Turns an order-8 allocation into slab-sized ones, thereby preventing
allocation failures with memory fragmentation.

This likely saves memory as well, as the slab allocator can pack objects
more tightly than the buddy allocator.

(nab: Fix lio-core patch fuzz)
Signed-off-by: NJoern Engel <joern@logfs.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Turns an order-10 allocation into slab-sized ones, thereby preventing
allocation failures with memory fragmentation.

This likely saves memory as well, as the slab allocator can pack objects
more tightly than the buddy allocator.
Signed-off-by: NJoern Engel <joern@logfs.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

If we don't handle a given service action, we're supposed to return
INVALID FIELD IN CDB, since we do handle the SERVICE ACTION IN opcode.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Remove the hack that has READ CAPACITY(10) return 0xFFFFFFFF as the
number of sectors when thin provisioning is enabled.  This is supposed
to trigger the initiator to use READ CAPACITY(16) in this case so that
it finds out about thin provisioning.  But an initiator that cares about
thin provisioning is going to ask anyway, and an initiator that doesn't
know about READ CAPACITY(16) is going to get the wrong capacity.  So
just have READ CAPACITY(10) return the size it's supposed to.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Bump core version to v4.1.0-rc2-ml, and for versions from the
following mainline fabric modules:

loopback: v2.1-rc2
tcm_fc: v0.4
iscsi-target: v4.1.0-rc2
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch fixes a bug in tcm_fc where fc_exch memory from fc_exch_mgr->ep_pool
is currently being leaked by ft_send_resp_status() usage.  Following current
code in ft_queue_status() response path, using lport->tt.seq_send() needs to be
followed by a lport->tt.exch_done() in order to release fc_exch memory back into
libfc_em kmem_cache.

ft_send_resp_status() code is currently used in pre submit se_cmd ft_send_work()
error exceptions, TM request setup exceptions, and main TM response callback
path in ft_queue_tm_resp().  This bugfix addresses the leak in these cases.

Cc: Mark D Rustad <mark.d.rustad@intel.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Robert Love <robert.w.love@intel.com>
Cc: Andy Grover <agrover@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch drops the following unused legacy API callers from target_core_fabric.h:

*) TFO->fall_back_to_erl0()
*) TFO->stop_session()
*) TFO->sess_logged_in()
*) TFO->is_state_remove()

This patch also removes the stub usage in loopback, tcm_fc, iscsi_target,
and ib_srpt fabric modules.
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch converts iscsi-target session code to use se_sess->sess_kref
counting for iscsi session shutdown.  The following cases include:

*) last iscsit_close_connection() shutdown path to invoke close session
*) iscsit_logout_post_handler_closesession() for explict logout
*) iscsit_free_session() caller for explict shutdown

It also moves iscsit_stop_session() call from lio_tpg_close_session()
into lio_tpg_shutdown_session() TFO callbacks to invoke an explict
shutdown, and also changes iscsi_check_for_session_reinstatement()
login code to use se_sess->sess_kref.

(v2: Make iscsit_handle_time2retain_timeout() use target_put_session)

Cc: Andy Grover <agrover@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch converts core_tpg_del_initiator_node_acl() shutdown from configfs
context to use se_node_acl->acl_kref and ->acl_free_comp in order to wait for
outstanding fabric callbacks to complete via transport_deregister_session()
callbacks before waking ->acl_free_comp from the last ->acl_kref put.

It also changes core_tpg_del_initiator_node_acl() to setup a local sess_list
with target_get_session() + acl->acl_stop = 1 for active sessions that will
be shutdown, and changes transport_deregister_session_configfs() to check
for ->acl_stop usage.

Cc: Roland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Joern Engel <joern@logfs.org>
Cc: Andy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch adds se_node_acl->acl_kref for use with ->acl_free_comp
during explict se_node_acl release.  It adds kref_init() during
se_node_acl setup, kref_get() during __transport_register_session()
-> target_put_nacl() with existing transport_deregister_session()
fabric callback usage.

It also moves transport_free_session() to release *se_sess memory
after target_put_nacl() execution in transport_deregister_session()

Cc: Roland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Joern Engel <joern@logfs.org>
Cc: Andy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Add se_node_acl->acl_free_comp for NodeACL release path to wait for outstanding
fabric session shutdown to complete in transport_deregister_session() before
finishing NodeACL release from configfs process context.

Also make transport_deregister_session() clear the comp_nacl bit
to skip se_node_acl->acl_free_comp completion for dynamically generated
NodeACL during fabric session shutdown.

