<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/drivers/vhost/scsi.c, branch v3.13</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending</title>
<updated>2013-11-22T18:52:03+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2013-11-22T18:52:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=b0e3636f656c98bdeded5aaa78601e3256b18d6d'/>
<id>b0e3636f656c98bdeded5aaa78601e3256b18d6d</id>
<content type='text'>
Pull SCSI target updates from Nicholas Bellinger:
 "Things have been quiet this round with mostly bugfixes, percpu
  conversions, and other minor iscsi-target conformance testing changes.

  The highlights include:

   - Add demo_mode_discovery attribute for iscsi-target (Thomas)
   - Convert tcm_fc(FCoE) to use percpu-ida pre-allocation
   - Add send completion interrupt coalescing for ib_isert
   - Convert target-core to use percpu-refcounting for se_lun
   - Fix mutex_trylock usage bug in iscsit_increment_maxcmdsn
   - tcm_loop updates (Hannes)
   - target-core ALUA cleanups + prep for v3.14 SCSI Referrals support (Hannes)

  v3.14 is currently shaping to be a busy development cycle in target
  land, with initial support for T10 Referrals and T10 DIF currently on
  the roadmap"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (40 commits)
  iscsi-target: chap auth shouldn't match username with trailing garbage
  iscsi-target: fix extract_param to handle buffer length corner case
  iscsi-target: Expose default_erl as TPG attribute
  target_core_configfs: split up ALUA supported states
  target_core_alua: Make supported states configurable
  target_core_alua: Store supported ALUA states
  target_core_alua: Rename ALUA_ACCESS_STATE_OPTIMIZED
  target_core_alua: spellcheck
  target core: rename (ex,im)plict -&gt; (ex,im)plicit
  percpu-refcount: Add percpu-refcount.o to obj-y
  iscsi-target: Do not reject non-immediate CmdSNs exceeding MaxCmdSN
  iscsi-target: Convert iscsi_session statistics to atomic_long_t
  target: Convert se_device statistics to atomic_long_t
  target: Fix delayed Task Aborted Status (TAS) handling bug
  iscsi-target: Reject unsupported multi PDU text command sequence
  ib_isert: Avoid duplicate iscsit_increment_maxcmdsn call
  iscsi-target: Fix mutex_trylock usage in iscsit_increment_maxcmdsn
  target: Core does not need blkdev.h
  target: Pass through I/O topology for block backstores
  iser-target: Avoid using FRMR for single dma entry requests
  ...
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pull SCSI target updates from Nicholas Bellinger:
 "Things have been quiet this round with mostly bugfixes, percpu
  conversions, and other minor iscsi-target conformance testing changes.

  The highlights include:

   - Add demo_mode_discovery attribute for iscsi-target (Thomas)
   - Convert tcm_fc(FCoE) to use percpu-ida pre-allocation
   - Add send completion interrupt coalescing for ib_isert
   - Convert target-core to use percpu-refcounting for se_lun
   - Fix mutex_trylock usage bug in iscsit_increment_maxcmdsn
   - tcm_loop updates (Hannes)
   - target-core ALUA cleanups + prep for v3.14 SCSI Referrals support (Hannes)

  v3.14 is currently shaping to be a busy development cycle in target
  land, with initial support for T10 Referrals and T10 DIF currently on
  the roadmap"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (40 commits)
  iscsi-target: chap auth shouldn't match username with trailing garbage
  iscsi-target: fix extract_param to handle buffer length corner case
  iscsi-target: Expose default_erl as TPG attribute
  target_core_configfs: split up ALUA supported states
  target_core_alua: Make supported states configurable
  target_core_alua: Store supported ALUA states
  target_core_alua: Rename ALUA_ACCESS_STATE_OPTIMIZED
  target_core_alua: spellcheck
  target core: rename (ex,im)plict -&gt; (ex,im)plicit
  percpu-refcount: Add percpu-refcount.o to obj-y
  iscsi-target: Do not reject non-immediate CmdSNs exceeding MaxCmdSN
  iscsi-target: Convert iscsi_session statistics to atomic_long_t
  target: Convert se_device statistics to atomic_long_t
  target: Fix delayed Task Aborted Status (TAS) handling bug
  iscsi-target: Reject unsupported multi PDU text command sequence
  ib_isert: Avoid duplicate iscsit_increment_maxcmdsn call
  iscsi-target: Fix mutex_trylock usage in iscsit_increment_maxcmdsn
  target: Core does not need blkdev.h
  target: Pass through I/O topology for block backstores
  iser-target: Avoid using FRMR for single dma entry requests
  ...
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost/scsi: Fix incorrect usage of get_user_pages_fast write parameter</title>
<updated>2013-10-25T18:03:34+00:00</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-10-25T17:44:15+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=60a01f558af9c48b0bb31f303c479e32721add3f'/>
<id>60a01f558af9c48b0bb31f303c479e32721add3f</id>
<content type='text'>
This patch addresses a long-standing bug where the get_user_pages_fast()
write parameter used for setting the underlying page table entry permission
bits was incorrectly set to write=1 for data_direction=DMA_TO_DEVICE, and
passed into get_user_pages_fast() via vhost_scsi_map_iov_to_sgl().

