<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux/include/target/iscsi, branch v3.16</title>
<subtitle>Linux kernel source tree</subtitle>
<id>https://git.amat.us/linux/atom/include/target/iscsi?h=v3.16</id>
<link rel='self' href='https://git.amat.us/linux/atom/include/target/iscsi?h=v3.16'/>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/'/>
<updated>2014-06-11T18:52:39Z</updated>
<entry>
<title>Target/iscsi: Fix sendtargets response pdu for iser transport</title>
<updated>2014-06-11T18:52:39Z</updated>
<author>
<name>Sagi Grimberg</name>
<email>sagig@mellanox.com</email>
</author>
<published>2014-06-10T15:27:59Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=22c7aaa57e80853b4904a46c18f97db0036a3b97'/>
<id>urn:sha1:22c7aaa57e80853b4904a46c18f97db0036a3b97</id>
<content type='text'>
In case the transport is iser we should not include the
iscsi target info in the sendtargets text response pdu.
This causes sendtargets response to include the target
info twice.

Modify iscsit_build_sendtargets_response to filter
transport types that don't match.

Signed-off-by: Sagi Grimberg &lt;sagig@mellanox.com&gt;
Reported-by: Slava Shwartsman &lt;valyushash@gmail.com&gt;
Cc: stable@vger.kernel.org # 3.11+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>target: Pass in transport supported PI at session initialization</title>
<updated>2014-04-07T08:48:54Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2014-04-02T19:52:38Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=e70beee783d6977d80eede88a3394f02eabddad1'/>
<id>urn:sha1:e70beee783d6977d80eede88a3394f02eabddad1</id>
<content type='text'>
In order to support local WRITE_INSERT + READ_STRIP operations for
non PI enabled fabrics, the fabric driver needs to be able signal
what protection offload operations are supported.

This is done at session initialization time so the modes can be
signaled by individual se_wwn + se_portal_group endpoints, as well
as optionally across different transports on the same endpoint.

For iser-target, set TARGET_PROT_ALL if the underlying ib_device
has already signaled PI offload support, and allow this to be
exposed via a new iscsit_transport-&gt;iscsit_get_sup_prot_ops()
callback.

For loopback, set TARGET_PROT_ALL to signal SCSI initiator mode
operation.

For all other drivers, set TARGET_PROT_NORMAL to disable fabric
level PI.

Cc: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
Cc: Sagi Grimberg &lt;sagig@mellanox.com&gt;
Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Quinn Tran &lt;quinn.tran@qlogic.com&gt;
Cc: Giridhar Malavali &lt;giridhar.malavali@qlogic.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>target: Add TFO-&gt;abort_task for aborted task resources release</title>
<updated>2014-04-07T08:48:51Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2014-03-22T21:55:56Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=131e6abc674edb9f9a59090bb35bf6650569b7e7'/>
<id>urn:sha1:131e6abc674edb9f9a59090bb35bf6650569b7e7</id>
<content type='text'>
Now that TASK_ABORTED status is not generated for all cases by
TMR ABORT_TASK + LUN_RESET, a new TFO-&gt;abort_task() caller is
necessary in order to give fabric drivers a chance to unmap
hardware / software resources before the se_cmd descriptor is
released via the normal TFO-&gt;release_cmd() codepath.

This patch adds TFO-&gt;aborted_task() in core_tmr_abort_task()
in place of the original transport_send_task_abort(), and
also updates all fabric drivers to implement this caller.

The fabric drivers that include changes to perform cleanup
via -&gt;aborted_task() are:

  - iscsi-target
  - iser-target
  - srpt
  - tcm_qla2xxx

The fabric drivers that currently set -&gt;aborted_task() to
NOPs are:

  - loopback
  - tcm_fc
  - usb-gadget
  - sbp-target
  - vhost-scsi

For the latter five, there appears to be no additional cleanup
required before invoking TFO-&gt;release_cmd() to release the
se_cmd descriptor.

v2 changes:
  - Move -&gt;aborted_task() call into transport_cmd_finish_abort (Alex)

