Currently, the FW logging facility is a load/boot time parameter which
requires the driver to be unloaded/reloaded or the system rebooted in order
to change its configuration.
Convert the logging facility to allow dynamic enablement and configuration.
Specifically:
- Convert the feature so that it can be enabled dynamically via an
attribute. Additionally, the size of the buffer can be configured
dynamically.
- Add locks around states that now may be changing.
- Tie the feature into debugfs so that the logs can be read at any time.
Link: https://lore.kernel.org/r/20191018211832.7917-12-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Lower IOps performance with write operations. Perf tool shows lock
contention in dma_pool_alloc and dma_pool_free related to the
txrdy_payload_pool.
The allocations are for dma buffers for XFER_RDY's, which actually are not
needed for the FCP_TRECEIVE command as the command contents are used by the
adapter to generate the IU.
Remove the allocations and the associated buffer pool. Rather than leaving
NULLs in buffer pointer locations, set command and sgl to indicate skipped
SGLE indexes.
Link: https://lore.kernel.org/r/20191018211832.7917-10-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The nvme-fc transport may call to abort an io on controller reset. If the
driver is out of resources to issue an abort command, it just gives up and
does nothing. The transport expects the lldd to always be able to terminate
an io it has issued. At that point, the controller hangs waiting for
aborted ios to be returned. Note: flaged by "6136" and "6176" error
messages.
Root issue was the adapter mis-allocated the number resources it allocated
for command entries for the adapter.
Convert the driver to allocate command resources based on the number of
xris supported by the FC port - 1 resource for the original command and 1
resource for the abort request.
Link: https://lore.kernel.org/r/20190922035906.10977-5-jsmart2021@gmail.com
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This is mostly update of the usual drivers: qla2xxx, ufs, smartpqi,
lpfc, hisi_sas, qedf, mpt3sas; plus a whole load of minor updates.
The only core change this time around is the addition of request
batching for virtio. Since batching requires an additional flag to
use, it should be invisible to the rest of the drivers.
Signed-off-by: James E.J. Bottomley <jejb@linux.ibm.com>
-----BEGIN PGP SIGNATURE-----
iJwEABMIAEQWIQTnYEDbdso9F2cI+arnQslM7pishQUCXYQE/yYcamFtZXMuYm90
dG9tbGV5QGhhbnNlbnBhcnRuZXJzaGlwLmNvbQAKCRDnQslM7pishXs9AP4usPY5
OpMlF6OiKFNeJrCdhCScVghf9uHbc7UA6cP+EgD/bCtRgcDe1ZjOTYWdeTwvwWqA
ltWYonnv6Lg3b1f9yqI=
=jRC/
-----END PGP SIGNATURE-----
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI updates from James Bottomley:
"This is mostly update of the usual drivers: qla2xxx, ufs, smartpqi,
lpfc, hisi_sas, qedf, mpt3sas; plus a whole load of minor updates. The
only core change this time around is the addition of request batching
for virtio. Since batching requires an additional flag to use, it
should be invisible to the rest of the drivers"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (264 commits)
scsi: hisi_sas: Fix the conflict between device gone and host reset
scsi: hisi_sas: Add BIST support for phy loopback
scsi: hisi_sas: Add hisi_sas_debugfs_alloc() to centralise allocation
scsi: hisi_sas: Remove some unused function arguments
scsi: hisi_sas: Remove redundant work declaration
scsi: hisi_sas: Remove hisi_sas_hw.slot_complete
scsi: hisi_sas: Assign NCQ tag for all NCQ commands
scsi: hisi_sas: Update all the registers after suspend and resume
scsi: hisi_sas: Retry 3 times TMF IO for SAS disks when init device
scsi: hisi_sas: Remove sleep after issue phy reset if sas_smp_phy_control() fails
scsi: hisi_sas: Directly return when running I_T_nexus reset if phy disabled
scsi: hisi_sas: Use true/false as input parameter of sas_phy_reset()
scsi: hisi_sas: add debugfs auto-trigger for internal abort time out
scsi: virtio_scsi: unplug LUNs when events missed
scsi: scsi_dh_rdac: zero cdb in send_mode_select()
scsi: fcoe: fix null-ptr-deref Read in fc_release_transport
scsi: ufs-hisi: use devm_platform_ioremap_resource() to simplify code
scsi: ufshcd: use devm_platform_ioremap_resource() to simplify code
scsi: hisi_sas: use devm_platform_ioremap_resource() to simplify code
scsi: ufs: Use kmemdup in ufshcd_read_string_desc()
...
