linux-stable.git/drivers/scsi/scsi_debug.c, branch linux-5.14.y

scsi: core: Kill DRIVER_SENSE

2021-06-01T02:48:21+00:00

Replace the check for DRIVER_SENSE with a check for
scsi_status_is_check_condition().

Audit all callsites to ensure the SAM status is set correctly. For
backwards compability move the DRIVER_SENSE definition to sg.h, and update
sg, bsg, and scsi_ioctl to set the DRIVER_SENSE driver_status whenever
SAM_STAT_CHECK_CONDITION is present.

[mkp: fix zeroday srp warning]

Link: https://lore.kernel.org/r/20210427083046.31620-10-hare@suse.de
Signed-off-by: Hannes Reinecke 
Signed-off-by: Martin K. Petersen 

fix

scsi: core: Introduce scsi_build_sense()

2021-06-01T02:48:21+00:00

Introduce scsi_build_sense() as a wrapper around scsi_build_sense_buffer()
to format the buffer and set the correct SCSI status.

Link: https://lore.kernel.org/r/20210427083046.31620-8-hare@suse.de
Reviewed-by: Bart Van Assche 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Hannes Reinecke 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: Drop if with an always false condition

2021-05-15T02:34:31+00:00

to_sdebug_host() is a container_of operation, so it never returns NULL.

Link: https://lore.kernel.org/r/20210506203206.254258-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Uwe Kleine-König 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: Fix cmd_per_lun, set to max_queue

2021-05-03T02:41:44+00:00

Make sure that the cmd_per_lun value placed in the host template never
exceeds the can_queue value. If the max_queue driver parameter is not
specified then both cmd_per_lun and can_queue are set to CAN_QUEUE.
CAN_QUEUE is a compile time constant and is used to dimension an array to
hold queued requests. If the max_queue driver parameter is given it is must
be less than or equal to CAN_QUEUE and if so, the host template values are
adjusted.

Remove undocumented code that allowed queue_depth to exceed CAN_QUEUE and
cause stack full type errors. There is a documented way to do that with
every_nth and

    echo 0x8000 > /sys/bus/pseudo/drivers/scsi_debug/opts

See: https://sg.danny.cz/sg/scsi_debug.html

Tweak some formatting, and add a suggestion to the "trim poll_queues"
warning.

Link: https://lore.kernel.org/r/20210415015031.607153-1-dgilbert@interlog.com
Reported-by: Kashyap Desai 
Reviewed-by: John Garry 
Signed-off-by: Douglas Gilbert 
Signed-off-by: John Garry 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: Fix cmd duration calculation

2021-03-04T22:37:03+00:00

In some cases, sdebug_defer::cmpl_ts (completion timestamp) wasn't being
properly set when REQ_HIPRI was given. Fix that and improve code to only
call ktime_get_boottime_ns() for commands with REQ_HIPRI set as cmpl_ts is
only used in that case.

Link: https://lore.kernel.org/r/20210304014107.307625-1-dgilbert@interlog.com
Signed-off-by: Douglas Gilbert 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: Add new defer type for mq_poll

2021-03-04T22:37:03+00:00

Add a new sdeb_defer_type enumeration: SDEB_DEFER_POLL for requests that
have REQ_HIPRI set in cmd_flags field. It is expected that these requests
will be polled via the mq_poll entry point which is driven by calls to
blk_poll() in the block layer. Therefore timer events are not 'wired up' in
the normal fashion.

There are still cases with short delays (e.g. < 10 microseconds) where by
the time the command response processing occurs, the delay is already
exceeded in which case the code calls scsi_done() directly. In such cases
there is no window for mq_poll() to be called.

Add 'mq_polls' counter that increments on each scsi_done() called via the
mq_poll entry point. Can be used to show (with 'cat
/proc/scsi/scsi_debug/') that blk_poll() is causing completions
rather than some other mechanism.