However, this parameter is intended to signal WRITEs to pinned userspace
PTEs for the virtio-scsi DMA_FROM_DEVICE -&gt; READ payload case, and *not*
for the virtio-scsi DMA_TO_DEVICE -&gt; WRITE payload case.

This bug would manifest itself as random process segmentation faults on
KVM host after repeated vhost starts + stops and/or with lots of vhost
endpoints + LUNs.

Cc: Stefan Hajnoczi &lt;stefanha@redhat.com&gt;
Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # 3.6+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch addresses a long-standing bug where the get_user_pages_fast()
write parameter used for setting the underlying page table entry permission
bits was incorrectly set to write=1 for data_direction=DMA_TO_DEVICE, and
passed into get_user_pages_fast() via vhost_scsi_map_iov_to_sgl().

However, this parameter is intended to signal WRITEs to pinned userspace
PTEs for the virtio-scsi DMA_FROM_DEVICE -&gt; READ payload case, and *not*
for the virtio-scsi DMA_TO_DEVICE -&gt; WRITE payload case.

This bug would manifest itself as random process segmentation faults on
KVM host after repeated vhost starts + stops and/or with lots of vhost
endpoints + LUNs.

Cc: Stefan Hajnoczi &lt;stefanha@redhat.com&gt;
Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; # 3.6+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>target: Remove TF_CIT_TMPL macro</title>
<updated>2013-10-16T20:35:02+00:00</updated>
<author>
<name>Andy Grover</name>
<email>agrover@redhat.com</email>
</author>
<published>2013-10-09T18:05:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=d80e224dd52dfc448e820aefa5f86c441ab76e1a'/>
<id>d80e224dd52dfc448e820aefa5f86c441ab76e1a</id>
<content type='text'>
Remove a lingering macro that just hid a dereference.

Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Andy Grover &lt;agrover@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Remove a lingering macro that just hid a dereference.

Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Andy Grover &lt;agrover@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost/scsi: Use GFP_ATOMIC with percpu_ida_alloc for obtaining tag</title>
<updated>2013-10-02T04:27:31+00:00</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-09-23T18:42:28+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=4a47d3a1ff10e564bf04f27ac14552df64f60cdf'/>
<id>4a47d3a1ff10e564bf04f27ac14552df64f60cdf</id>
<content type='text'>
Fix GFP_KERNEL -&gt; GFP_ATOMIC usage of percpu_ida_alloc() within
vhost_scsi_get_tag(), as this code is expected to be called directly
from interrupt context.

v2 changes:

  - Handle possible tag &lt; 0 failure with GFP_ATOMIC

Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Acked-by: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Fix GFP_KERNEL -&gt; GFP_ATOMIC usage of percpu_ida_alloc() within
vhost_scsi_get_tag(), as this code is expected to be called directly
from interrupt context.

v2 changes:

  - Handle possible tag &lt; 0 failure with GFP_ATOMIC

Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Acked-by: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost-scsi: whitespace tweak</title>
<updated>2013-09-17T19:56:09+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2013-09-17T19:54:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=d3d665a654a35c47463d2aa6353bac3ce293f4f5'/>
<id>d3d665a654a35c47463d2aa6353bac3ce293f4f5</id>
<content type='text'>
Remove space at start of line that sneaked in.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Remove space at start of line that sneaked in.

Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost/scsi: use vmalloc for order-10 allocation</title>
<updated>2013-09-17T19:55:46+00:00</updated>
<author>
<name>Michael S. Tsirkin</name>
<email>mst@redhat.com</email>
</author>
<published>2013-09-17T06:30:34+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=595cb754983d4387cb25b3dcf08f5129663d634e'/>
<id>595cb754983d4387cb25b3dcf08f5129663d634e</id>
<content type='text'>
As vhost scsi device struct is large, if the device is
created on a busy system, kzalloc() might fail, so this patch does a
fallback to vzalloc().

As vmalloc() adds overhead on data-path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.

Reviewed-by: Asias He &lt;asias@redhat.com&gt;
Reported-by: Dan Aloni &lt;alonid@stratoscale.com&gt;
Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
As vhost scsi device struct is large, if the device is
created on a busy system, kzalloc() might fail, so this patch does a
fallback to vzalloc().

As vmalloc() adds overhead on data-path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.