Cc: Roland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Joern Engel <joern@logfs.org>
Cc: Andy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch adds basic se_session->sess_kref and get/put helpers for fabric
session reference counting.  It sets the initial kref in transport_init_session()
and adds a target_release_session() callback to invoke TFO->close_session()
for final session shutdown.

Cc: Roland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Joern Engel <joern@logfs.org>
Cc: Andy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch converts the remaining struct se_portal_group->session_lock
usage to use irqsave+irqrestore to address the following warnings for
hardware target mode interrupt context usage.  This change generate
other warnings for current iscsi-target mode still using ->session_lock
with spin_lock_bh, which will need to be converted in a seperate patch.

[  492.480728] [ INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected ]
[  492.488194] 3.0.0+ #23
[  492.490820] ------------------------------------------------------
[  492.497704] sh/7162 [HC0[0]:SC0[2]:HE0:SE0] is trying to acquire:
[  492.504493] (&(&se_tpg->session_lock)->rlock){+.....}, at: [<ffffffffa022364d>] transport_deregister_session+0x2d/0x163 [target_core_mod]
  492.518390]
[  492.518390] and this task is already holding:
[  492.524897] (&(&ha->hardware_lock)->rlock){-.-...}, at: [<ffffffffa00b9146>] qla_tgt_stop_phase1+0x5e/0x27e [qla2xxx]
[  492.536856] which would create a new lock dependency:
[  492.542481] (&(&ha->hardware_lock)->rlock){-.-...} -> (&(&se_tpg->session_lock)->rlock){+.....}
[  492.552321]
[  492.552321] but this new dependency connects a HARDIRQ-irq-safe lock:
[  492.561149] (&(&ha->hardware_lock)->rlock){-.-...}
[  492.566400] ... which became HARDIRQ-irq-safe at:
[  492.571841]   [<ffffffff81064720>] __lock_acquire+0x68f/0x921
[  492.578247]   [<ffffffff81064eff>] lock_acquire+0xe0/0x10d
[  492.584367]   [<ffffffff813a74c6>] _raw_spin_lock_irqsave+0x44/0x56
[  492.591358]   [<ffffffffa009b1be>] qla24xx_msix_default+0x5c/0x2aa [qla2xxx]
[  492.599227]   [<ffffffff81088582>] handle_irq_event_percpu+0x5a/0x197
[  492.606413]   [<ffffffff810886fb>] handle_irq_event+0x3c/0x5c
[  492.612822]   [<ffffffff8108a6dc>] handle_edge_irq+0xcc/0xf1
[  492.619138]   [<ffffffff810039b9>] handle_irq+0x83/0x8e
[  492.624971]   [<ffffffff8100333e>] do_IRQ+0x48/0xaf
[  492.630413]   [<ffffffff813a7cd3>] ret_from_intr+0x0/0x1a
[  492.636437]   [<ffffffff81001dc1>] cpu_idle+0x5b/0x8d
[  492.642073]   [<ffffffff81392709>] rest_init+0xad/0xb4
[  492.647809]   [<ffffffff81a1cbbc>] start_kernel+0x366/0x371
[  492.654030]   [<ffffffff81a1c2b1>] x86_64_start_reservations+0xb8/0xbc
[  492.661311]   [<ffffffff81a1c3b6>] x86_64_start_kernel+0x101/0x110
[  492.668204]
[  492.668205] to a HARDIRQ-irq-unsafe lock:
[  492.674324] (&(&se_tpg->session_lock)->rlock){+.....}
[  492.679862] ... which became HARDIRQ-irq-unsafe at:
[  492.685497] ...  [<ffffffff8106479a>] __lock_acquire+0x709/0x921
[  492.692209]   [<ffffffff81064eff>] lock_acquire+0xe0/0x10d
[  492.698330]   [<ffffffff813a75ed>] _raw_spin_lock_bh+0x31/0x40
[  492.704836]   [<ffffffffa021c208>] core_tpg_del_initiator_node_acl+0x89/0x336 [target_core_mod]
[  492.714546]   [<ffffffffa02fb075>] tcm_qla2xxx_drop_nodeacl+0x20/0x2d [tcm_qla2xxx]
[  492.723087]   [<ffffffffa02108d9>] target_fabric_nacl_base_release+0x22/0x24 [target_core_mod]
[  492.732698]   [<ffffffffa01661c8>] config_item_release+0x7d/0xa3 [configfs]
[  492.740465]   [<ffffffff811d48fe>] kref_put+0x43/0x4d
[  492.746101]   [<ffffffffa0166149>] config_item_put+0x19/0x1b [configfs]
[  492.753481]   [<ffffffffa0164987>] configfs_rmdir+0x1eb/0x258 [configfs]
[  492.760957]   [<ffffffff810ecc54>] vfs_rmdir+0x79/0xd0
[  492.766690]   [<ffffffff810eec4a>] do_rmdir+0xc2/0x111
[  492.772423]   [<ffffffff810eecd0>] sys_rmdir+0x11/0x13
[  492.778156]   [<ffffffff813ae4d2>] system_call_fastpath+0x16/0x1b
[  492.784953]
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Correct spelling "exeeds" to "exceeds" in iscsi/iscsi_target_nego.c
Correct spelling "diabled" to "disabled" in target_core_configfs.c
Signed-off-by: NMasanari Iida <standby24x7@gmail.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch fixes a free after use in lio_target_make_nodeacl() where
iscsi_node_acl was referenced from the original se_nacl_new allocation,
instead of from core_tpg_add_initiator_node_acl() in the case of dynamic
-> explict NodeACL conversion.

Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch allows target_submit_tmr() to pass gfp_t for se_cmd->se_tmr_req
allocation, and also set up se_cmd->se_tmr_req->ref_task_tag for passed
tag with TMR_ABORT_TASK.

Also update tcm_fc(fcoe) parameter usgae and add ref_task_tag FIXME
for TMR_ABORT_TASK usage,

Cc: Andy Grover <agrover@redhat.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

In order to handle incoming target_submit_tmr() callers using interrupt
context, use system_wq process context for transport_lookup_tmr_lun()
failures to return TMR_LUN_DOES_NOT_EXIST status for fabric TM response.

Cc: Andy Grover <agrover@redhat.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch makes ft_send_tm() handle target_submit_tmr() allocation failures
via an ft_send_resp_code_and_free() w/ FCP_TMF_FAILED status.

Cc: Andy Grover <agrover@redhat.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch makes target_submit_tmr() se_tmr_req allocation occur before
target_get_sess_cmd(), and changes target_submit_tmr() to return a failure
w/ non zero status to the fabric caller upon core_tmr_alloc_req() failure.

Cc: Andy Grover <agrover@redhat.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch changes target_submit_tmr() LUN lookup failures to set response
status TMR_LUN_DOES_NOT_EXIST and invoke TFO->queue_tm_rsp(), instead of
sending CHECK_CONDITION status that can cause problems with Linux FCoE
clients.

Tested with tcm_fc ports and explict LUN_RESET lookup failures.

Cc: Andy Grover <agrover@redhat.com>
Cc: Kiran Patil <kiran.patil@intel.com>
Cc: Arun Easi <arun.easi@qlogic.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Make iscsit_alloc_buffs() failure case for page_alloc_failed use correct
__free_page() SGL pointer, and return -ENOMEM for iscsit_allocate_iovecs
failure to push se_cmd->t_mem_sg release into iscsit_release_cmd()
callback during iscsit_add_reject_from_cmd() connection reset.

Also drop cmd->t_mem_sg = NULL assignment from page_alloc_failed
failure case.
Reported-by: NRoland Dreier <roland@purestorage.com>
Cc: Andy Grover <agrover@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

The old code did (MSB << 8) & 0xff, which always evaluates to 0.  Just use
get_unaligned_be16() so we don't have to worry about whether our open-coded
version is correct or not.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Signed-off-by: NSpencer Krum <nibz@cat.pdx.edu>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

cdb_offset is always equal to offset - 8, so remove that one.  More
importantly, the existing code only worked correct if
se_cmd->data_length is a multiple of 8.  Pass in a length of, say, 9 and
we will happily overwrite 7 bytes of "unallocated" memory.

Now, afaics this bug is currently harmless, as allocations will
implicitly be padded to multiples of 8 bytes.  But depending on such a
fact wouldn't qualify as sound engineering practice.
Signed-off-by: NJoern Engel <joern@logfs.org>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

transport_kmap_data_sg can return NULL.  I never saw this trigger, but
returning -ENOMEM seems better than a crash.  Also removes a pointless
case while at it.
Signed-off-by: NJoern Engel <joern@logfs.org>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Fix possible NULL pointer dereference in target_report_luns failure path.
Signed-off-by: NJoern Engel <joern@logfs.org>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

If the target core signals an over- or under-run, tcm_loop should call
scsi_set_resid() to tell the SCSI midlayer about the residual data length.