Capturing and downloading dif command data and dif data was done a dozen
years ago and no longer being used. Also creates a potential security hole.
Remove the debugfs buffer for dif debugging.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
CC: KyleMahlkuch <kmahlkuc@linux.vnet.ibm.com>
CC: Hannes Reinecke <hare@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Currently, each hardware queue, typically allocated per-cpu, consists of a
WQ/CQ pair per protocol. Meaning if both SCSI and NVMe are supported 2
WQ/CQ pairs will exist for the hardware queue. Separate queues are
unnecessary. The current implementation wastes memory backing the 2nd set
of queues, and the use of double the SLI-4 WQ/CQ's means less hardware
queues can be supported which means there may not always be enough to have
a pair per cpu. If there is only 1 pair per cpu, more cpu's may get their
own WQ/CQ.
Rework the implementation to use a single WQ/CQ pair by both protocols.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
FC-NVMe-2 added support for sequence level error recovery in the FC-NVME
protocol. This allows for the detection of errors and lost frames and
immediate retransmission of data to avoid exchange termination, which
escalates into NVMeoFC connection and association failures. A significant
RAS improvement.
The driver is modified to indicate support for SLER in the NVMe PRLI is
issues and to check for support in the PRLI response. When both sides
support it, the driver will set a bit in the WQE to enable the recovery
behavior on the exchange. The adapter will take care of all detection and
retransmission.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Typical SLI-4 hardware supports up to 2 4KB pages to be registered per XRI
to contain the exchanges Scatter/Gather List. This caps the number of SGL
elements that can be in the SGL. There are not extensions to extend the
list out of the 2 pages.
The G7 hardware adds a SGE type that allows the SGL to be vectored to a
different scatter/gather list segment. And that segment can contain a SGE
to go to another segment and so on. The initial segment must still be
pre-registered for the XRI, but it can be a much smaller amount (256Bytes)
as it can now be dynamically grown. This much smaller allocation can
handle the SG list for most normal I/O, and the dynamic aspect allows it to
support many MB's if needed.
The implementation creates a pool which contains "segments" and which is
initially sized to hold the initial small segment per xri. If an I/O
requires additional segments, they are allocated from the pool. If the
pool has no more segments, the pool is grown based on what is now
needed. After the I/O completes, the additional segments are returned to
the pool for use by other I/Os. Once allocated, the additional segments are
not released under the assumption of "if needed once, it will be needed
again". Pools are kept on a per-hardware queue basis, which is typically
1:1 per cpu, but may be shared by multiple cpus.
The switch to the smaller initial allocation significantly reduces the
memory footprint of the driver (which only grows if large ios are
issued). Based on the several K of XRIs for the adapter, the 8KB->256B
reduction can conserve 32MBs or more.
It has been observed with per-cpu resource pools that allocating a resource
on CPU A, may be put back on CPU B. While the get routines are distributed
evenly, only a limited subset of CPUs may be handling the put routines.
This can put a strain on the lpfc_put_cmd_rsp_buf_per_cpu routine because
all the resources are being put on a limited subset of CPUs.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When SCSI-MQ is enabled, the SCSI-MQ layers will do pre-allocation of MQ
resources based on shost values set by the driver. In newer cases of the
driver, which attempts to set nr_hw_queues to the cpu count, the
multipliers become excessive, with a single shost having SCSI-MQ
pre-allocation reaching into the multiple GBytes range. NPIV, which
creates additional shosts, only multiply this overhead. On lower-memory
systems, this can exhaust system memory very quickly, resulting in a system
crash or failures in the driver or elsewhere due to low memory conditions.
After testing several scenarios, the situation can be mitigated by limiting
the value set in shost->nr_hw_queues to 4. Although the shost values were
changed, the driver still had per-cpu hardware queues of its own that
allowed parallelization per-cpu. Testing revealed that even with the
smallish number for nr_hw_queues for SCSI-MQ, performance levels remained
near maximum with the within-driver affiinitization.