Cc: Alex Leung &lt;amleung21@yahoo.com&gt;
Cc: Mark Rustad &lt;mark.d.rustad@intel.com&gt;
Cc: Roland Dreier &lt;roland@kernel.org&gt;
Cc: Vu Pham &lt;vu@mellanox.com&gt;
Cc: Chris Boot &lt;bootc@bootc.net&gt;
Cc: Sebastian Andrzej Siewior &lt;bigeasy@linutronix.de&gt;
Cc: Michael S. Tsirkin &lt;mst@redhat.com&gt;
Cc: Giridhar Malavali &lt;giridhar.malavali@qlogic.com&gt;
Cc: Saurav Kashyap &lt;saurav.kashyap@qlogic.com&gt;
Cc: Quinn Tran &lt;quinn.tran@qlogic.com&gt;
Cc: Sagi Grimberg &lt;sagig@mellanox.com&gt;
Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi/iser-target: Fix isert_conn-&gt;state hung shutdown issues</title>
<updated>2014-03-05T01:54:09Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2014-02-03T20:54:39Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=defd884845297fd5690594bfe89656b01f16d87e'/>
<id>urn:sha1:defd884845297fd5690594bfe89656b01f16d87e</id>
<content type='text'>
This patch addresses a couple of different hug shutdown issues
related to wait_event() + isert_conn-&gt;state.  First, it changes
isert_conn-&gt;conn_wait + isert_conn-&gt;conn_wait_comp_err from
waitqueues to completions, and sets ISER_CONN_TERMINATING from
within isert_disconnect_work().

Second, it splits isert_free_conn() into isert_wait_conn() that
is called earlier in iscsit_close_connection() to ensure that
all outstanding commands have completed before continuing.

Finally, it breaks isert_cq_comp_err() into seperate TX / RX
related code, and adds logic in isert_cq_rx_comp_err() to wait
for outstanding commands to complete before setting ISER_CONN_DOWN
and calling complete(&amp;isert_conn-&gt;conn_wait_comp_err).

Acked-by: Sagi Grimberg &lt;sagig@mellanox.com&gt;
Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: &lt;stable@vger.kernel.org&gt; #3.10+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi-target: Convert gfp_t parameter to task state bitmask</title>
<updated>2014-01-25T06:58:52Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2014-01-20T03:36:44Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=676687c69697d2081d25afd14ee90937d1fb0c8e'/>
<id>urn:sha1:676687c69697d2081d25afd14ee90937d1fb0c8e</id>
<content type='text'>
This patch propigates the use of task state bitmask now used by
percpu_ida_alloc() up the iscsi-target callchain, replacing the
use of GFP_ATOMIC for TASK_RUNNING, and GFP_KERNEL for
TASK_INTERRUPTIBLE.

Also, drop the unnecessary gfp_t parameter to isert_allocate_cmd(),
and just pass TASK_INTERRUPTIBLE into iscsit_allocate_cmd().

Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi/iser-target: Convert to command priv_size usage</title>
<updated>2013-09-09T21:29:21Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@daterainc.com</email>
</author>
<published>2013-08-17T21:27:56Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=d703ce2f7f4de20c03d71c22a9d5e3708798047b'/>
<id>urn:sha1:d703ce2f7f4de20c03d71c22a9d5e3708798047b</id>
<content type='text'>
This command converts iscsi/isert-target to use allocations based on
iscsit_transport-&gt;priv_size within iscsit_allocate_cmd(), instead of
using an embedded isert_cmd-&gt;iscsi_cmd.

This includes removing iscsit_transport-&gt;alloc_cmd() usage, along
with updating isert-target code to use iscsit_priv_cmd().

Also, remove left-over iscsit_transport-&gt;release_cmd() usage for
direct calls to iscsit_release_cmd(), and drop the now unused
lio_cmd_cache and isert_cmd_cache.

Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Kent Overstreet &lt;kmo@daterainc.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@daterainc.com&gt;
</content>
</entry>
<entry>
<title>iser-target: Fix session reset bug with RDMA_CM_EVENT_DISCONNECTED</title>
<updated>2013-07-08T01:35:56Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-07-03T10:05:37Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=b2cb96494d83b894a43ba8b9023eead8ff50684b'/>
<id>urn:sha1:b2cb96494d83b894a43ba8b9023eead8ff50684b</id>
<content type='text'>
This patch addresses a bug where RDMA_CM_EVENT_DISCONNECTED may occur
before the connection shutdown has been completed by rx/tx threads,
that causes isert_free_conn() to wait indefinately on -&gt;conn_wait.

