There are three timing problems in the kthread usages of iscsi_target_mod:

 - np_thread of struct iscsi_np
 - rx_thread and tx_thread of struct iscsi_conn

In iscsit_close_connection(), it calls

 send_sig(SIGINT, conn->tx_thread, 1);
 kthread_stop(conn->tx_thread);

In conn->tx_thread, which is iscsi_target_tx_thread(), when it receive
SIGINT the kthread will exit without checking the return value of
kthread_should_stop().

So if iscsi_target_tx_thread() exit right between send_sig(SIGINT...)
and kthread_stop(...), the kthread_stop() will try to stop an already
stopped kthread.

This is invalid according to the documentation of kthread_stop().

(Fix -ECONNRESET logout handling in iscsi_target_tx_thread and
 early iscsi_target_rx_thread failure case - nab)
Signed-off-by: NJiang Yi <jiangyilism@gmail.com>
Cc: <stable@vger.kernel.org> # v3.12+
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

While testing modification of per se_node_acl queue_depth forcing
session reinstatement via lio_target_nacl_cmdsn_depth_store() ->
core_tpg_set_initiator_node_queue_depth(), a hung task bug triggered
when changing cmdsn_depth invoked session reinstatement while an iscsi
login was already waiting for session reinstatement to complete.

This can happen when an outstanding se_cmd descriptor is taking a
long time to complete, and session reinstatement from iscsi login
or cmdsn_depth change occurs concurrently.

To address this bug, explicitly set session_fall_back_to_erl0 = 1
when forcing session reinstatement, so session reinstatement is
not attempted if an active session is already being shutdown.

This patch has been tested with two scenarios.  The first when
iscsi login is blocked waiting for iscsi session reinstatement
to complete followed by queue_depth change via configfs, and
second when queue_depth change via configfs us blocked followed
by a iscsi login driven session reinstatement.

Note this patch depends on commit d36ad77f to handle multiple
sessions per se_node_acl when changing cmdsn_depth, and for
pre v4.5 kernels will need to be included for stable as well.
Reported-by: NGary Guo <ghg@datera.io>
Tested-by: NGary Guo <ghg@datera.io>
Cc: Gary Guo <ghg@datera.io>
Cc: <stable@vger.kernel.org> # v4.1+
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Replace the specification of four data structures by pointer dereferences
as the parameter for the operator "sizeof" to make the corresponding size
determinations a bit safer according to the Linux coding style convention.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

The script "checkpatch.pl" pointed information out like the following.

WARNING: Possible unnecessary 'out of memory' message

Thus remove such statements here.

Link: http://events.linuxfoundation.org/sites/events/files/slides/LCJ16-Refactor_Strings-WSang_0.pdfSigned-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

* A multiplication for the size determination of a memory allocation
  indicated that an array data structure should be processed.
  Thus use the corresponding function "kcalloc".

  This issue was detected by using the Coccinelle software.

* Replace the specification of a data structure by a pointer dereference
  to make the corresponding size determination a bit safer according to
  the Linux coding style convention.
Signed-off-by: NMarkus Elfring <elfring@users.sourceforge.net>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch changes iscsi-target to propagate iscsit_transport
->iscsit_queue_data_in() and ->iscsit_queue_status() callback
errors, back up into target-core.

This allows target-core to retry failed iscsit_transport
callbacks using internal queue-full logic.
Reported-by: NPotnuri Bharat Teja <bharat@chelsio.com>
Reviewed-by: NPotnuri Bharat Teja <bharat@chelsio.com>
Tested-by: NPotnuri Bharat Teja <bharat@chelsio.com>
Cc: Potnuri Bharat Teja <bharat@chelsio.com>
Reported-by: NSteve Wise <swise@opengridcomputing.com>
Cc: Steve Wise <swise@opengridcomputing.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

We are going to split <linux/sched/signal.h> out of <linux/sched.h>, which
will have to be picked up from other headers and a couple of .c files.

Create a trivial placeholder <linux/sched/signal.h> file that just
maps to <linux/sched.h> to make this patch obviously correct and
bisectable.

