linux.git/drivers/scsi, branch v4.12-rc4

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

2017-06-04T18:15:43+00:00

Pull SCSI fixes from James Bottomley:
 "This is nine fixes, seven of which are for the qedi driver (new as of
  4.10) the other two are a use after free in the cxgbi drivers and a
  potential NULL dereference in the rdac device handler"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: libcxgbi: fix skb use after free
  scsi: qedi: Fix endpoint NULL panic during recovery.
  scsi: qedi: set max_fin_rt default value
  scsi: qedi: Set firmware tcp msl timer value.
  scsi: qedi: Fix endpoint NULL panic in qedi_set_path.
  scsi: qedi: Set dma_boundary to 0xfff.
  scsi: qedi: Correctly set firmware max supported BDs.
  scsi: qedi: Fix bad pte call trace when iscsiuio is stopped.
  scsi: scsi_dh_rdac: Use ctlr directly in rdac_failover_get()

Merge git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending

2017-06-01T17:40:41+00:00

Pull SCSI target fixes from Nicholas Bellinger:
 "Here are the target-pending fixes for v4.12-rc4:

   - ibmviscsis ABORT_TASK handling fixes that missed the v4.12 merge
     window. (Bryant Ly and Michael Cyr)

   - Re-add a target-core check enforcing WRITE overflow reject that was
     relaxed in v4.3, to avoid unsupported iscsi-target immediate data
     overflow. (nab)

   - Fix a target-core-user OOPs during device removal. (MNC + Bryant
     Ly)

   - Fix a long standing iscsi-target potential issue where kthread exit
     did not wait for kthread_should_stop(). (Jiang Yi)

   - Fix a iscsi-target v3.12.y regression OOPs involving initial login
     PDU processing during asynchronous TCP connection close. (MNC +
     nab)

  This is a little larger than usual for an -rc4, primarily due to the
  iscsi-target v3.12.y regression OOPs bug-fix.

  However, it's an important patch as MNC + Hannes where both able to
  trigger it using a reduced iscsi initiator login timeout combined with
  a backend taking a long time to complete I/Os during iscsi login
  driven session reinstatement"

* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
  iscsi-target: Always wait for kthread_should_stop() before kthread exit
  iscsi-target: Fix initial login PDU asynchronous socket close OOPs
  tcmu: fix crash during device removal
  target: Re-add check to reject control WRITEs with overflow data
  ibmvscsis: Fix the incorrect req_lim_delta
  ibmvscsis: Clear left-over abort_cmd pointers

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

2017-05-25T03:29:53+00:00

Pull SCSI fixes from James Bottomley:
"This is quite a big update because it includes a rework of the lpfc
driver to separate the NVMe part from the FC part.

The reason for doing this is because two separate trees (the nvme and
scsi trees respectively) want to update the individual components and
this separation will prevent a really nasty cross tree entanglement by
the time we reach the next merge window.

The rest of the fixes are the usual minor sort with no significant
security implications"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (25 commits)
scsi: zero per-cmd private driver data for each MQ I/O
scsi: csiostor: fix use after free in csio_hw_use_fwconfig()
scsi: ufs: Clean up some rpm/spm level SysFS nodes upon remove
scsi: lpfc: fix build issue if NVME_FC_TARGET is not defined
scsi: lpfc: Fix NULL pointer dereference during PCI error recovery
scsi: lpfc: update version to 11.2.0.14
scsi: lpfc: Add MDS Diagnostic support.
scsi: lpfc: Fix NVMEI's handling of NVMET's PRLI response attributes
scsi: lpfc: Cleanup entry_repost settings on SLI4 queues
scsi: lpfc: Fix debugfs root inode "lpfc" not getting deleted on driver unload.
scsi: lpfc: Fix NVME I+T not registering NVME as a supported FC4 type
scsi: lpfc: Added recovery logic for running out of NVMET IO context resources
scsi: lpfc: Separate NVMET RQ buffer posting from IO resources SGL/iocbq/context
scsi: lpfc: Separate NVMET data buffer pool fir ELS/CT.
scsi: lpfc: Fix NMI watchdog assertions when running nvmet IOPS tests
scsi: lpfc: Fix NVMEI driver not decrementing counter causing bad rport state.
scsi: lpfc: Fix nvmet RQ resource needs for large block writes.
scsi: lpfc: Adding additional stats counters for nvme.
scsi: lpfc: Fix system crash when port is reset.
scsi: lpfc: Fix used-RPI accounting problem.
...

scsi: libcxgbi: fix skb use after free

2017-05-24T02:39:14+00:00

skb->data is assigned to task->hdr in cxgbi_conn_alloc_pdu(),
skb gets freed after tx but task->hdr is still dereferenced in
iscsi_tcp_task_xmit() to avoid this call skb_get() after allocating skb
and free the skb in cxgbi_cleanup_task() or before allocating new skb in
cxgbi_conn_alloc_pdu().

Signed-off-by: Varun Prakash 
Signed-off-by: Martin K. Petersen

scsi: qedi: Fix endpoint NULL panic during recovery.

2017-05-24T02:16:43+00:00

Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen

scsi: qedi: set max_fin_rt default value

2017-05-24T02:16:43+00:00

max_fin_rt is the maximum re-transmission of FIN packets
as part of the termination flow. After reaching this value
the FW will send a single RESET.

Signed-off-by: Nilesh Javali 
Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen

scsi: qedi: Set firmware tcp msl timer value.

2017-05-24T02:16:43+00:00

Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen

scsi: qedi: Fix endpoint NULL panic in qedi_set_path.

2017-05-24T02:16:43+00:00

RIP: 0010:qedi_set_path+0x114/0x570 [qedi]
 Call Trace:
  [] iscsi_if_recv_msg+0x623/0x14a0
  [] ? rhashtable_lookup_compare+0x36/0x70
  [] iscsi_if_rx+0x8e/0x1f0
  [] netlink_unicast+0xed/0x1b0
  [] netlink_sendmsg+0x330/0x770
  [] sock_sendmsg+0xb0/0xf0
  [] ? __switch_to+0x17b/0x4b0
  [] ? __schedule+0x2d8/0x900
  [] ___sys_sendmsg+0x3a9/0x3c0
  [] ? get_futex_key+0x1c8/0x2b0
  [] ? futex_wake+0x80/0x160

Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen

scsi: qedi: Set dma_boundary to 0xfff.

2017-05-24T02:16:43+00:00

Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen

scsi: qedi: Correctly set firmware max supported BDs.

2017-05-24T02:16:43+00:00

Signed-off-by: Manish Rangankar 
Signed-off-by: Martin K. Petersen