The difference can be seen by doing something like

    strace -eioctl sg_raw -r 1024 /dev/sda 8 0 0 0 1 0 > /dev/null

and looking at the "resid=" part of the SG_IO ioctl -- after this patch,
the field is correctly reported as 512.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

se_dev_attrib.max_sectors currently has two independent meanings:

 - It is reported in the block limits VPD page as the maximum transfer
   length, ie the largest IO that the front-end (fabric) can handle.
   Also the target core doesn't enforce this maximum transfer length.

 - It is used to hold the size of the largest IO that the back-end can
   handle, so we know when to split SCSI commands into multiple tasks.

Fix this by adding a new se_dev_attrib.fabric_max_sectors to hold the
maximum transfer length, and checking incoming IOs against that limit.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

SPC-4 says about the WBUS16 and SYNC bits:

    The meanings of these fields are specific to SPI-5 (see 6.4.3).
    For SCSI transport protocols other than the SCSI Parallel
    Interface, these fields are reserved.

We don't have a SPI fabric module, so we should never set these bits.
(The comment was misleading, since it only mentioned Sync but the
actual code set WBUS16 too).
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Current code sets the peripheral device type to 0x3f == "not present
unknown" for virtual LUN 0 for standard INQUIRY commands, but leaves it
as 0 == "connected direct access block" for VPD INQUIRY commands.  This
is just because the check for LUN 0 only happens in some code paths.

Make our peripheral device type consistent by moving the LUN 0 check
into the common emulate_inquiry() code.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

If the initiator sends us an INQUIRY command with an allocation length
that's shorter than what we want to return, we're simply supposed to
truncate our response and return what the initiator gave us space for,
without signaling any error.  Current target code has various tests that
don't fill out the full response if the buffer is too short and
sometimes return errors incorrectly.

Fix this up by allocating a bounce buffer for INQUIRY responses if we
need to, ie if we have cmd->data_length too small as well as
SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC set in cmd->se_cmd_flags -- for most
fabrics, we always allocate at least a full page, but for tcm_loop we
may have a small buffer coming directly from the SCSI stack.

This lets us delete a lot of cmd->data_length checking, and also makes
our INQUIRY handling correct per SPC in a lot more cases.
Signed-off-by: NRoland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch adds initial support for TMR_ABORT_TASK ops for se_cmd
descriptors using se_sess->sess_cmd_list and se_cmd->cmd_kref counting.

It will perform an explict abort for all outstanding se_cmd ops based
upon tmr->ref_task_tag that have not been set CMD_T_COMPLETE.
It will cancel se_cmd->work and wait for backing I/O to complete before
attempting to send SAM_STAT_TASK_ABORTED and perform
target_put_sess_cmd() to release the referenced descriptor.

It also adds a CMD_T_ABORTED check into transport_complete_task() to
catch the completion from backend I/O that has been aborted, and
updates transport_wait_for_tasks() to allow CMD_T_ABORTED usage with
core_tmr_abort_task() context.
Reported-by: NRoland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This patch changes target_release_cmd_kref() to make TFO->release_cmd()
call when list_empty(&se_cmd->se_cmd_list) is TRUE.  This is required
for TMR_ABORT_TASK operation where the referenced tag descriptor may
have already been pulled of the session command list.
Reported-by: NRoland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

When TARGET_SCF_ACK_KREF is in use with target_submit_cmd() for
setting the extra acknowledgement reference to se_cmd->cmd_kref,
go ahead and set SCF_ACK_KREF in order to be used later by
abort task.
Reported-by: NRoland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

transport_generic_request_failure() is a wrapper around calling
transport_send_check_condition_and_sense() that is required once
an se_cmd->cmd_kref has been obtained via target_submit_cmd() ->
target_get_sess_cmd().

tcm_qla2xxx currently requires this, and since it's necessary for
other callers using target_submit_cmd() make it exportable now.
Reported-by: NRoland Dreier <roland@purestorage.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This code isn't broken per se, but it's scary to look at! It looks like
in the t_data_nents==1 case we're doing both a kunmap and a vunmap,
what's saving us is that t_data_vmap in this case is 0, so vunmap
doesn't do anything.

Return after kunmap, so the handling of the three cases does not overlap.
Signed-off-by: NAndy Grover <agrover@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Now that we use a workqueue for I/O submission there is no need to use
transport_generic_handle_cdb_map any more.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

Apply the qla2xxx model of submitting all commands from a workqueue.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>

This function makes little sense as a separate abstraction as it's deeply
interwinded with the control flow of its only caller.  Merged it into
tcm_loop_queuecommand after factoring out a helper to convert the task
attribute representation.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>