When the host is frozen in SCSI EH state, at any point after the LLDD
sets SAS_TASK_STATE_DONE for the sas_task task state, libsas may free
the task; see sas_scsi_find_task().
This puts the LLDD in a difficult position, in that once it sets
SAS_TASK_STATE_DONE for the task state it should not reference the
sas_task again. But the LLDD needs will check the sas_task indirectly in
calling task->task_done()->sas_scsi_task_done() or sas_ata_task_done()
(to check if the host is frozen state actually).
And the LLDD cannot set SAS_TASK_STATE_DONE for the task state after
task->task_done() is called (as the sas_task is free'd at this point).
This situation would seem to be a problem made by libsas.
To work around, check in the LLDD whether the host is in frozen state to
ensure it is ok to call task->task_done() function. If in the frozen
state, we rely on SCSI EH and libsas to free the sas_task directly.
We do not do this for the following IO types:
- SMP - they are managed in libsas directly, outside SCSI EH
- Any internally originated IO, for similar reason
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If the SCSI host enters EH, any pending IO will be processed by SCSI
EH. However it is possible that SCSI EH will try to abort the IO and
also at the same time the IO completes in the driver. In this situation
there is a small chance of freeing the sas_task twice.
Then if another IO re-uses freed sas_task before the second time of
free'ing sas_task, it is possible to free incorrect sas_task.
To avoid this situation, add some checks to increase reliability. The
sas_task task state flag SAS_TASK_STATE_ABORTED is used to mutually
protect the LLDD and libsas freeing the task.
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In the DQ tasklet processing it is not necessary to take the DQ lock, as
there is no contention between adding slots to the CQ and removing slots
from the matching DQ.
In addition, since we run each DQ in a separate tasklet context, there
would be no possible contention between DQ processing running for the
same queue in parallel.
It is still necessary to take hisi_hba lock when free'ing slots.
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Fix small formatting and wording nits in Broadcom copyright header
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Update the driver version to 12.0.0.3
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Enhance log messages for CQEs as they were not reporting certain fields.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Fix up log messages and add an fcp error stat counter in the IO submit
code path to make diagnosing problems easier
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
If the cpu count is larger than the number of WQ resources available,
adapter attachment eventually failes due to a WQ_CREATE failure.
Calculate the number of WQs desired (which initializes to cpu count)
after accounting for the number of queues the adapter supports and the
number allocated to SCSI and the control/ELS path, and scale down if
necessary.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The driver encounters a link event ACQE with a fault code it doesn't
recognize, it logs an "Invalid" fault type and futher treats the unknown
value as a mailbox command failure. First off, there is no "invalid"
value, only values that are unknown. Secondly, the fault code doesn't
indicate status - the rest of the ACQE contains that status so there is
no reason to "fail the commands".
Change the "Invalid" to "Unknown". There is no "invalid" code value.
Separate fault code parsing and message genaration from any mbx handling
status.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In situations when the firmware image in inappropriate for the chip
type, initial validation checks were light, allowing the checks to pass,
thus allowing the firmware to be downloaded. Eventually, after the
download, the chip rejects the firmware but it is logged as a generic
firmware download error.
Revise the initial checks to validate the image vs asic type so that the
correct message is displayed and the download process is avoided.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The driver builds the control structures in host memory using
definitions that are based on 32-bit words. After building the structure
it is then written to the adapter.
This patch slightly optimizes LE hosts by copying the structures via
64-bit copies. This is doable as the adapter interface is LE thus there
is no byteswapping as the copy is performed.
The same optimization would be nice on BE systems, but when byteswapping
occurs, it swaps 32-bit words as well, thus trashing the control
structure. Given amount of code that is dependent upon the 32-bit word
definition, it was decided to not change things for the minor
optimization. Thus PPC 64-bit systems sticks with doing 32-bit copies.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
I/O submission paths in the lpfc nvme path are rejecting the io with an
error code that reflects back to the callee as a hard io failure. Many
of these conditions are transient and would likely resolve if retried.
