We are only holding the lun_tg_pt_gp_lock in target_alua_state_check() to
make sure tg_pt_gp is not freed from under us while we copy the state,
delay, ID values. We can instead use RCU here to access the tg_pt_gp.
With this patch IOPs can increase up to 10% for jobs like:
fio --filename=/dev/sdX --direct=1 --rw=randrw --bs=4k \
--ioengine=libaio --iodepth=64 --numjobs=N
when there are multiple sessions (running that fio command to each /dev/sdX
or using multipath and there are over 8 paths), or more than 8 queues for
the loop or vhost with multiple threads case and numjobs > 8.
Link: https://lore.kernel.org/r/20210930020422.92578-5-michael.christie@oracle.com
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This patch fixes the following bugs:
1. If there are multiple ordered cmds queued and multiple simple cmds
completing, target_restart_delayed_cmds() could be called on different
CPUs and each instance could start a ordered cmd. They could then run in
different orders than they were queued.
2. target_restart_delayed_cmds() and target_handle_task_attr() can race
where:
1. target_handle_task_attr() has passed the simple_cmds == 0 check.
2. transport_complete_task_attr() then decrements simple_cmds to 0.
3. transport_complete_task_attr() runs target_restart_delayed_cmds() and
it does not see any cmds on the delayed_cmd_list.
4. target_handle_task_attr() adds the cmd to the delayed_cmd_list.
The cmd will then end up timing out.
3. If we are sent > 1 ordered cmds and simple_cmds == 0, we can execute
them out of order, because target_handle_task_attr() will hit that
simple_cmds check first and return false for all ordered cmds sent.
4. We run target_restart_delayed_cmds() after every cmd completion, so if
there is more than 1 simple cmd running, we start executing ordered cmds
after that first cmd instead of waiting for all of them to complete.
5. Ordered cmds are not supposed to start until HEAD OF QUEUE and all older
cmds have completed, and not just simple.
6. It's not a bug but it doesn't make sense to take the delayed_cmd_lock
for every cmd completion when ordered cmds are almost never used. Just
replacing that lock with an atomic increases IOPs by up to 10% when
completions are spread over multiple CPUs and there are multiple
sessions/ mqs/thread accessing the same device.
This patch moves the queued delayed handling to a per device work to
serialze the cmd executions for each device and adds a new counter to track
HEAD_OF_QUEUE and SIMPLE cmds. We can then check the new counter to
determine when to run the work on the completion path.
Link: https://lore.kernel.org/r/20210930020422.92578-3-michael.christie@oracle.com
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
It may not always be best to complete the IO on same CPU as it was
submitted on. This commit allows userspace to configure it.
This has been useful for vhost-scsi where we have a single thread for
submissions and completions. If we force the completion on the submission
CPU we may be adding conflicts with what the user has setup in the lower
levels with settings like the block layer rq_affinity or the driver's IRQ
or softirq (the network's rps_cpus value) settings.
We may also want to set it up where the vhost thread runs on CPU N and does
its submissions/completions there, and then have LIO do its completion
booking on CPU M, but can't configure the lower levels due to issues like
using dm-multipath with lots of paths (the path selector can throw commands
all over the system because it's only taking into account latency/throughput
at its level).
The new setting is in:
/sys/kernel/config/target/$fabric/$target/param/cmd_completion_affinity
Writing:
-1 -> Gives the current default behavior of completing on the
submission CPU.
-2 -> Completes the cmd on the CPU the lower layers sent it to us from.
> 0 -> Completes on the CPU userspace has specified.
Link: https://lore.kernel.org/r/20210227170006.5077-26-michael.christie@oracle.com
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
target_core_iblock is plugging and unplugging on every command and this is
causing perf issues for drivers that prefer batched cmds. With recent
patches we can now take multiple cmds from a fabric driver queue and then
pass them down the backend drivers in a batch. This patch adds this support
by adding 2 callouts to the backend for plugging and unplugging the
device. Subsequent commits will add support for iblock and tcmu device
plugging.
Link: https://lore.kernel.org/r/20210227170006.5077-22-michael.christie@oracle.com
Reviewed-by: Bodo Stroesser <bostroesser@gmail.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
loop and vhost/scsi do their target cmd submission from driver
workqueues. This allows them to avoid an issue where the backend may block
waiting for resources like tags/requests, mem/locks, etc and that ends up
blocking their entire submission path and for the case of vhost-scsi both
the submission and completion path.
This patch adds a helper drivers can use to submit from a LIO workqueue.
This code will then be extended in the next patches to fix the plugging of
backend devices.
We are only converting vhost/loop initially, but the workqueue based
submission will work for other drivers and have similar benefits where the
main target loops will not end up blocking one some backend resource.
Link: https://lore.kernel.org/r/20210227170006.5077-17-michael.christie@oracle.com
Tested-by: Laurence Oberman <loberman@redhat.com>
Reviewed-by: Bodo Stroesser <bostroesser@gmail.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
According to FCP-4 (9.4.2):
If the command requested that data beyond the length specified by the
FCP_DL field be transferred, then the device server shall set the
FCP_RESID_OVER bit (see 9.5.8) to one in the FCP_RSP IU and:
a) process the command normally except that data beyond the FCP_DL count
shall not be requested or transferred;
b) transfer no data and return CHECK CONDITION status with the sense key
set to ILLEGAL REQUEST and the additional sense code set to INVALID FIELD
IN COMMAND INFORMATION UNIT; or
c) may transfer data and return CHECK CONDITION status with the sense key
set to ABORTED COMMAND and the additional sense code set to INVALID FIELD
IN COMMAND INFORMATION UNIT.