Reviewed-by: Asias He &lt;asias@redhat.com&gt;
Reported-by: Dan Aloni &lt;alonid@stratoscale.com&gt;
Signed-off-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>target: Update copyright ownership/year information to 2013</title>
<updated>2013-09-11T03:23:36+00:00</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-09-05T22:29:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=4c76251e8e8f704ed7eace603a8041458a1ed16f'/>
<id>4c76251e8e8f704ed7eace603a8041458a1ed16f</id>
<content type='text'>
Update copyright ownership/year information for target-core,
loopback, iscsi-target, tcm_qla2xx, vhost and iser-target.

Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Update copyright ownership/year information for target-core,
loopback, iscsi-target, tcm_qla2xx, vhost and iser-target.

Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost/scsi: Add pre-allocation for tv_cmd SGL + upages memory</title>
<updated>2013-09-09T21:29:19+00:00</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-06-21T21:32:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=3aee26b4ae91048c933dc622f00b7bb01f7f0ff1'/>
<id>3aee26b4ae91048c933dc622f00b7bb01f7f0ff1</id>
<content type='text'>
This patch adds support for pre-allocation of per tv_cmd descriptor
scatterlist + user-space page pointer memory using se_sess-&gt;sess_cmd_map
within tcm_vhost_make_nexus() code.

This includes sanity checks within vhost_scsi_map_to_sgl()
to reject I/O that exceeds these initial hardcoded values, and
the necessary cleanup in tcm_vhost_make_nexus() failure path +
tcm_vhost_drop_nexus().

v3 changes:
  - Rebase to v3.11-rc5 code

Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Reviewed-by: Asias He &lt;asias@redhat.com&gt;
Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch adds support for pre-allocation of per tv_cmd descriptor
scatterlist + user-space page pointer memory using se_sess-&gt;sess_cmd_map
within tcm_vhost_make_nexus() code.

This includes sanity checks within vhost_scsi_map_to_sgl()
to reject I/O that exceeds these initial hardcoded values, and
the necessary cleanup in tcm_vhost_make_nexus() failure path +
tcm_vhost_drop_nexus().

v3 changes:
  - Rebase to v3.11-rc5 code

Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Reviewed-by: Asias He &lt;asias@redhat.com&gt;
Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>vhost/scsi: Convert to per-cpu ida_alloc + ida_free command map</title>
<updated>2013-09-09T21:29:18+00:00</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-06-08T00:47:46+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=4824d3bfb9097ac59cf7633c0bc4fb7c5b549a80'/>
<id>4824d3bfb9097ac59cf7633c0bc4fb7c5b549a80</id>
<content type='text'>
This patch changes vhost/scsi to use transport_init_session_tags()
pre-allocation logic for per-cpu session tag pooling with internal
ida_alloc() + ida_free() calls based upon the saved se_cmd-&gt;map_tag id.

FIXME: Make transport_init_session_tags() number of tags setup
configurable per vring client setting via configfs

v5 changes:
 - Convert to percpu_ida.h include

v3 changes:
 - Update to percpu-ida usage
 - Rebase to v3.11-rc5 code

Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch changes vhost/scsi to use transport_init_session_tags()
pre-allocation logic for per-cpu session tag pooling with internal
ida_alloc() + ida_free() calls based upon the saved se_cmd-&gt;map_tag id.

FIXME: Make transport_init_session_tags() number of tags setup
configurable per vring client setting via configfs

v5 changes:
 - Convert to percpu_ida.h include

v3 changes:
 - Update to percpu-ida usage
 - Rebase to v3.11-rc5 code

Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Asias He &lt;asias@redhat.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Acked-by: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost</title>
<updated>2013-07-23T21:38:20+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2013-07-23T21:38:20+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=a030cbc35091630261194745389cace7706abafa'/>
<id>a030cbc35091630261194745389cace7706abafa</id>
<content type='text'>
Pull vhost fixes from Michael Tsirkin:
 "vhost: more fixes for 3.11

  This includes some fixes for vhost net and scsi drivers.

  The test module has already been reworked to avoid rcu usage, but the
  necessary core changes are missing, we fixed this.

  Unlikely to affect any real-world users, but it's early in the cycle
  so, let's merge them"

(It was earlier when Michael originally sent the email, but it somehot
got missed in the flood, so here it is after -rc2)

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  vhost: Remove custom vhost rcu usage
  vhost-scsi: Always access vq-&gt;private_data under vq mutex
  vhost-net: Always access vq-&gt;private_data under vq mutex
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pull vhost fixes from Michael Tsirkin:
 "vhost: more fixes for 3.11

  This includes some fixes for vhost net and scsi drivers.

  The test module has already been reworked to avoid rcu usage, but the
  necessary core changes are missing, we fixed this.

  Unlikely to affect any real-world users, but it's early in the cycle
  so, let's merge them"

(It was earlier when Michael originally sent the email, but it somehot
got missed in the flood, so here it is after -rc2)

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  vhost: Remove custom vhost rcu usage
  vhost-scsi: Always access vq-&gt;private_data under vq mutex
  vhost-net: Always access vq-&gt;private_data under vq mutex
</pre>
</div>
</content>
</entry>
</feed>