Include the new header in the files that are going to need it.
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: NIngo Molnar <mingo@kernel.org>

Split iscsit_check_dataout_hdr() into two functions
1. __iscsit_check_dataout_hdr() - This function
   validates data out hdr.
2. iscsit_check_dataout_hdr() - This function finds
   iSCSI cmd using iscsit_find_cmd_from_itt_or_dump(),
   then it calls __iscsit_check_dataout_hdr() to
   validate iSCSI hdr.

This split is required to support Chelsio T6 iSCSI
DDP completion feature. T6 adapters reduce number of
completions to host by generating single completion
for all directly placed(DDP) iSCSI pdus in a sequence,
DDP completion contains iSCSI hdr of the last pdu in a
sequence.

On receiving DDP completion cxgbit driver will first
find iSCSI cmd using iscsit_find_cmd_from_itt_or_dump()
then updates cmd->write_data_done, cmd->next_burst_len,
cmd->data_sn and calls  __iscsit_check_dataout_hdr()
to validate iSCSI hdr.

(Move XRDSL check ahead of itt lookup / dump - nab)
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Allocate a task management request structure for all task management
requests, including task reassignment. This change avoids that the
se_tmr->response assignment dereferences an uninitialized se_tmr
pointer.
Reported-by: NMoshe David <mdavid@infinidat.com>
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Moshe David <mdavid@infinidat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch does not change any functionality.
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Nicholas A. Bellinger <nab@linux-iscsi.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Fix the spelling of this word in a function name, messages and
source code comments.
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Nicholas A. Bellinger <nab@linux-iscsi.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Change two occurrences of "preform" into "perform".
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NHannes Reinecke <hare@suse.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Cc: Nicholas A. Bellinger <nab@linux-iscsi.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Remove superfluous #include directives from the include/target/*.h
files. Add missing #include directives to other *.h and *.c files.
Use forward declarations for structures where possible. This
change reduces the build time for make M=drivers/target on my
laptop from 27.1s to 18.7s or by about 30%.
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Bryant G. Ly <bryantly@linux.vnet.ibm.com>

If iscsi-target receives NOP OUT with ITT and TTT
set to 0xffffffff it allocates iscsi_cmd but
does not free the cmd, so free iscsi_cmd in this case.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Trivial fix to spelling mistakes in pr_debug message and comments
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Reviewed-by: NBart Van Assche <bart.vanassche@sandisk.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch fixes a race in iscsit_release_commands_from_conn() ->
iscsit_free_cmd() -> transport_generic_free_cmd() + wait_for_tasks=1,
where CMD_T_FABRIC_STOP could end up being set after the final
kref_put() is called from core_tmr_abort_task() context.

This results in transport_generic_free_cmd() blocking indefinately
on se_cmd->cmd_wait_comp, because the target_release_cmd_kref()
check for CMD_T_FABRIC_STOP returns false.

To address this bug, make iscsit_release_commands_from_conn()
do list_splice and set CMD_T_FABRIC_STOP early while holding
iscsi_conn->cmd_lock.  Also make iscsit_aborted_task() only
remove iscsi_cmd_t if CMD_T_FABRIC_STOP has not already been
set.

Finally in target_release_cmd_kref(), only honor fabric_stop
if CMD_T_ABORTED has been set.

Cc: Mike Christie <mchristi@redhat.com>
Cc: Quinn Tran <quinn.tran@qlogic.com>
Cc: Himanshu Madhani <himanshu.madhani@qlogic.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: stable@vger.kernel.org # 3.14+
Tested-by: NNicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Instead of special casing the handful of callers that check for
iser-target rdma verbs specific shutdown, use a simple flag at
iscsit_transport->rdma_shutdown so each driver can signal this.

Also, update iscsi-target/tcp + cxgbit to rdma_shutdown = false.

Cc: Varun Prakash <varun@chelsio.com>
Cc: Hariprasad Shenai <hariprasad@chelsio.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Ensures the first page entry is within bounds. A failed check would
terminate the iSCSI connection instead of causing a NULL-dereference.

This violation seems to happen with certain iSCSI commands where the
computed CDB length is zero but the expected transfer length is
non-zero. The real problem is probably on the iSCSI initiator side
since there is a discrepancy between the iSCSI header and the
encapsulated CDB Opcode.
Signed-off-by: NImran Haider <imran1008@gmail.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

->shutdown session only decideѕ if the target core calls ->close_session
directly, or if the fabrics drivers calls it manually later through
target_put_session, which at this point will always close the session as
it has been removed from the lookup list and thus no new references will
be acquired from the core.