Correct by returning -EBUSY, which the FC transport triggers off of to
return busy status codes to the blk-mq layer.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
During an uplink toggle test all error handling is done via timeout and
firmware error conditions which can occur concurrently:
- SCSI layer timeouts
- Error detect CQEs
- Firmware detected underruns
- ABTS timeouts
All these concurrent events require more defensive checks in the driver
including:
- Check both internally and externally generated aborts to make sure the
xid is not already been aborted in another context or in cleanup.
- Check back pointers in qedf_cmd_timeout to verify the context of the
io_req, fcport and qedf_ctx
- Check rport state in host reset handler to not reset the whole host
if the rport is already uploaded or in the process of relogin
- Check to state for an fcport before initiating a middle path ELS
request
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Similar to what we do when we remove a PCI function, set the
QEDF_UNLOADING flag to prevent any requests from being queued while a
vport is being deleted. This prevents any requests from getting stuck
in limbo when the vport is unloaded or deleted.
Fixes the crash:
PID: 106676 TASK: ffff9a436aa90000 CPU: 12 COMMAND: "multipathd"
#0 [ffff9a43567d3550] machine_kexec+522 at ffffffffaca60b2a
#1 [ffff9a43567d35b0] __crash_kexec+114 at ffffffffacb13512
#2 [ffff9a43567d3680] crash_kexec+48 at ffffffffacb13600
#3 [ffff9a43567d3698] oops_end+168 at ffffffffad117768
#4 [ffff9a43567d36c0] no_context+645 at ffffffffad106f52
#5 [ffff9a43567d3710] __bad_area_nosemaphore+116 at ffffffffad106fe9
#6 [ffff9a43567d3760] bad_area+70 at ffffffffad107379
#7 [ffff9a43567d3788] __do_page_fault+1247 at ffffffffad11a8cf
#8 [ffff9a43567d37f0] do_page_fault+53 at ffffffffad11a915
#9 [ffff9a43567d3820] page_fault+40 at ffffffffad116768
[exception RIP: qedf_init_task+61]
RIP: ffffffffc0e13c2d RSP: ffff9a43567d38d0 RFLAGS: 00010046
RAX: 0000000000000000 RBX: ffffbe920472c738 RCX: ffff9a434fa0e3e8
RDX: ffff9a434f695280 RSI: ffffbe920472c738 RDI: ffff9a43aa359c80
RBP: ffff9a43567d3950 R8: 0000000000000c15 R9: ffff9a3fb09b9880
R10: ffff9a434fa0e3e8 R11: ffff9a43567d35ce R12: 0000000000000000
R13: ffff9a434f695280 R14: ffff9a43aa359c80 R15: ffff9a3fb9e005c0
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
There are a couple of kernel cases when we restart a remote port due to
ABTS timeout that we need to handle:
1. Flush any outstanding ABTS requests when flushing I/Os so that we do
not hold up the eh_abort handler indefinitely causing process hangs.
2. Check if we are currently uploading a connection before issuing an
ABTS.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Get all firmware debug data instead of just a grc dump.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
PROBLEM DESCRIPTION:
According to the logs, STAG was changing and it was triggering soft
reset. In soft reset we used to virtual link down and up and also we
were disabling DCBx flag. Since this was virtual link flap, DCBx never
used to converge again.
SOLUTION:
Code change is to remove disabling DCBx flag from soft reset.
Signed-off-by: Saurav Kashyap <saurav.kashyap@cavium.com>
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Helps to corroborate which requests we can't get reference on and if
it's real bug or not.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When an RRQ request times out the reference is not getting decremented
correctly as there are still ELS commands leftover when we flush any
pending I/Os during offload:
[ 281.788553] [0000:21:00.3]:[qedf_cmd_timeout:58]:4: ELS timeout, xid=0x96a.
...
[ 281.788553] [0000:21:00.3]:[qedf_cmd_timeout:58]:4: ELS timeout, xid=0x96a.
[ 281.788772] [0000:21:00.3]:[qedf_rrq_compl:182]:4: Entered.
[ 281.788774] [0000:21:00.3]:[qedf_rrq_compl:200]:4: rrq_compl: orig io = ffffc90004c556f8, orig xid = 0x81b, rrq_xid = 0x96a, refcount=1
...
[ 331.448032] [0000:21:00.3]:[qedf_flush_els_req:1512]:4: Flushing ELS request xid=0x96a refcount=2.
