linux-stable.git/drivers/scsi, branch v4.16.3

Revert "scsi: core: return BLK_STS_OK for DID_OK in __scsi_error_from_host_byte()"

2018-04-19T06:54:10+00:00

commit cbe095e2b584623b882ebaf6c18e0b9077baa3f7 upstream.

The description of commit e39a97353e53 is wrong: it mentions that commit
2a842acab109 introduced a bug in __scsi_error_from_host_byte() although that
commit did not change the behavior of that function.  Additionally, commit
e39a97353e53 introduced a bug: it causes commands that fail with
hostbyte=DID_OK and driverbyte=DRIVER_SENSE to be completed with
BLK_STS_OK. Hence revert that commit.

Fixes: e39a97353e53 ("scsi: core: return BLK_STS_OK for DID_OK in __scsi_error_from_host_byte()")
Reported-by: Damien Le Moal 
Signed-off-by: Bart Van Assche 
Cc: Hannes Reinecke 
Cc: Douglas Gilbert 
Cc: Damien Le Moal 
Cc: Christoph Hellwig 
Cc: Lee Duncan 
Cc: stable@vger.kernel.org
Reviewed-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

scsi: qla2xxx: Fix small memory leak in qla2x00_probe_one on probe failure

2018-04-19T06:54:10+00:00

commit 6d6340672ba3a99c4cf7af79c2edf7aa25595c84 upstream.

The code that fixes the crashes in the following commit introduced a small
memory leak:

commit 6a2cf8d3663e ("scsi: qla2xxx: Fix crashes in qla2x00_probe_one on probe failure")

Fixing this requires a bit of reworking, which I've explained. Also provide
some code cleanup.

There is a small window in qla2x00_probe_one where if qla2x00_alloc_queues
fails, we end up never freeing req and rsp and leak 0xc0 and 0xc8 bytes
respectively (the sizes of req and rsp).

I originally put in checks to test for this condition which were based on
the incorrect assumption that if ha->rsp_q_map and ha->req_q_map were
allocated, then rsp and req were allocated as well. This is incorrect.
There is a window between these allocations:

       ret = qla2x00_mem_alloc(ha, req_length, rsp_length, &req, &rsp);
                goto probe_hw_failed;

[if successful, both rsp and req allocated]

       base_vha = qla2x00_create_host(sht, ha);
                goto probe_hw_failed;

       ret = qla2x00_request_irqs(ha, rsp);
                goto probe_failed;

       if (qla2x00_alloc_queues(ha, req, rsp)) {
                goto probe_failed;

[if successful, now ha->rsp_q_map and ha->req_q_map allocated]

To simplify this, we should just set req and rsp to NULL after we free
them. Sounds simple enough? The problem is that req and rsp are pointers
defined in the qla2x00_probe_one and they are not always passed by reference
to the routines that free them.

Here are paths which can free req and rsp:

PATH 1:
qla2x00_probe_one
   ret = qla2x00_mem_alloc(ha, req_length, rsp_length, &req, &rsp);
   [req and rsp are passed by reference, but if this fails, we currently
    do not NULL out req and rsp. Easily fixed]

PATH 2:
qla2x00_probe_one
   failing in qla2x00_request_irqs or qla2x00_alloc_queues
      probe_failed:
         qla2x00_free_device(base_vha);
            qla2x00_free_req_que(ha, req)
            qla2x00_free_rsp_que(ha, rsp)

PATH 3:
qla2x00_probe_one:
   failing in qla2x00_mem_alloc or qla2x00_create_host
      probe_hw_failed:
         qla2x00_free_req_que(ha, req)
         qla2x00_free_rsp_que(ha, rsp)

PATH 1: This should currently work, but it doesn't because rsp and rsp are
not set to NULL in qla2x00_mem_alloc. Easily remedied.

PATH 2: req and rsp aren't passed in at all to qla2x00_free_device but are
derived from ha->req_q_map[0] and ha->rsp_q_map[0]. These are only set up if
qla2x00_alloc_queues succeeds.

