commit 49cb77e297dc611a1b795cfeb79452b3002bd331 upstream.
This patch closes a race between se_lun deletion during configfs
unlink in target_fabric_port_unlink() -> core_dev_del_lun()
-> core_tpg_remove_lun(), when transport_clear_lun_ref() blocks
waiting for percpu_ref RCU grace period to finish, but a new
NodeACL mappedlun is added before the RCU grace period has
completed.
This can happen in target_fabric_mappedlun_link() because it
only checks for se_lun->lun_se_dev, which is not cleared until
after transport_clear_lun_ref() percpu_ref RCU grace period
finishes.
This bug originally manifested as NULL pointer dereference
OOPsen in target_stat_scsi_att_intr_port_show_attr_dev() on
v4.1.y code, because it dereferences lun->lun_se_dev without
a explicit NULL pointer check.
In post v4.1 code with target-core RCU conversion, the code
in target_stat_scsi_att_intr_port_show_attr_dev() no longer
uses se_lun->lun_se_dev, but the same race still exists.
To address the bug, go ahead and set se_lun>lun_shutdown as
early as possible in core_tpg_remove_lun(), and ensure new
NodeACL mappedlun creation in target_fabric_mappedlun_link()
fails during se_lun shutdown.
Reported-by: James Shen <jcs@datera.io>
Cc: James Shen <jcs@datera.io>
Tested-by: James Shen <jcs@datera.io>
Cc: stable@vger.kernel.org # 3.10+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
commit 95cadace8f3959282e76ebf8b382bd0930807d2c upstream.
This patch allows FILEIO to update hw_max_sectors based on the current
max_bytes_per_io. This is required because vfs_[writev,readv]() can accept
a maximum of 2048 iovecs per call, so the enforced hw_max_sectors really
needs to be calculated based on block_size.
This addresses a >= v3.5 bug where block_size=512 was rejecting > 1M
sized I/O requests, because FD_MAX_SECTORS was hardcoded to 2048 for
the block_size=4096 case.
(v2: Use max_bytes_per_io instead of ->update_hw_max_sectors)
Reported-by: Henrik Goldman <hg@x-formation.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Switch back to pre commit 1c7b13fe65 list splicing logic for active I/O
shutdown with tcm_qla2xxx + ib_srpt fabrics.
The original commit was done under the incorrect assumption that it's safe to
walk se_sess->sess_cmd_list unprotected in target_wait_for_sess_cmds() after
sess->sess_tearing_down = 1 has been set by target_sess_cmd_list_set_waiting()
during session shutdown.
So instead of adding sess->sess_cmd_lock protection around sess->sess_cmd_list
during target_wait_for_sess_cmds(), switch back to sess->sess_wait_list to
allow wait_for_completion() + TFO->release_cmd() to occur without having to
walk ->sess_cmd_list after the list_splice.
Also add a check to exit if target_sess_cmd_list_set_waiting() has already
been called, and add a WARN_ON to check for any fabric bug where new se_cmds
are added to sess->sess_cmd_list after sess->sess_tearing_down = 1 has already
been set.
Cc: Joern Engel <joern@logfs.org>
Cc: Roland Dreier <roland@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Some were incremented, but never used anywhere from what I could tell.
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
SBC-3 (revision 35) says:
The PARAMETER LIST LENGTH field specifies the length in bytes of the
UNMAP parameter list that is available to be transferred from the
Data-Out Buffer. If the parameter list length is greater than zero
and less than 0008h (i.e., eight), then the device server shall
terminate the command with CHECK CONDITION status with the sense key
set to ILLEGAL REQUEST and the additional sense code set to
PARAMETER LIST LENGTH ERROR. A PARAMETER LIST LENGTH set to zero
specifies that no data shall be sent.
so our sense code for too-short descriptors was wrong, and we were
incorrectly failing commands that didn't transfer any descriptors.
While we're at it, also handle the UNMAP check:
If the ANCHOR bit is set to one, and the ANC_SUP bit in the Logical
Block Provisioning VPD page (see 6.6.4) is set to zero, then the
device server shall terminate the command with CHECK CONDITION
status with the sense key set to ILLEGAL REQUEST and the additional
sense code set to INVALID FIELD IN CDB.
(chris boot: Fix wrong cut+paste comment in transport_send_check_condition_and_sense)
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch changes LIO to use the configfs backend device name as the
model if you echo '1' to an individual device's emulate_model_alias attribute.
This is a valid operation only on devices with an export count of 0.
Signed-off-by: Tregaron Bayly <tbayly@bluehost.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
"buf" is 128 characters and "vpd->device_identifier" is 256. It makes
the static checkers complain.
Also bump VPD_TMP_BUF_SIZE to match INQUIRY_VPD_DEVICE_IDENTIFIER_LEN.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Introduce TCM_NO_SENSE, mapping to sense code
'Not ready, no additional sense information'.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Cc: Nicholas Bellinger <nab@risingtidesystems.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Thanks for reviews, looking a lot better.
