<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/drivers/infiniband/core/device.c, branch v3.11.4</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>IB/core: Handle table with full and partial membership for the same P_Key</title>
<updated>2012-10-01T03:33:29+00:00</updated>
<author>
<name>Jack Morgenstein</name>
<email>jackm@dev.mellanox.co.il</email>
</author>
<published>2012-08-03T08:40:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=ff7166c447df23a61e4f51bf748319dc6728dc74'/>
<id>ff7166c447df23a61e4f51bf748319dc6728dc74</id>
<content type='text'>
Extend the cached and non-cached P_Key table lookups to handle limited
and full membership of the same P_Key to co-exist in the P_Key table.

This is necessary for SR-IOV, to allow for some guests would to have
the full membership P_Key in their virtual P_Key table, while other
guests on the same physical HCA would have the limited one.
To support this, we need both the limited and full membership P_Keys
to be present in the master's (hypervisor physical port) P_Key table.

The algorithm for handling P_Key tables which contain both the limited
and the full membership versions of the same P_Key works as follows:

When scanning the P_Key table for a 15-bit P_Key:

A. If there is a full member version of that P_Key anywhere in the
    table, return its index (even if a limited-member version of the
    P_Key exists earlier in the table).

B. If the full member version is not in the table, but the
   limited-member version is in the table, return the index of the
   limited P_Key.

Signed-off-by: Liran Liss &lt;liranl@mellanox.com&gt;
Signed-off-by: Jack Morgenstein &lt;jackm@dev.mellanox.co.il&gt;
Signed-off-by: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Extend the cached and non-cached P_Key table lookups to handle limited
and full membership of the same P_Key to co-exist in the P_Key table.

This is necessary for SR-IOV, to allow for some guests would to have
the full membership P_Key in their virtual P_Key table, while other
guests on the same physical HCA would have the limited one.
To support this, we need both the limited and full membership P_Keys
to be present in the master's (hypervisor physical port) P_Key table.

The algorithm for handling P_Key tables which contain both the limited
and the full membership versions of the same P_Key works as follows:

When scanning the P_Key table for a 15-bit P_Key:

A. If there is a full member version of that P_Key anywhere in the
    table, return its index (even if a limited-member version of the
    P_Key exists earlier in the table).

B. If the full member version is not in the table, but the
   limited-member version is in the table, return the index of the
   limited P_Key.

Signed-off-by: Liran Liss &lt;liranl@mellanox.com&gt;
Signed-off-by: Jack Morgenstein &lt;jackm@dev.mellanox.co.il&gt;
Signed-off-by: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>RDMA: Allow for NULL .modify_device() and .modify_port() methods</title>
<updated>2011-07-18T23:44:30+00:00</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2011-06-18T16:35:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=10e1b54bbb51ac134c51996022f3bd10579fbd3b'/>
<id>10e1b54bbb51ac134c51996022f3bd10579fbd3b</id>
<content type='text'>
These methods don't make sense for iWARP devices, so rather than
forcing them to implement stubs, just return -ENOSYS in the core if
the hardware driver doesn't set .modify_device and/or .modify_port.

Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
These methods don't make sense for iWARP devices, so rather than
forcing them to implement stubs, just return -ENOSYS in the core if
the hardware driver doesn't set .modify_device and/or .modify_port.

Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>RDMA: Add netlink infrastructure</title>
<updated>2011-05-20T18:46:11+00:00</updated>
<author>
<name>Roland Dreier</name>
<email>roland@purestorage.com</email>
</author>
<published>2011-05-20T18:46:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=b2cbae2c248776d81cc265ff7d48405b6a4cc463'/>
<id>b2cbae2c248776d81cc265ff7d48405b6a4cc463</id>
<content type='text'>
Add basic RDMA netlink infrastructure that allows for registration of
RDMA clients for which data is to be exported and supplies message
construction callbacks.

Signed-off-by: Nir Muchtar &lt;nirm@voltaire.com&gt;

[ Reorganize a few things, add CONFIG_NET dependency.  - Roland ]

Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add basic RDMA netlink infrastructure that allows for registration of
RDMA clients for which data is to be exported and supplies message
construction callbacks.