A module parameter was created to allow the value set for the nr_hw_queues
to be tunable.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
To support scenarios which aren't bound to nvmetcli add port scenarios,
which is currently where the nvmet_fc transport invokes the discovery
event callbacks, a syfs attribute is added to lpfc which can be written
to cause an RSCN to be generated for the nport.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This patch adds general RSCN support:
- The ability to transmit an RSCN to the port on the other end of
the link (regular port if pt2pt, or fabric controller if fabric).
- And general recognition of an RSCN ELS when an ELS is received.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The driver is currently reporting the firmware revision not the actual boot
bios version in FDMI data.
Modify the driver to obtain the boot bios version from the adapter and use
that data in the FMDI data sent to the switch.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
For files modified as part of 12.2.0.0 patches, update copyright to 2019
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When the transport calls into the lpfc target to release an IO job
structure, which corresponds to an exchange, and if the driver was waiting
for an exchange in order to post a previously received command to the
transport, the driver immediately takes the IO job and reuses the context
for the prior command and calls nvmet_fc_rcv_fcp_req() to tell the
transport about a newly received command.
Problem is, the execution of the IO job release may be in the context of
the back end driver and its bio completion handlers, thus it may be in a
irq context and protection code kicks in in the bio and request layers that
are subsequently called.
Rework lpfc so that instead of immediately upcalling, queue it to a
deferred work thread and have the thread make the upcall.
Took advantage of this change to remove duplicated code with the normal
command receive path that preps the IO job and upcalls nvmet_fc. Created a
common routine both paths use.
Also corrected some errors that were found during review of the context
freeing and reuse - basically unlocked operations and a somewhat disjoint
set of calls to release associated job elements. Cleaned up this path and
added locks for coherency.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The conversion to enable SCSI and NVME fc4 support ran into an issue with
NPIV support. With NVME, NPIV is not currently supported, but with SCSI it
was. The driver reverted to its lowest setting meaning NPIV with SCSI was
not allowed.
Convert the NPIV checks and implementation so that SCSI can continue to
allow NPIV support.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When driving high iop counts, auto_imax coalescing kicks in and drives the
performance to extremely small iops levels.
There are two issues:
1) auto_imax is enabled by default. The auto algorithm, when iops gets
high, divides the iops by the hdwq count and uses that value to
calculate EQ_Delay. The EQ_Delay is set uniformly on all EQs whether
they have load or not. The EQ_delay is only manipulated every 5s (a
long time). Thus there were large 5s swings of no interrupt delay
followed by large/maximum delay, before repeating.
2) When processing a CQ, the driver got mixed up on the rate of when
to ring the doorbell to keep the chip appraised of the eqe or cqe
consumption as well as how how long to sit in the thread and
process queue entries. Currently, the driver capped its work at
64 entries (very small) and exited/rearmed the CQ. Thus, on heavy
loads, additional overheads were taken to exit and re-enter the
interrupt handler. Worse, if in the large/maximum coalescing
windows,k it could be a while before getting back to servicing.
The issues are corrected by the following:
- A change in defaults. Auto_imax is turned OFF and fcp_imax is set
to 0. Thus all interrupts are immediate.
- Cleanup of field names and their meanings. Existing names were
non-intuitive or used for duplicate things.
- Added max_proc_limit field, to control the length of time the
handlers would service completions.
- Reworked EQ handling:
Added common routine that walks eq, applying notify interval and max
processing limits. Use queue_claimed to claim ownership of the queue
while processing. Always rearm the queue whenever the common routine
is called.
Rework queue element processing, namely to eliminate hba_index vs
host_index. Only one index is necessary. The queue entry can be
marked invalid and the host_index updated immediately after eqe
processing.
After rework, xx_release routines are now DB write functions. Renamed
the routines as such.
Moved lpfc_sli4_eq_flush(), which does similar action, to same area.
Replaced the 2 individual loops that walk an eq with a call to the
common routine.
Slightly revised lpfc_sli4_hba_handle_eqe() calling syntax.
Added per-cpu counters to detect interrupt rates and scale
interrupt coalescing values.