TCM follows b) and transfers no data for residual writes but returns
INVALID FIELD IN CDB instead of INVALID FIELD IN COMMAND INFORMATION UNIT.
Change the ASCQ to INVALID FIELD IN COMMAND INFORMATION UNIT to meet the
standard.
Link: https://lore.kernel.org/r/20201203082035.54566-4-a.kovaleva@yadro.com
Signed-off-by: Anastasia Kovaleva <a.kovaleva@yadro.com>
Signed-off-by: Roman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
cmd.bad_sector currently gets packed into the sense INFORMATION field for
TCM_LOGICAL_BLOCK_{GUARD,APP_TAG,REF_TAG}_CHECK_FAILED errors, which carry
an .add_sector_info flag in the sense_detail_table to ensure this.
In preparation for propagating a byte offset on COMPARE AND WRITE
TCM_MISCOMPARE_VERIFY error, rename cmd.bad_sector to cmd.sense_info and
sense_detail.add_sector_info to sense_detail.add_sense_info so that it
better reflects the sense INFORMATION field destination.
[ddiss: update previously overlooked ib_isert]
Link: https://lore.kernel.org/r/20201031233211.5207-3-ddiss@suse.de
Reviewed-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
pgr_support and alua_support device attributes show the inverted value of
the transport_flags:
* TRANSPORT_FLAG_PASSTHROUGH_PGR
* TRANSPORT_FLAG_PASSTHROUGH_ALUA
These attributes are per device, while the flags are per backend. Rename
the transport_flags in backend/transport to transport_flags_default and use
this value to initialize the new transport_flags field in the se_device
structure.
Now data and attribute both are per se_device.
Link: https://lore.kernel.org/r/20200427150823.15350-4-bstroesser@ts.fujitsu.com
Reviewed-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Bodo Stroesser <bstroesser@ts.fujitsu.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The emulate_ua_intlck_ctrl device attribute accepts values of 0, 1 or 2 via
ConfigFS, which map to unit attention interlocks control codes in the MODE
SENSE control Mode Page. Use an enum to track these values so that it's
clear that, unlike the remaining emulate_X attributes,
emulate_ua_intlck_ctrl isn't boolean.
Link: https://marc.info/?l=target-devel&m=158227825428798
Suggested-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The LIO unmap_zeroes_data device attribute is mapped to the LBPRZ flag in
the READ CAPACITY(16) and Thin Provisioning VPD INQUIRY responses.
The unmap_zeroes_data attribute is exposed via configfs, where any write
value is correctly validated via strtobool(). However, when initialised via
target_configure_unmap_from_queue() it takes the value of the device's
max_write_zeroes_sectors queue limit, which is non-boolean.
A non-boolean value can be read from configfs, but attempting to write the
same value back results in -EINVAL, causing problems for configuration
utilities such as targetcli.
Link: https://marc.info/?l=target-devel&m=158213354011309
Fixes: 2237498f0b ("target/iblock: Convert WRITE_SAME to blkdev_issue_zeroout")
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Instead of tracking the initiator that established an SPC-2 reservation,
track the session through which the SPC-2 reservation has been
established. This patch does not change any functionality.
Cc: Mike Christie <mchristi@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Whether or not a session is being torn down does not affect whether or not
SCSI commands are in the task set. Hence remove the "tearing down" checks
from the TMF code. The TRANSPORT_ISTATE_PROCESSING check is left out
because it is now safe to wait for a command that is in that state. The
CMD_T_PRE_EXECUTE is left out because abort processing is postponed until
after commands have left the pre-execute state since the patch that makes
TMF processing synchronous.
See also commit 1c21a48055 ("target: Avoid early CMD_T_PRE_EXECUTE
failures during ABORT_TASK").
Cc: Nicholas Bellinger <nab@linux-iscsi.org>
Cc: Mike Christie <mchristi@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Hannes Reinecke <hare@suse.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In preparation for supporting user provided vendor strings, add an extra
byte to the vendor, model and revision arrays in struct t10_wwn. This
ensures that the full INQUIRY data can be carried in the arrays along with
a null-terminator.
Change a number of array readers and writers so that they account for
explicit null-termination:
- The pscsi_set_inquiry_info() and emulate_model_alias_store() codepaths
don't currently explicitly null-terminate; fix this.
- Existing t10_wwn field dumps use for-loops which step over
null-terminators for right-padding.
+ Use printf with width specifiers instead.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The TASK ABORTED STATUS (TAS) bit is defined as follows in SAM:
"TASK_ABORTED: this status shall be returned if a command is aborted by a
command or task management function on another I_T nexus and the control
mode page TAS bit is set to one". TAS handling is spread over the target
core and the iSCSI target driver. If a LUN RESET is received, the target
core will send the TASK_ABORTED response for all commands for which such a
response has to be sent. If an ABORT TASK is received, only the iSCSI
target driver will send the TASK_ABORTED response for the commands for
which that response has to be sent. That is a bug since all target drivers
have to honor the TAS bit. Fix this by moving the code that handles TAS
from the iSCSI target driver into the target core. Additionally, if a
command has been aborted, instead of sending the TASK_ABORTED status from
the context that processes the SCSI command send it from the context of the
ABORT TMF. The core_tmr_abort_task() change in this patch causes the
CMD_T_TAS flag to be set if a TASK_ABORTED status has to be sent back to
the initiator that submitted the command. If that flag has been set
transport_cmd_finish_abort() will send the TASK_ABORTED response.
Cc: Nicholas Bellinger <nab@linux-iscsi.org>
Cc: Mike Christie <mchristi@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Disseldorp <ddiss@suse.de>
Cc: Hannes Reinecke <hare@suse.de>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