In qla2x00_free_queues, we are protected from crashing if these don't exist
because req_qid_map and rsp_qid_map are only set on their allocation. We are
guarded in this way:

        for (cnt = 0; cnt < ha->max_req_queues; cnt++) {
                if (!test_bit(cnt, ha->req_qid_map))
                        continue;

PATH 3: This works. We haven't freed req or rsp yet (or they were never
allocated if qla2x00_mem_alloc failed), so we'll attempt to free them here.

To summarize, there are a few small changes to make this work correctly and
(and for some cleanup):

1) (For PATH 1) Set *rsp and *req to NULL in case of failure in
qla2x00_mem_alloc so these are correctly set to NULL back in
qla2x00_probe_one

2) After jumping to probe_failed: and calling qla2x00_free_device,
explicitly set rsp and req to NULL so further calls with these pointers do
not crash, i.e. the free queue calls in the probe_hw_failed section we fall
through to.

3) Fix return code check in the call to qla2x00_alloc_queues. We currently
drop the return code on the floor. The probe fails but the caller of the
probe doesn't have an error code, so it attaches to pci. This can result in
a crash on module shutdown.

4) Remove unnecessary NULL checks in qla2x00_free_req_que,
qla2x00_free_rsp_que, and the egregious NULL checks before kfrees and vfrees
in qla2x00_mem_free.

I tested this out running a scenario where the card breaks at various times
during initialization. I made sure I forced every error exit path in
qla2x00_probe_one.

Cc:  # v4.16
Fixes: 6a2cf8d3663e ("scsi: qla2xxx: Fix crashes in qla2x00_probe_one on probe failure")
Signed-off-by: Bill Kuzeja 
Acked-by: Himanshu Madhani 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

scsi: scsi_dh: Don't look for NULL devices handlers by name

2018-04-19T06:54:10+00:00

commit 2ee5671e3ae35e53bb5a53a89ac8f033e4b1721f upstream.

Currently scsi_dh_lookup() doesn't check for NULL as a device name. This
combined with nvme over dm-mpath results in the following messages
emitted by device-mapper:

 device-mapper: multipath: Could not failover device 259:67: Handler scsi_dh_(null) error 14.

Let scsi_dh_lookup() fail fast on NULL names.

[mkp: typo fix]

Cc:  # v4.16
Signed-off-by: Johannes Thumshirn 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Bart Van Assche 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

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

2018-03-28T00:11:46+00:00

Pull SCSI fixes from James Bottomley:
 "Two driver fixes (ibmvfc, iscsi_tcp) and a USB fix for devices that
  give the wrong return to Read Capacity and cause a huge log spew.

  The remaining five patches all try to fix commit 84676c1f21e8
  ("genirq/affinity: assign vectors to all possible CPUs") which broke
  the non-mq I/O path"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: iscsi_tcp: set BDI_CAP_STABLE_WRITES when data digest enabled
  scsi: sd: Remember that READ CAPACITY(16) succeeded
  scsi: ibmvfc: Avoid unnecessary port relogin
  scsi: virtio_scsi: unify scsi_host_template
  scsi: virtio_scsi: fix IO hang caused by automatic irq vector affinity
  scsi: core: introduce force_blk_mq
  scsi: megaraid_sas: fix selection of reply queue
  scsi: hpsa: fix selection of reply queue

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

2018-03-20T23:59:01+00:00

Pull SCSI fixes from James Bottomley:

 - one driver patch (qla2xxx) which fixes a problem caused by an
   existing regression fix (FCP discovery is failing)

 - one generic fix to a longstanding bug in libsas that causes I/O
   eventually to hang to the device in the face of ATA error recovery.

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: qla2xxx: Remove FC_NO_LOOP_ID for FCP and FC-NVMe Discovery
  scsi: libsas: defer ata device eh commands to libata

scsi: iscsi_tcp: set BDI_CAP_STABLE_WRITES when data digest enabled

2018-03-20T03:25:55+00:00

iscsi tcp will first send out data, then calculate and send data
digest. If we don't have BDI_CAP_STABLE_WRITES, the page cache will be
written in spite of the on going writeback. Consequently, wrong digest
will be got and sent to target.