- Reworked CQ handling:
Added common routine that walks cq, applying notify interval and max
processing limits. Use queue_claimed to claim ownership of the queue
while processing. Always rearm the queue whenever the common routine
is called.
Rework queue element processing, namely to eliminate hba_index vs
host_index. Only one index is necessary. The queue entry can be
marked invalid and the host_index updated immediately after cqe
processing.
After rework, xx_release routines are now DB write functions. Renamed
the routines as such.
Replaced the 3 individual loops that walk a cq with a call to the
common routine.
Redefined lpfc_sli4_sp_handle_mcqe() to commong handler definition with
queue reference. Add increment for mbox completion to handler.
- Added a new module/sysfs attribute: lpfc_cq_max_proc_limit To allow
dynamic changing of the CQ max_proc_limit value being used.
Although this leaves an EQ as an immediate interrupt, that interrupt will
only occur if a CQ bound to it is in an armed state and has cqe's to
process. By staying in the cq processing routine longer, high loads will
avoid generating more interrupts as they will only rearm as the processing
thread exits. The immediately interrupt is also beneficial to idle or
lower-processing CQ's as they get serviced immediately without being
penalized by sharing an EQ with a more loaded CQ.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
So far MSIX vector allocation assumed it would be 1:1 with hardware
queues. However, there are several reasons why fewer MSIX vectors may be
allocated than hardware queues such as the platform being out of vectors or
adapter limits being less than cpu count.
This patch reworks the MSIX/EQ relationships with the per-cpu hardware
queues so they can function independently. MSIX vectors will be equitably
split been cpu sockets/cores and then the per-cpu hardware queues will be
mapped to the vectors most efficient for them.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The XRI get/put lists were partitioned per hardware queue. However, the
adapter rarely had sufficient resources to give a large number of resources
per queue. As such, it became common for a cpu to encounter a lack of XRI
resource and request the upper io stack to retry after returning a BUSY
condition. This occurred even though other cpus were idle and not using
their resources.
Create as efficient a scheme as possible to move resources to the cpus that
need them. Each cpu maintains a small private pool which it allocates from
for io. There is a watermark that the cpu attempts to keep in the private
pool. The private pool, when empty, pulls from a global pool from the
cpu. When the cpu's global pool is empty it will pull from other cpu's
global pool. As there many cpu global pools (1 per cpu or hardware queue
count) and as each cpu selects what cpu to pull from at different rates and
at different times, it creates a radomizing effect that minimizes the
number of cpu's that will contend with each other when the steal XRI's from
another cpu's global pool.
On io completion, a cpu will push the XRI back on to its private pool. A
watermark level is maintained for the private pool such that when it is
exceeded it will move XRI's to the CPU global pool so that other cpu's may
allocate them.
On NVME, as heartbeat commands are critical to get placed on the wire, a
single expedite pool is maintained. When a heartbeat is to be sent, it will
allocate an XRI from the expedite pool rather than the normal cpu
private/global pools. On any io completion, if a reduction in the expedite
pools is seen, it will be replenished before the XRI is placed on the cpu
private pool.
Statistics are added to aid understanding the XRI levels on each cpu and
their behaviors.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Now that the lower half has much better per-cpu parallelization using the
hardware queues, the SCSI MQ support needs to be tied into it.
The involves the following mods:
- Use the hardware queue info from the midlayer to help select the
hardware queue to utilize. This required change to the get_scsi-buf_xxx
routines.
- Remove lpfc_sli4_scmd_to_wqidx_distr() routine. No longer needed.
- Includes fix for SLI-3 that does not have multi queue parallelization.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Many io statistics were being sampled and saved using adapter-based data
structures. This was creating a lot of contention and cache thrashing in
the I/O path.
Move the statistics to the hardware queue data structures. Given the
per-queue data structures, use of atomic types is lessened.
Add new sysfs and debugfs stat routines to collate the per hardware queue
values and report at an adapter level.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Similar to the io execution path that reports cpu context information, the
debugfs routines for cpu information needs to be aligned with new hardware
queue implementation.
Convert debugfs cnd nvme cpucheck statistics to report information per
Hardware Queue.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Both NVME and SCSI aborts are now processed off the CQ workqueue and do not
generate events for the slowpath any more.