The fix is to call kref_put on the rrq_req in case of timeout as the
timeout handler will call rrq_compl directly vs. a normal completion
where it is call from els_compl.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
We currently hard code the priority in the 8021q tag to 3 for FCoE
traffic. The vast majority of the time this is fine but if the priority
is something else besides 3, any VLAN ID comparison either in the
non-offload path or offload path will fail and cause dropped frames
where none are expected.
Change the behavior so that the driver default is 3 if we do not get any
DCBX convergence.
If DCBX does converge, then set the FIP/FCoE priority in the following
manner:
1. If the qedf_default_prio modparam is set use that
2. If the DCBX FCoE priority is not in range (0..7) use 3
3. Use the DCBX FCoE priority we get in the driver's DCBX handler
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This module parameter is to work around cases where we do not receive
the DCBX handler notification from qed but discovery is still possible
if we send out a FIP VLAN request irregardless of the DCBX state.
[mkp: zeroday warning]
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Some configurations need more than 30 seconds to respond to a FIP VLAN
request so increase the default to 60 seconds.
Signed-off-by: Chad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
For target mode, task management command is queued to specific cpu base
on where the SCSI command is residing. This prevent race condition of
task management command getting ahead of regular scsi command.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
- Uses predefine inline function to access add_cdb_len field in ATIO.
- Return SS_RESIDUAL_UNDER status when sending BUSY
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When a connection is established, the target core session may not be
created immediately. Current code will drop/terminate the command based
on the session state. This patch will return BUSY status for any
commands arriving on wire before the session is created.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Move GPSC & GFPNID commands out of session management to reduce time lag
in reporting the session state to remote port. These commands are not
essential when it comes to maintaining the rport state. Delay sending
these commands after rport state is set to Online.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
For each RSCN that triggers a rescan of the fabric, ADISC is used to
revalidate an existing session. If the RSCN is not affecting all
existing sessions, then driver should not send redundant ADISC for all
existing sessions.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This patch fixes rport state and session state getting out of sync.
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This patch fixes login_retry login for ADISC command.
when login_retry count reaches 0, further attempt to send ADISC command
is ignored by the code. Remove this redundant login_retry count check
from qla24xx_fcport_handle_login()
[mkp: fix typo]
Signed-off-by: Quinn Tran <quinn.tran@cavium.com>
Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Update driver version to match OOB/internal driver version.
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In ioctl exit path driver refers ioc_list to free memory associated with
diag buffers and event_log pointer used to save events by driver.
If ctl_exit() func is called after unregistering driver, then ioc_list will
be empty and hence driver will not be able to free the allocated memory
which in turn causes memory leak.
So call ctl_exit() function before unregistering mpt3sas driver.
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
1) Manufacturing Page 11 contains parameters to control internal
firmware behavior. Based on AddlFlags2 field FW/Driver behaviour can
be changed, (flag tm_custom_handling is used for this)
a) For PCIe device, protocol level reset should be used if flag
tm_custom_handling is 0. Since Abort Task Set, LUN reset and Target
reset will result in a protocol level reset. Drivers should issue
only one type of this reset, if that fails then it should escalate to
a controller reset (diag reset/OCR).
b) If the driver has control over the TM reset timeout value, then
driver should use the value exposed in PCIe Device Page 2 for pcie
device (field ControllerResetTO).
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Update MPI Files to support protocol level reset for NVMe device.
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Added function _base_display_fwpkg_version, which sends FWUpload request
to pull FW package version from FW Image Header. Now driver prints FW
package version in addition to FW version if the PackageVersion is
valid.
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In function _scsih_add_device, for each device connected to an
enclosure, driver reads the enclosure page(To get details like enclosure
handle, enclosure logical ID, enclosure level etc.)
With this patch, instead of reading enclosure page everytime, driver
maintains a list for enclosure device(During enclosure add event,
enclosure device is added to the list and removed from the list on
delete events) and uses the enclosure page from the list.
Signed-off-by: Chaitra P B <chaitra.basappa@broadcom.com>
Signed-off-by: Suganath Prabu S <suganath-prabu.subramani@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