So instead remove ->shutdown and have the core call ->close_session
directly, and replace all calls to target_put_session in the iSCSI target
with direct calls to iscsit_close_session.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

export symbols for ISCSI_HW_OFFLOAD
transport drivers.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

ISCSI_HW_OFFLOAD transport drivers waits
on conn_logout_comp as ISCSI_TCP driver
so call complete if transport type
is ISCSI_HW_OFFLOAD.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

clear tx_thread_active for ISCSI_HW_OFFLOAD
transport in logout_post_handler functions.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Use conn_transport->transport_type instead of
ISCSI_TCP to build text response.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Move iscsit_thread_check_cpumask() to header
file so that ISCSI_HW_OFFLOAD and other transport drivers
can call this function to ensure both tx and rx thread
runs on same cpu.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Add void (*iscsit_get_r2t_ttt)() to
struct iscsit_transport, iscsi-target
uses this callback to get
r2t->targ_xfer_tag.

cxgbit.ko needs this callback
for Direct Data Placement of Data Out
pdus, adapter uses ttt in Data Out pdus
for placing data directly in to the host
buffers.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

split iscsi_target_rx_thread() into two parts,
1. iscsi_target_rx_thread() is common to all
   transport drivers, it will call Rx function
   registered by transport driver.

2. iscsit_get_rx_pdu() is Rx function for
   ISCSI_TCP transport.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Add void (*iscsit_get_rx_pdu)() to
struct iscsit_transport, iscsi-target
uses this callback to receive and
process Rx iSCSI PDUs.

cxgbit.ko needs this callback to
reuse iscsi-target Rx thread.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Acked-by: NSagi Grimberg <sagi@grimberg.me>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Add int (*iscsit_xmit_pdu)() to
struct iscsit_transport, iscsi-target
uses this callback to transmit an
iSCSI PDU.

cxgbit.ko needs this callback to
avoid duplicating iscsit_immediate_queue()
and iscsit_response_queue() code.
Signed-off-by: NVarun Prakash <varun@chelsio.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch replaces uses of the long obsolete hash interface with
either shash (for non-SG users) or ahash.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch addresses a long standing race where obtaining
se_node_acl->acl_kref in __transport_register_session()
happens a bit too late, and leaves open the potential
for core_tpg_del_initiator_node_acl() to hit a NULL
pointer dereference.

Instead, take ->acl_kref in core_tpg_get_initiator_node_acl()
while se_portal_group->acl_node_mutex is held, and move the
final target_put_nacl() from transport_deregister_session()
into transport_free_session() so that fabric driver login
failure handling using the modern method to still work
as expected.

Also, update core_tpg_get_initiator_node_acl() to take
an extra reference for dynamically generated acls for
demo-mode, before returning to fabric caller.  Also
update iscsi-target sendtargets special case handling
to use target_tpg_has_node_acl() when checking if
demo_mode_discovery == true during discovery lookup.

Note the existing wait_for_completion(&acl->acl_free_comp)
in core_tpg_del_initiator_node_acl() does not change.

Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Andy Grover <agrover@redhat.com>
Cc: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Modify indentation such that the 'smatch' tool no longer complains
about incorrect indentation + unreachable code.
Signed-off-by: NBart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: NSagi Grimberg <sagig@mellanox.com>
Reviewed-by: NAndy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch addresses a case where iscsi_target_do_tx_login_io()
fails sending the last login response PDU, after the RX/TX
threads have already been started.

The case centers around iscsi_target_rx_thread() not invoking
allow_signal(SIGINT) before the send_sig(SIGINT, ...) occurs
from the failure path, resulting in RX thread hanging
indefinately on iscsi_conn->rx_login_comp.

Note this bug is a regression introduced by:

  commit e5419865
  Author: Nicholas Bellinger <nab@linux-iscsi.org>
  Date:   Wed Jul 22 23:14:19 2015 -0700

      iscsi-target: Fix iscsit_start_kthreads failure OOPs

To address this bug, complete ->rx_login_complete for good
measure in the failure path, and immediately return from
RX thread context if connection state did not actually reach
full feature phase (TARG_CONN_STATE_LOGGED_IN).

Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: <stable@vger.kernel.org> # v3.10+
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

It appears to be what the rest of the kernel does, so let's do it too.
Signed-off-by: NAndy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This is a more natural format that lets us format it with the appropriate
printk specifier as needed.

This also lets us handle v4-mapped ipv6 addresses a little more nicely, by
storing the addr as an actual v4 sockaddr in conn->local_sockaddr.

Finally, we no longer need to maintain variables for port, since this is
contained in sockaddr. Remove iscsi_np.np_port and iscsi_conn.local_port.
Signed-off-by: NAndy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Revert commit 1997e625, which causes double brackets on ipv6
inaddr_any addresses.

Since we have np_sockaddr, if we need a textual representation we can
use "%pISc".

Change iscsit_add_network_portal() and iscsit_add_np() signatures to remove
*ip_str parameter.

Fix and extend some comments earlier in the function.

Tested to work for :: and ::1 via iscsiadm, previously :: failed, see
https://bugzilla.redhat.com/show_bug.cgi?id=1249107 .

CC: stable@vger.kernel.org
Signed-off-by: NAndy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

In a performance profile, taking a mutex in iscsit_increment_maxcmdsn()
shows up very high.  However taking a mutex around "sess->max_cmd_sn += 1"
seems pretty silly: we're not serializing against other contexts in
any useful way.

I did a quick audit and there don't appear to be any other places that
use max_cmd_sn within the mutex more than once, so this lock can't be
providing any useful serialization.

(Get correct values for logging - fix whitespace damage)
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Signed-off-by: NSpencer Baugh <sbaugh@catern.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch adds a new tpg_enabled_sendtargets configfs attribute
to allow in-band sendtargets discovery information to include
target-portal-groups (TPGs) in !TPG_STATE_ACTIVE state.

This functionality is useful for clustered iSCSI targets, where
TPGTs handled on remote cluster nodes should be advertised in
the SendTargets response.

By default, this new attribute retains the default behaviour of
existing code which to ignore portal-groups in !TPG_STATE_ACTIVE
state.
Signed-off-by: NDavid Disseldorp <ddiss@suse.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Current implementation assumes that all the buffers of an IO are linked
with a single SG list, which is OK because target-core is only allocating
a contigious scatterlist region.  However, this assumption is wrong for
se_cmd descriptors that want to use chaining across multiple SGL regions.

Fix this up by using proper SGL accessors for digest payload computation.
Signed-off-by: NAlexei Potashnik <alexei@purestorage.com>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Make sure all non-READ SCSI commands get targ_xfer_tag initialized
to 0xffffffff, not just WRITEs.

Double-free of a TUR cmd object occurs under the following scenario:

1. TUR received (targ_xfer_tag is uninitialized and left at 0)
2. TUR status sent
3. First unsolicited NOPIN is sent to initiator (gets targ_xfer_tag of 0)
4. NOPOUT for NOPIN (with TTT=0) arrives
 - its ExpStatSN acks TUR status, TUR is queued for removal
 - LIO tries to find NOPIN with TTT=0, but finds the same TUR instead,
   TUR is queued for removal for the 2nd time

(Drop unbalanced conditional bracket usage - nab)
Signed-off-by: NAlexei Potashnik <alexei@purestorage.com>
Signed-off-by: NSpencer Baugh <sbaugh@catern.com>
Cc: <stable@vger.kernel.org> # v3.1+
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch fixes a regression introduced with the following commit
in v4.0-rc1 code, where an explicit iser-target logout would result
in ->tx_thread_active being incorrectly cleared by the logout post
handler, and subsequent TX kthread leak:

    commit 88dcd2da
    Author: Nicholas Bellinger <nab@linux-iscsi.org>
    Date:   Thu Feb 26 22:19:15 2015 -0800

        iscsi-target: Convert iscsi_thread_set usage to kthread.h

To address this bug, change iscsit_logout_post_handler_closesession()
and iscsit_logout_post_handler_samecid() to only cmpxchg() on
->tx_thread_active for traditional iscsi/tcp connections.

This is required because iscsi/tcp connections are invoking logout
post handler logic directly from TX kthread context, while iser
connections are invoking logout post handler logic from a seperate
workqueue context.

Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: <stable@vger.kernel.org> # v3.10+
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>