Remove the unused event code.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Once the IO buff allocations were made shared, there was a single XRI
buffer list shared by all hardware queues. A single list isn't great for
performance when shared across the per-cpu hardware queues.
Create a separate XRI IO buffer get/put list for each Hardware Queue. As
SGLs and associated IO buffers get allocated/posted to the firmware; round
robin their assignment across all available hardware Queues so that there
is an equitable assignment.
Modify SCSI and NVME IO submit code paths to use the Hardware Queue logic
for XRI allocation.
Add a debugfs interface to display hardware queue statistics
Added new empty_io_bufs counter to track if a cpu runs out of XRIs.
Replace common_ variables/names with io_ to make meanings clearer.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Currently, both nvme and fcp each have their own concept of an io_channel,
which is a combination wq/cq and associated msix. Different cpus would
share an io_channel.
The driver is now moving to per-cpu wq/cq pairs and msix vectors. The
driver will still use separate wq/cq pairs per protocol on each cpu, but
the protocols will share the msix vector.
Given the elimination of the nvme and fcp io channels, the module
parameters will be removed. A new parameter, lpfc_hdw_queue is added which
allows the wq/cq pair allocation per cpu to be overridden and allocated to
lesser value. If lpfc_hdw_queue is zero, the number of pairs allocated will
be based on the number of cpus. If non-zero, the parameter specifies the
number of queues to allocate. At this time, the maximum non-zero value is
64.
To manage this new paradigm, a new hardware queue structure is created to
track queue activity and relationships.
As MSIX vector allocation must be known before setting up the
relationships, msix allocation now occurs before queue datastructures are
allocated. If the number of vectors allocated is less than the desired
hardware queues, the hardware queue counts will be reduced to the number of
vectors
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Currently, both NVME and SCSI get their IO buffers from separate
pools. XRI's are associated 1:1 with IO buffers, so XRI's are also split
between protocols.
Eliminate the independent pools and use a single pool. Each buffer
structure now has a common section and a protocol section. Per protocol
routines for SGL initialization are removed and replaced by common
routines. Initialization of the buffers is only done on the common area.
All other fields, which are protocol specific, are initialized when the
buffer is allocated for use in the per-protocol allocation routine.
In the past, the SCSI side allocated IO buffers as part of slave_alloc
calls until the maximum XRIs for SCSI was reached. As all XRIs are now
common and may be used for either protocol, allocation for everything is
done as part of adapter initialization and the scsi side has no action in
slave alloc.
As XRI's are no longer split, the lpfc_xri_split module parameter is
removed.
Adapters based on SLI3 will continue to use the older scsi_buf_list_get/put
routines. All SLI4 adapters utilize the new IO buffer scheme
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The current discovery state machine the driver treated FLOGI oddly. When
point to point, an FLOGI is to be exchanged by the two ports, with the port
with the most significant WWN then proceeding with PLOGI. The
implementation in the driver was keyed to closely with "what have I sent",
not with what has happened between the two endpoints. Thus, it blatantly
would ACC an FLOGI, but reject PLOGI's until it had its FLOGI ACC'd. The
problem is - the sending of FLOGI may be delayed for some reason, or the
response to FLOGI held off by the other side. In the failing situation the
other side sent an FLOGI, which was ACC'd, then sent PLOGIs which were then
rjt'd until the retry count for the PLOGIs were exceeded and the port gave
up. The FLOGI may have been very late in transmit, or the response held off
until the PLOGIs failed. Given the other port had the higher WWN, no PLOGIs
would occur and communication stopped.
Correct the situation by changing the FLOGI handling. Defer any response to
an FLOGI until the driver has sent its FLOGI as well. Then, upon either
completion of the sent FLOGI, or upon sending an ACC to a received FLOGI
(which may be received before or just after FLOGI was sent). the driver
will act on who has the higher WWN. if the other port does, the driver will
noop any handling of an FLOGI response (if outstanding) and wait for PLOGI.
If the local port does, the driver will transition to sending PLOGI and
will noop any action on responding to an FLOGI (if not yet received).
