptable functions within the protected area.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Jon Maloy &lt;jon.maloy@ericsson.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: rename tipc_createport_raw to tipc_createport</title>
<updated>2013-06-17T22:53:01Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:44Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=3c5db8e4eca36e4f312b49bba99f4c1f6ce0563a'/>
<id>urn:sha1:3c5db8e4eca36e4f312b49bba99f4c1f6ce0563a</id>
<content type='text'>
After the removal of the native API, there is now only one way to
to create a TIPC port instance -- the function tipc_createport_raw().
We make it more readable by renaming it to tipc_createport().

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Jon Maloy &lt;jon.maloy@ericsson.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>tipc: remove user_port instance from tipc_port structure</title>
<updated>2013-06-17T22:53:00Z</updated>
<author>
<name>Ying Xue</name>
<email>ying.xue@windriver.com</email>
</author>
<published>2013-06-17T14:54:43Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=f1733d7580ff94deb8ea071a293c23939ae0d450'/>
<id>urn:sha1:f1733d7580ff94deb8ea071a293c23939ae0d450</id>
<content type='text'>
After the native API has been completely removed, the 'user_port'
field in struct tipc_port becomes unused, and can be removed.
As a consequence, the "usrmem" argument in tipc_msg_build() is no
longer needed, and so we remove that one too.

Signed-off-by: Ying Xue &lt;ying.xue@windriver.com&gt;
Signed-off-by: Jon Maloy &lt;jon.maloy@ericsson.com&gt;
Signed-off-by: Paul Gortmaker &lt;paul.gortmaker@windriver.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
</feed>