This patch allows isert_disconnect_work code to invoke rdma_disconnect
when isert_disconnect_work() process context is started by client
session reset before isert_free_conn() code has been reached.

It also adds isert_conn-&gt;conn_mutex protection for -&gt;state within
isert_disconnect_work(), isert_cq_comp_err() and isert_free_conn()
code, along with isert_check_state() for wait_event usage.

(v2: Add explicit iscsit_cause_connection_reinstatement call
     during isert_disconnect_work() to force conn reset)

Cc: stable@vger.kernel.org  # 3.10+
Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi-target: Fix iscsit_sequence_cmd reject handling for iser</title>
<updated>2013-07-07T04:59:54Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-07-03T10:58:58Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=561bf15892375597ee59d473a704a3e634c4f311'/>
<id>urn:sha1:561bf15892375597ee59d473a704a3e634c4f311</id>
<content type='text'>
This patch moves ISCSI_OP_REJECT failures into iscsit_sequence_cmd()
in order to avoid external iscsit_reject_cmd() reject usage for all
PDU types.

It also updates PDU specific handlers for traditional iscsi-target
code to not reset the session after posting a ISCSI_OP_REJECT during
setup.

(v2: Fix CMDSN_LOWER_THAN_EXP for ISCSI_OP_SCSI to call
     target_put_sess_cmd() after iscsit_sequence_cmd() failure)

Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Mike Christie &lt;michaelc@cs.wisc.edu&gt;
Cc: stable@vger.kernel.org  # 3.10+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi-target: Fix iscsit_add_reject* usage for iser</title>
<updated>2013-07-07T04:59:53Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-07-03T10:48:24Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=ba159914086f06532079fc15141f46ffe7e04a41'/>
<id>urn:sha1:ba159914086f06532079fc15141f46ffe7e04a41</id>
<content type='text'>
This patch changes iscsit_add_reject() + iscsit_add_reject_from_cmd()
usage to not sleep on iscsi_cmd-&gt;reject_comp to address a free-after-use
usage bug in v3.10 with iser-target code.

It saves -&gt;reject_reason for use within iscsit_build_reject() so the
correct value for both transport cases.  It also drops the legacy
fail_conn parameter usage throughput iscsi-target code and adds
two iscsit_add_reject_cmd() and iscsit_reject_cmd helper functions,
along with various small cleanups.

(v2: Re-enable target_put_sess_cmd() to be called from
     iscsit_add_reject_from_cmd() for rejects invoked after
     target_get_sess_cmd() has been called)

Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Mike Christie &lt;michaelc@cs.wisc.edu&gt;
Cc: stable@vger.kernel.org  # 3.10+
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
<entry>
<title>iscsi-target: Refactor ISCSI_OP_TEXT_RSP TX handling</title>
<updated>2013-06-26T07:33:32Z</updated>
<author>
<name>Nicholas Bellinger</name>
<email>nab@linux-iscsi.org</email>
</author>
<published>2013-06-15T01:49:55Z</published>
<link rel='alternate' type='text/html' href='https://git.amat.us/linux/commit/?id=889c8a68b8483a8b3482ac440af3ad7368c58555'/>
<id>urn:sha1:889c8a68b8483a8b3482ac440af3ad7368c58555</id>
<content type='text'>
This patch refactoring existing iscsit_send_text_rsp() in order
to handle iscsi_text_rsp payloads in a transport specific manner.

This includes the addition of iscsit_build_text_rsp() to build
the response payload and initialize ISCSI_OP_TEXT_RSP.

v2: Make iscsit_build_text_rsp() determine extra padding bytes, and
    drop legacy padding calculation for traditional iSCSI text
    responses within iscsit_send_text_rsp()

Reported-by: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Or Gerlitz &lt;ogerlitz@mellanox.com&gt;
Cc: Mike Christie &lt;michaelc@cs.wisc.edu&gt;
Signed-off-by: Nicholas Bellinger &lt;nab@linux-iscsi.org&gt;
</content>
</entry>
</feed>