Fortunately, to implement this, it only took another state flag and
deferring any FLOGI response if the FLOGI has yet to be transmit. All
subsequent actions were already in place.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Depending on the chipset, the number of NPIV vports may vary and be in
excess of what most switches support (256). To avoid confusion with the
users, limit the reported NPIV vports to 256.
Additionally correct the 16G adapter which is reporting a bogus NPIV vport
number if the link is down.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Driver is hitting null pring pointers in lpfc_do_work().
Pointer assignment occurs based on SLI-revision. If recovering after an
error, its possible the sli revision for the port was cleared, making the
lpfc_phba_elsring() not return a ring pointer, thus the null pointer.
Add SLI revision checking to lpfc_phba_elsring() and status checking to all
callers.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add trunking support to the driver. Trunking is found on more recent
asics. In general, trunking appears as a single "port" to the driver
and overall behavior doesn't differ. Link speed is reported as an
aggregate value, while link speed control is done on a per-physical
link basis with all links in the trunk symmetrical. Some commands
returning port information are updated to additionally provide
trunking information. And new ACQEs are generated to report physical
link events relative to the trunk.
This patch contains the following modifications:
- Added link speed settings of 128GB and 256GB.
- Added handling of trunk-related ACQEs, mainly logging and trapping
of physical link statuses.
- Added additional bsg interface to query trunk state by applications.
- Augment link_state sysfs attribtute to display trunk link status
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The switches seem to respond faster to GID_PT vs GID_FT NameServer
queries. Add support for GID_PT to be used over GID_FT to enable
faster storage failover detection. Includes addition of new module
parameter to select between GID_PT and GID_FT (GID_FT is default).
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Testing a point-to-point topology and a case of re-FLOGI without
intervening link bouncing, showed an odd interaction with firmware and
a resulting scenario where the driver no longer probed after accepting
the new FLOGI.
Work around the firmware issue by issuing a link bounce if a FLOGI is
received after the link is already up and FLOGI's accepted.
While debugging the issue, realized that some debug traces should be
clarified to help in the future.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This is mostly updates of the usual drivers: UFS, esp_scsi, NCR5380,
qla2xxx, lpfc, libsas, hisi_sas. In addition there's a set of mostly
small updates to the target subsystem a set of conversions to the
generic DMA API, which do have some potential for issues in the older
drivers but we'll handle those as case by case fixes. A new myrs for
the DAC960/mylex raid controllers to replace the block based DAC960
which is also being removed by Jens in this merge window. Plus the
usual slew of trivial changes.
Signed-off-by: James E.J. Bottomley <jejb@linux.ibm.com>
-----BEGIN PGP SIGNATURE-----
iJwEABMIAEQWIQTnYEDbdso9F2cI+arnQslM7pishQUCW9BQJSYcamFtZXMuYm90
dG9tbGV5QGhhbnNlbnBhcnRuZXJzaGlwLmNvbQAKCRDnQslM7pishU3MAP41T8yW
UJQDCprj65pCR+9mOUWzgMvgAW/15ouK89x/7AD/XAEQZqoAgpFUbgnoZWGddZkS
LykIzSiLHP4qeDOh1TQ=
=2JMU
-----END PGP SIGNATURE-----
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI updates from James Bottomley:
"This is mostly updates of the usual drivers: UFS, esp_scsi, NCR5380,
qla2xxx, lpfc, libsas, hisi_sas.
In addition there's a set of mostly small updates to the target
subsystem a set of conversions to the generic DMA API, which do have
some potential for issues in the older drivers but we'll handle those
as case by case fixes.
A new myrs driver for the DAC960/mylex raid controllers to replace the
block based DAC960 which is also being removed by Jens in this merge
window.