Signed-off-by: Nir Muchtar &lt;nirm@voltaire.com&gt;

[ Reorganize a few things, add CONFIG_NET dependency.  - Roland ]

Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>RDMA: Add error handling to ib_core_init()</title>
<updated>2011-05-20T18:46:10+00:00</updated>
<author>
<name>Nir Muchtar</name>
<email>nirm@voltaire.com</email>
</author>
<published>2011-05-20T18:46:10+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=fd75c789abf7948e16fe50917a6acb809927719a'/>
<id>fd75c789abf7948e16fe50917a6acb809927719a</id>
<content type='text'>
Fail RDMA midlayer initialization if sysfs setup fails.

Signed-off-by: Nir Muchtar &lt;nirm@voltaire.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Fail RDMA midlayer initialization if sysfs setup fails.

Signed-off-by: Nir Muchtar &lt;nirm@voltaire.com&gt;
Signed-off-by: Roland Dreier &lt;roland@purestorage.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>RDMA: Update workqueue usage</title>
<updated>2011-01-17T05:16:31+00:00</updated>
<author>
<name>Tejun Heo</name>
<email>tj@kernel.org</email>
</author>
<published>2010-10-19T15:24:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f06267104dd9112f11586830d22501d0e26245ea'/>
<id>f06267104dd9112f11586830d22501d0e26245ea</id>
<content type='text'>
* ib_wq is added, which is used as the common workqueue for infiniband
  instead of the system workqueue.  All system workqueue usages
  including flush_scheduled_work() callers are converted to use and
  flush ib_wq.

* cancel_delayed_work() + flush_scheduled_work() converted to
  cancel_delayed_work_sync().

* qib_wq is removed and ib_wq is used instead.

This is to prepare for deprecation of flush_scheduled_work().

Signed-off-by: Tejun Heo &lt;tj@kernel.org&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
* ib_wq is added, which is used as the common workqueue for infiniband
  instead of the system workqueue.  All system workqueue usages
  including flush_scheduled_work() callers are converted to use and
  flush ib_wq.

* cancel_delayed_work() + flush_scheduled_work() converted to
  cancel_delayed_work_sync().

* qib_wq is removed and ib_wq is used instead.

This is to prepare for deprecation of flush_scheduled_work().

Signed-off-by: Tejun Heo &lt;tj@kernel.org&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>IB/core: Allow device-specific per-port sysfs files</title>
<updated>2010-05-21T17:34:44+00:00</updated>
<author>
<name>Ralph Campbell</name>
<email>ralph.campbell@qlogic.com</email>
</author>
<published>2010-05-07T00:03:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=9a6edb60ec10d86b1025a0cdad68fd89f1ddaf02'/>
<id>9a6edb60ec10d86b1025a0cdad68fd89f1ddaf02</id>
<content type='text'>
Add a new parameter to ib_register_device() so that low-level device
drivers can pass in a pointer to a callback function that will be
called for each port that is registered in sysfs.  This allows
low-level device drivers to create files in

    /sys/class/infiniband/&lt;hca&gt;/ports/&lt;N&gt;/

without having to poke through the internals of the RDMA sysfs handling.

There is no need for an unregister function since the kobject
reference will go to zero when ib_unregister_device() is called.

Signed-off-by: Ralph Campbell &lt;ralph.campbell@qlogic.com&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Add a new parameter to ib_register_device() so that low-level device
drivers can pass in a pointer to a callback function that will be
called for each port that is registered in sysfs.  This allows
low-level device drivers to create files in

    /sys/class/infiniband/&lt;hca&gt;/ports/&lt;N&gt;/

without having to poke through the internals of the RDMA sysfs handling.

There is no need for an unregister function since the kobject
reference will go to zero when ib_unregister_device() is called.

Signed-off-by: Ralph Campbell &lt;ralph.campbell@qlogic.com&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>IB: Remove sysfs files before unregistering device</title>
<updated>2009-02-25T21:27:46+00:00</updated>
<author>
<name>Roland Dreier</name>
<email>rolandd@cisco.com</email>
</author>
<published>2009-02-25T21:27:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=9206dff15705267c19f8fed391c4fb95975540a3'/>
<id>9206dff15705267c19f8fed391c4fb95975540a3</id>
<content type='text'>
Move the ib_device_unregister_sysfs() call from ib_dealloc_device() to
ib_unregister_device().  The old code allows device unregister to
proceed even if some sysfs files are open, which leaves a window where
userspace can open a file before a device is removed but then end up
reading the file after the device is removed, which leads to various
kernel crashes either because the device data structure is freed or
because the low-level driver code is gone after module removal.