Link: https://lore.kernel.org/r/20210215074048.19424-5-kashyap.desai@broadcom.com
Tested-by: Kashyap Desai 
Signed-off-by: Douglas Gilbert 
Signed-off-by: Kashyap Desai 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: mq_poll support

2021-03-04T22:37:03+00:00

Add support of the mq_poll interface to scsi_debug.  This feature
requires shared host tag support in kernel and driver.

Signed-off-by: Kashyap Desai 
Acked-by: Douglas Gilbert 
Tested-by: Douglas Gilbert 
Reviewed-by: Hannes Reinecke 

Link: https://lore.kernel.org/r/20210215074048.19424-4-kashyap.desai@broadcom.com
Cc: dgilbert@interlog.com
Cc: linux-block@vger.kernel.org
Signed-off-by: Martin K. Petersen

Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

2021-02-22T18:24:58+00:00

Pull SCSI updates from James Bottomley:
 "This series consists of the usual driver updates (ufs, ibmvfc,
  qla2xxx, hisi_sas, pm80xx) plus the removal of the gdth driver (which
  is bound to cause conflicts with a trivial change somewhere).

  The only big major rework of note is the one from Hannes trying to
  clean up our result handling code in the drivers to make it
  consistent"

* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (194 commits)
  scsi: MAINTAINERS: Adjust to reflect gdth scsi driver removal
  scsi: ufs: Give clk scaling min gear a value
  scsi: lpfc: Fix 'physical' typos
  scsi: megaraid_mbox: Fix spelling of 'allocated'
  scsi: qla2xxx: Simplify the calculation of variables
  scsi: message: fusion: Fix 'physical' typos
  scsi: target: core: Change ASCQ for residual write
  scsi: target: core: Signal WRITE residuals
  scsi: target: core: Set residuals for 4Kn devices
  scsi: hisi_sas: Add trace FIFO debugfs support
  scsi: hisi_sas: Flush workqueue in hisi_sas_v3_remove()
  scsi: hisi_sas: Enable debugfs support by default
  scsi: hisi_sas: Don't check .nr_hw_queues in hisi_sas_task_prep()
  scsi: hisi_sas: Remove deferred probe check in hisi_sas_v2_probe()
  scsi: lpfc: Add auto select on IRQ_POLL
  scsi: ncr53c8xx: Fix typos
  scsi: lpfc: Fix ancient double free
  scsi: qla2xxx: Fix some memory corruption
  scsi: qla2xxx: Remove redundant NULL check
  scsi: megaraid: Fix ifnullfree.cocci warnings
  ...

scsi: scsi_debug: Fix a memory leak

2021-02-09T02:51:25+00:00

The sdebug_q_arr pointer must be freed when the module is unloaded.

$ cat /sys/kernel/debug/kmemleak
unreferenced object 0xffff888e1cfb0000 (size 4096):
  comm "modprobe", pid 165555, jiffies 4325987516 (age 685.194s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000458f4f5d>] 0xffffffffc06702d9
    [<000000003edc4b1f>] do_one_initcall+0xe9/0x57d
    [<00000000da7d518c>] do_init_module+0x1d1/0x6f0
    [<000000009a6a9248>] load_module+0x36bd/0x4f50
    [<00000000ddb0c3ce>] __do_sys_init_module+0x1db/0x260
    [<000000009532db57>] do_syscall_64+0xa5/0x420
    [<000000002916b13d>] entry_SYSCALL_64_after_hwframe+0x6a/0xdf

Fixes: 87c715dcde63 ("scsi: scsi_debug: Add per_host_store option")
Link: https://lore.kernel.org/r/20210208111734.34034-1-mlombard@redhat.com
Acked-by: Douglas Gilbert 
Signed-off-by: Maurizio Lombardi 
Signed-off-by: Martin K. Petersen

scsi: scsi_debug: Do not set COMMAND_COMPLETE

2021-01-23T02:14:10+00:00

COMMAND_COMPLETE is defined as '0', so setting it is quite pointless.

Link: https://lore.kernel.org/r/20210113090500.129644-23-hare@suse.de
Reviewed-by: Christoph Hellwig 
Acked-by: Douglas Gilbert 
Signed-off-by: Hannes Reinecke 
Signed-off-by: Martin K. Petersen