Plus the usual slew of trivial changes"
[ "myrs" stands for "MYlex Raid Scsi". Obviously. Silly of me to even
wonder. There's also a "myrb" driver, where the 'b' stands for
'block'. Truly, somebody has got mad naming skillz. - Linus ]
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (237 commits)
scsi: myrs: Fix the processor absent message in processor_show()
scsi: myrs: Fix a logical vs bitwise bug
scsi: hisi_sas: Fix NULL pointer dereference
scsi: myrs: fix build failure on 32 bit
scsi: fnic: replace gross legacy tag hack with blk-mq hack
scsi: mesh: switch to generic DMA API
scsi: ips: switch to generic DMA API
scsi: smartpqi: fully convert to the generic DMA API
scsi: vmw_pscsi: switch to generic DMA API
scsi: snic: switch to generic DMA API
scsi: qla4xxx: fully convert to the generic DMA API
scsi: qla2xxx: fully convert to the generic DMA API
scsi: qla1280: switch to generic DMA API
scsi: qedi: fully convert to the generic DMA API
scsi: qedf: fully convert to the generic DMA API
scsi: pm8001: switch to generic DMA API
scsi: nsp32: switch to generic DMA API
scsi: mvsas: fully convert to the generic DMA API
scsi: mvumi: switch to generic DMA API
scsi: mpt3sas: switch to generic DMA API
...
This patch adds the ability to read firmware logs from the adapter. The driver
registers a buffer with the adapter that is then written to by the adapter.
The adapter posts CQEs to indicate content updates in the buffer. While the
adapter is writing to the buffer in a circular fashion, an application will
poll the driver to read the next amount of log data from the buffer.
Driver log buffer size is configurable via the ras_fwlog_buffsize sysfs
attribute. Verbosity to be used by firmware when logging to host memory is
controlled through the ras_fwlog_level attribute. The ras_fwlog_func
attribute enables or disables loggy by firmware.
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>
When taking the board offline while performing i/o, unsafe locking errors
occurred and irq level isn't properly managed.
In lpfc_sli_hba_down, spin_lock_irqsave(&phba->hbalock, flags) does not
disable softirqs raised from timer expiry. It is possible that a softirq is
raised from the lpfc_els_retry_delay routine and recursively requests the same
phba->hbalock spinlock causing deadlock.
Address the deadlocks by creating a new port_list lock. The softirq behavior
can then be managed a level deeper into the calling sequences.
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 allocates a sg list per io struture based on a fixed maximum
size. When it registers with the protocol transports and indicates the max sg
list size it supports, the driver manipulates the fixed value to report a
lesser amount so that it has reserved space for sg elements that are used for
DIF.
The driver initialization path sets the cfg_sg_seg_cnt field to the
manipulated value for scsi. NVME initialization ran afterward and capped it's
maximum by the manipulated value for SCSI. This erroneously made NVME report
the SCSI-reduce-for-DIF value that reduced the max io size for nvme and wasted
sg elements.
Rework the driver so that cfg_sg_seg_cnt becomes the overall maximum size and
allow the max size to be tunable. A separate (new) scsi sg count is then
setup with the scsi-modified reduced value. NVME then initializes based off
the overall maximum.
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>
A recent change added some MDS processing in the lpfc_drain_txq routine
that relies on the fcp_wq being allocated. For nvmet operation the fcp_wq
is not allocated because it can only be an nvme-target. When the original
MDS support was added LS_MDS_LOOPBACK was defined wrong, (0x16) it should
have been 0x10 (decimal value used for hex setting). This incorrect value
allowed MDS_LOOPBACK to be set simultaneously with LS_NPIV_FAB_SUPPORTED,
causing the driver to crash when it accesses the non-existent fcp_wq.
Correct the bad value setting for LS_MDS_LOOPBACK.
Fixes: ae9e28f36a ("lpfc: Add MDS Diagnostic support.")
Cc: <stable@vger.kernel.org> # v4.12+
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Tested-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Change references from "Broadcom Limited" to "Broadcom Inc." in the
copyright message. Update copyright duration if not yet updated for 2018.
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 PBDE optimizations aren't supported in all firmware revs.
Make optimizations configurable in case there's a side effect on old
firmware.
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>
Under large io load, the current sizing of asynchronous buffer counts
could be exceeded, indicated by a 2885 log message:
2885 Port Status Event: port status reg 0x81800000, port smphr
reg 0xc000, error 1=0x52004a01, error 2=0x0
Enlarge the async receive queue size. Allow for a configurable number
of buffers to be posted to each RQ, using the new attribute
lpfc_nvmet_mrq_post.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When debugging various issues, per IO channel IO statistics were useful
to understand what was happening. However, many of the stats were on a
port basis rather than an io channel basis.