By not returning from ib_unregister_device() until after all sysfs
entries are removed, we make sure that data structures and/or module
code is not freed until after all sysfs access is done.

Reported-by: Jack Morgenstein &lt;jackm@dev.mellanox.co.il&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Move the ib_device_unregister_sysfs() call from ib_dealloc_device() to
ib_unregister_device().  The old code allows device unregister to
proceed even if some sysfs files are open, which leaves a window where
userspace can open a file before a device is removed but then end up
reading the file after the device is removed, which leads to various
kernel crashes either because the device data structure is freed or
because the low-level driver code is gone after module removal.

By not returning from ib_unregister_device() until after all sysfs
entries are removed, we make sure that data structures and/or module
code is not freed until after all sysfs access is done.

Reported-by: Jack Morgenstein &lt;jackm@dev.mellanox.co.il&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>RDMA: Remove subversion $Id tags</title>
<updated>2008-07-15T06:48:44+00:00</updated>
<author>
<name>Roland Dreier</name>
<email>rolandd@cisco.com</email>
</author>
<published>2008-07-15T06:48:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f3781d2e89f12dd5afa046dc56032af6e39bd116'/>
<id>f3781d2e89f12dd5afa046dc56032af6e39bd116</id>
<content type='text'>
They don't get updated by git and so they're worse than useless.

Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
They don't get updated by git and so they're worse than useless.

Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>IB: find_first_zero_bit() takes unsigned pointer</title>
<updated>2007-10-10T02:59:04+00:00</updated>
<author>
<name>Roland Dreier</name>
<email>rolandd@cisco.com</email>
</author>
<published>2007-10-10T02:59:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=65d470b3ea52ee1402499d6fcb4632296452e5b1'/>
<id>65d470b3ea52ee1402499d6fcb4632296452e5b1</id>
<content type='text'>
Fix sparse warning

    drivers/infiniband/core/device.c:142:6: warning: incorrect type in argument 1 (different signedness)
    drivers/infiniband/core/device.c:142:6:    expected unsigned long const *addr
    drivers/infiniband/core/device.c:142:6:    got long *[assigned] inuse

by making the local variable inuse unsigned.  Does not affect generated
code at all.

Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Fix sparse warning

    drivers/infiniband/core/device.c:142:6: warning: incorrect type in argument 1 (different signedness)
    drivers/infiniband/core/device.c:142:6:    expected unsigned long const *addr
    drivers/infiniband/core/device.c:142:6:    got long *[assigned] inuse

by making the local variable inuse unsigned.  Does not affect generated
code at all.

Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>IB/core: Ignore membership bit in ib_find_pkey()</title>
<updated>2007-08-03T17:45:17+00:00</updated>
<author>
<name>Moni Shoua</name>
<email>monisonlists@gmail.com</email>
</author>
<published>2007-07-23T07:07:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=36026ecc20e5df722bbe2ea9e451c73d686ef107'/>
<id>36026ecc20e5df722bbe2ea9e451c73d686ef107</id>
<content type='text'>
ib_find_pkey() is used as a replacement for ib_find_cached_pkey(), and
the original function ignored the membership bit when searching for a
P_Key, so ib_find_pkey() should ignore the bit too.

In particular, IPoIB turns on the P_Key membership bit of limited
membership P_Keys when creating a child interface and looks for the
full membership P_key.  This broke if a port was a partial member of a
partition when IPoIB switched from ib_find_cached_pkey() to
ib_find_pkey(), and this change fixes things again.

Signed-off-by: Moni Shoua &lt;monis@voltaire.com&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
ib_find_pkey() is used as a replacement for ib_find_cached_pkey(), and
the original function ignored the membership bit when searching for a
P_Key, so ib_find_pkey() should ignore the bit too.

In particular, IPoIB turns on the P_Key membership bit of limited
membership P_Keys when creating a child interface and looks for the
full membership P_key.  This broke if a port was a partial member of a
partition when IPoIB switched from ib_find_cached_pkey() to
ib_find_pkey(), and this change fixes things again.

Signed-off-by: Moni Shoua &lt;monis@voltaire.com&gt;
Signed-off-by: Roland Dreier &lt;rolandd@cisco.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