---- 8< ----
Initiator access config could be easier. The way other storage vendors
have addressed this is to support initiator groups: the admin adds
initiator WWNs to the group, and then LUN permissions can be granted for
the entire group at once.
Instead of changing ktarget's configfs interface, this patch keeps
the configfs interface per-initiator-wwn and just adds a 'tag' field
for each. This should be enough for user tools like targetcli to group
initiator ACLs and sync their configurations.
acl_tag is not used internally, but needs to be kept in configfs so that
all user tools can avoid dependencies on each other.
Code tested to work, although userspace pieces still to be implemented.
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds [dev,lun]_link_magic value assignment + checks within generic
target_fabric_port_link() and target_fabric_mappedlun_link() code to ensure
destination config_item *target_item sent from configfs_symlink() ->
config_item_operations->allow_link() is the underlying se_device->dev_group
and se_lun->lun_group that we expect to symlink.
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds a new max_write_same_len device attribute for use with
WRITE_SAME w/ UNMAP=0 backend emulation. This can be useful for
lowering the default backend value (IBLOCK uses 0xFFFF).
Also, update block limits VPD emulation code in spc_emulate_evpd_b0() to
report MAXIMUM WRITE SAME LENGTH, and enforce max_write_same_len during
sbc_parse() -> sbc_setup_write_same() CDB sanity checking for all emulated
WRITE_SAME w/ UNMAP=0 cases.
(Robert: Move max_write_same_len check in sbc_setup_write_same() to
check both WRITE_SAME w/ UNMAP=1 and w/ UNMAP=0 cases)
Cc: Christoph Hellwig <hch@lst.de>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Robert Elliott <Elliott@hp.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Pass the sense reason as an explicit return value from the I/O submission
path instead of storing it in struct se_cmd and using negative return
values. This cleans up a lot of the code pathes, and with the sparse
annotations for the new sense_reason_t type allows for much better
error checking.
(nab: Convert spc_emulate_modesense + spc_emulate_modeselect to use
sense_reason_t with Roland's MODE SELECT changes)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
We always support ALUA for virtual backends, and never for physical ones. Simplify
the code to just deal with these two cases and remove the superflous abstractions.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
We do not support host-level reservations for the pscsi backend, and all
virtual backends are newere than SCSI-2, so just make the combined
SPC-3 + SCSI-2 support the only supported variant and kill the switches
for the different implementations, given that this code handles the no-op
version just fine.
(hch: Update DRF_SPC2_RESERVATIONS lock usage)
Signed-off-by: Christoph Hellwig <hch@lst.de>
We can just key off ordered tag emulation of the transport_type field.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Simplify the code a lot by killing the superflous struct se_subsystem_dev.
Instead se_device is allocated early on by the backend driver, which allocates
it as part of its own per-device structure, borrowing the scheme that is for
example used for inode allocation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Similar to INQUIRY and MODE SENSE, construct the sense data in a
buffer and later copy it to the scatterlist. Do not do anything,
but still clear a pending unit attention condition, if the allocation
length is zero.
However, SPC tells us that "If a REQUEST SENSE command is terminated with
CHECK CONDITION status [and] the REQUEST SENSE command was received on
an I_T nexus with a pending unit attention condition (i.e., before the
device server reports CHECK CONDITION status), then the device server
shall not clear the pending unit attention condition." Do the
transport_kmap_data_sg early to detect this case.
It also tells us "Device servers shall not adjust the additional sense
length to reflect truncation if the allocation length is less than the
sense data available", so do not do that! Note that the err variable
is write-only.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This was originally for helping fabrics to determine overflow/underflow
status, and has been superceeded by SCF_OVERFLOW_BIT + SCF_UNDERFLOW_BIT.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Many SCSI commands are defined to return a CHECK CONDITION / ILLEGAL
REQUEST with ASC set to LOGICAL BLOCK ADDRESS OUT OF RANGE if the
initiator sends a command that accesses a too-big LBA. Add an enum
value and case entries so that target code can return this status.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Cc: stable@vger.kernel.org
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Since we set se_session.sess_tearing_down and stop new commands from
being added to se_session.sess_cmd_list before we wait for commands to
finish when freeing a session, there's no need for a separate
sess_wait_list -- if we let new commands be added to sess_cmd_list
after setting sess_tearing_down, that would be a bug that breaks the
logic of waiting in-flight commands.
Also rename target_splice_sess_cmd_list() to
target_sess_cmd_list_set_waiting(), since we are no longer splicing
onto a separate list.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The last functionality of the target processing thread is offloading possibly
long running task management requests from the submitter context. To keep
TMR semantics the same we need a single threaded ordered queue, which can
be provided by a per-device workqueue with the right flags.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Remove this command submission path which is not used by any in-tree driver.