Move statistics to an io channel basis.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The max_scsicmpl_time parameter can be used to perform scsi cmd queue
depth mgmt based on io completion time: the queue depth is reduced to
make completion time shorter. However, as soon as an io completes and
the completion time is within limits, the code immediately bumps the
queue depth limit back up to the target queue depth. Thus the procedure
restarts, effectively limiting the usefulness of adjusting queue depth
to help completion time.
This patch makes the following changes:
- Removes the code at io completion that resets the queue depth as soon
as within limits.
- As the code removed was where the target queue depth was first
applied, change target queue depth application so that it occurs when
the parameter is changed.
- Makes target queue depth a standard parameter: both a module
parameter and a sysfs parameter.
- Optimizes the command pending count by using atomics rather than
locks.
- Updates the debugfs nodelist stats to allow better debugging of
pending command counts.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The hardware offload for NVME commands was created when the
FC-NVME standard was setting SGL Descriptor Type to SGL Data
Block Descriptor (0h) and SGL Descriptor Sub Type to Address (0h).
A late change in NVMe-over-Fabrics obsoleted these values, creating
a transport SGL descriptor type with new values to go into these
fields.
For initial hardware support, in order to be compliant to the spec,
use host-supplied cmd IU buffers instead of the adapter generated
values. Later hardware will correct this.
Add a module parameter to override this offload disablement if looking
for lowest latency. This is reasonable as nothing in FC-NVME uses
the SQE SGL values.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The current driver isn't taking advantage of a performance hint whereby
the initial data buffer descriptor can be placed in the WQE as well as
the SGL.
Add the logic to detect support for the feature and to use it when
supported.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The G7 adapter supports 64G link speeds. Add support to the driver.
In addition, a small cleanup to replace the odd bitmap logic with
a switch case.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
New if_type=6 adapters support an additional BAR that provides
apertures to allow direct WQE to adapter push support - termed
Direct Packet Push (DPP). WQ creation differs slightly to ask for
a WQ to be DPP-ized. When submitting a WQE to a DPP WQ, it is
submitted to the host memory for the WQ normally, but is also
written by the host cpu directly to a BAR aperture. Write buffer
coalescing in hardware is (hopefully) turned on, enabling single
pci write operation support. The doorbell is thing rung to indicate
the WQE is available and was pushed to the aperture.
This patch:
- Updates the WQ Create commands for the DPP options
- Adds the bar mapping for if_type=6 DPP bar
- Adds the WQE pushing to the DDP aperture received from WQ create
- Adds a new module parameter to disable DPP operation if desired.
Default is enabled.
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>
Updated Copyright in files updated 11.4.0.7
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>
Increased CQ and WQ sizes for SCSI FCP, matching those used for NVMe
development.
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>
NVME targets appear to randomly disconnect from the initiator when
running heavy IO.
The error is due to the host aggregate (across all controllers) io load
was beyond the maximum exchange count for nvme on the adapter. The
driver was properly returning a resource busy status, but the io load
was so great heartbeat commands would be bounced and not have a
successful retry within the fuzz amount for the nvme heartbeat (yes, a
very high io load!). Thus the target was terminating the controller due
to a keep alive failure.
Resolve by reserving a few exchanges (by counters) which can be used
when the adapter is out of normal exchanges and the command is a NVME
heartbeat command. As counters are used, while the reserved command is
outstanding, as soon as any other exchange completes, the counters are
adjusted and the reserved count is replenished. The heartbeat completes
execution in a normal fashion.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The logic for sg_seg_cnt is a bit convoluted. This patch tries to clean
up a couple of areas, especially around the +2 and +1 logic.
This patch:
- Cleans up the lpfc_sg_seg_cnt attribute to specify a real minimum
rather than making the minimum be whatever the default is.
- Removes the hardcoding of +2 (for the number of elements we use in a
sgl for cmd iu and rsp iu) and +1 (an additional entry to compensate
for nvme's reduction of io size based on a possible partial page)
logic in sg list initialization. In the case where the +1 logic is
referenced in host and target io checks, use the values set in the
transport template as that value was properly set.
There can certainly be more done in this area and it will be addressed
in combined host/target driver effort.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Raise the maximum NVME sg list size allowed to 256 elements.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