To fix this, set BDI_CAP_STABLE_WRITES when data digest is enabled
in iscsi_tcp .slave_configure callback.

Signed-off-by: Jianchao Wang 
Acked-by: Chris Leech 
Acked-by: Lee Duncan 
Signed-off-by: Martin K. Petersen

scsi: sd: Remember that READ CAPACITY(16) succeeded

2018-03-20T03:23:04+00:00

The USB storage glue sets the try_rc_10_first flag in an attempt to
avoid wedging poorly implemented legacy USB devices.

If the device capacity is too large to be expressed in the provided
response buffer field of READ CAPACITY(10), a well-behaved device will
set the reported capacity to 0xFFFFFFFF. We will then attempt to issue a
READ CAPACITY(16) to obtain the real capacity.

Since this part of the discovery logic is not covered by the first_scan
flag, a warning will be printed a couple of times times per revalidate
attempt if we upgrade from READ CAPACITY(10) to READ CAPACITY(16).

Remember that we have successfully issued READ CAPACITY(16) so we can
take the fast path on subsequent revalidate attempts.

Reported-by: Menion 
Reviewed-by: Laurence Oberman 
Signed-off-by: Martin K. Petersen

scsi: ibmvfc: Avoid unnecessary port relogin

2018-03-15T04:52:33+00:00

Following an RSCN, ibmvfc will issue an ADISC to determine if the
underlying target has changed, comparing the SCSI ID, WWPN, and WWNN to
determine how to handle the rport in discovery. However, the comparison
of the WWPN and WWNN was performing a memcmp between a big endian field
against a CPU endian field, which resulted in the wrong answer on LE
systems. This was observed as unexpected errors getting logged at boot
time as targets were getting relogins when not needed.

Signed-off-by: Brian King 
Signed-off-by: Martin K. Petersen

scsi: virtio_scsi: unify scsi_host_template

2018-03-15T03:31:13+00:00

Now that virtio_scsi uses blk-mq exclusively, we can remove the
scsi_host_template and associated plumbing for the legacy I/O path.

[mkp: commit desc]

Cc: Omar Sandoval ,
Cc: "Martin K. Petersen" ,
Cc: James Bottomley ,
Cc: Christoph Hellwig ,
Cc: Paolo Bonzini 
Cc: Mike Snitzer 
Cc: Laurence Oberman 
Cc: Hannes Reinecke 
Signed-off-by: Ming Lei 
Suggested-by: Christoph Hellwig ,
Reviewed-by: Christoph Hellwig 
Signed-off-by: Martin K. Petersen

scsi: virtio_scsi: fix IO hang caused by automatic irq vector affinity

2018-03-15T03:31:13+00:00

Since commit 84676c1f21e8ff5 ("genirq/affinity: assign vectors to all
possible CPUs") it is possible to end up in a scenario where only
offline CPUs are mapped to an interrupt vector.

This is only an issue for the legacy I/O path since with blk-mq/scsi-mq
an I/O can't be submitted to a hardware queue if the queue isn't mapped
to an online CPU.

Fix this issue by forcing virtio-scsi to use blk-mq.

[mkp: commit desc]

Cc: Omar Sandoval ,
Cc: "Martin K. Petersen" ,
Cc: James Bottomley ,
Cc: Christoph Hellwig ,
Cc: Don Brace 
Cc: Kashyap Desai 
Cc: Paolo Bonzini 
Cc: Mike Snitzer 
Cc: Laurence Oberman 
Fixes: 84676c1f21e8 ("genirq/affinity: assign vectors to all possible CPUs")
Signed-off-by: Ming Lei 
Reviewed-by: Hannes Reinecke 
Acked-by: Paolo Bonzini 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Martin K. Petersen