This also removes the now unused new_cmd_map fabtric method, which a few
drivers implemented despite never calling transport_generic_handle_cdb_map.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Remove the dead SCF_SE_ALLOW_EOO and SCF_DELAYED_CMD_FROM_SAM_ATTR
from se_cmd_flags_table.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Also remove the unused ref_task_lun field in struct se_tmr_req.
(nab: Add missing TASK_REASSIGN ref_lun vs. ref_cmd orig_fe_lun checks
in iscsit_tmr_task_reassign)
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Since "target: Drop se_device TCQ queue_depth usage from I/O path" we always
submit all commands (or back then, tasks) from __transport_execute_tasks.
That means the the execute list has lots its purpose, as we can simply
submit the commands that are restarted in transport_complete_task_attr
directly while we walk the list. In fact doing so also solves a race
in the way it currently walks to delayed_cmd_list as well.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
We don't need three flags to classifiy the CDB as we can check for a NULL S/G
list for a dataless command, and can infer from the absence of the data flag
that we deal with a control CDB. Also remove the _SG_IO from the data CDB
flag as all I/O is dont on S/G lists now.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds support for ALUA MI_REPORT_TARGET_PGS extended header
format defined within SPC-4. It changes target core ALUA emulation logic
within target_emulate_report_target_port_groups() to support both the
extended and original length only header formats.
It includes adding a new 'implict_trans_secs' attribute for each ALUA
target port group to control the value returned to the application client
for an recommended implict translation timeout in seconds. By default
this value is currently set to zero, and limited up to 255 by virtue of
using a single byte in the extended header format.
This value is used by target_emulate_report_target_port_groups() within
the extended header logic to set IMPLICIT TRANSITION TIME as defined by
spc4r30.
Cc: Hannes Reinecke <hare@suse.de>
Cc: Rob Evers <revers@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This patch removes the original usage of dev_attr->max_sectors in favor of
dev_attr->hw_max_sectors that is now being enforced by target core from
within transport_generic_cmd_sequencer() for SCF_SCSI_DATA_SG_IO_CDB ops.
After the recent se_task removal patches from hch, this value for IBLOCK
backends being set via configfs by userspace from an saved max_sectors
value that is turning out to be problematic, so it makes sense to go ahead
and remove this now legacy attribute all-together.
This patch also continues to make se_dev_set_default_attribs() do
(sectors / block_size) alignment for what actually get used by
target_core_mod to be safe here, following the same alignment currently
used by fabric_max_sectors.
Reported-by: Andy Grover <agrover@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that tasks are gone we are guaranteed to only get a single completion
per command, and thus don't need this counter.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that tasks are gone we are guaranteed to only get a single completion
per command, and thus don't need this counter.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that tasks are gone we are guaranteed to only get a single completion
per command, and thus don't need this counter.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
We can use struct se_cmd for everything it did. Make sure to pass the S/G
list and data direction to the execution function to ease adding back BIDI
support later on.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that we only have a single task per command we can use a direct pointer
to it instead of list.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Make CDB emulation work on commands instead of tasks again as a preparation
of removing tasks completely.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Remove the task_sectors field that isn't used anywhere.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that we don't split commands the size field in the task is always
equivalent to the one in the CDB, even in cases where we have two tasks
due to a BIDI transfer. Just refer the the size in the command instead
of duplicating it in the task.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that we don't split commands the lba field in the task is always
equivalent to the one in the CDB, even in cases where we have two tasks
due to a BIDI transfer. Just refer the the lba in the command instead
of duplicating it in the task.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Now that all fabrics are converted over to using se_cmd->t_data_sg
directly, we can drop the task sg chaining support. With the modern
memory allocation in target core, task sg chaining is needless
overhead -- we would split up the main cmd sglist into pieces, and
then splice those pieces back together instead of just using the
original list directly.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
The UASP protocol does not inform the target device upfront how much
data it should expect so we have to learn in from the CDB. So in order
to handle this case, add a TARGET_SCF_UNKNOWN_SIZE to target_submit_cmd()
and perform an explictly assignment for se_cmd->data_length from the
extracted CDB size in transport_generic_cmd_sequencer().
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
This was used at one time as a hack by FILEIO backend registration to
allow a struct block_device that was claimed with blkdev_get (by a local
filesystem mount for example) to be exported as read-only (SCSI WP=1).
Since FILEIO backend registration will no longer attempt to obtain
exclusive access to an underlying struct block_device here, this flag is
now obsolete.
Reported-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
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: Joern Engel <joern@logfs.org>
Signed-off-by: Nicholas 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: Joern Engel <joern@logfs.org>
Signed-off-by: Nicholas 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: Nicholas 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: Nicholas 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: Nicholas 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: Nicholas Bellinger <nab@linux-iscsi.org>
Nicholas Bellinger