| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
SLI4 nvme functions are passing the SLI3 ring number when posting wqe to
hardware. This should be indicating the hardware queue to use, not the ring
number.
Replace ring number with the hardware queue that should be used.
Note: SCSI avoided this issue as it utilized an older lfpc_issue_iocb
routine that properly adapts.
Signed-off-by: Dick Kennedy <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
There is a extra queue and msix vector for expresslane. Now that the driver
will be doing queues per cpu, this oddball queue is no longer needed.
Expresslane will utilize the normal per-cpu queues.
Updated debugfs sli4 queue output to go along with the change
Signed-off-by: Dick Kennedy <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
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 <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
lpfc_nvme_prep_io_cmd() checks for null pnode, but caller
lpfc_nvme_fcp_io_submit() has already ensured it's non-null.
Remove the pnode null check.
Signed-off-by: Dick Kennedy <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
An hba-wide lock is taken in the nvme io completion routine. The lock
covers null'ing of the nrport pointer in the cmd structure.
The nrport member isn't necessary. After extracting the pointer from the
command, the pointer was dereferenced to get the fc discovery node
pointer. But the fc discovery node pointer is alrady in the command
structure so the dereferrence was unnecessary.
Eliminated the nrport structure member and its use, which also eliminates
the port-wide lock.
Signed-off-by: Dick Kennedy <[email protected]>
Signed-off-by: James Smart <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch removes unnecessary code to handle RSCN, instead performs full
scan everytime driver receives RSCN
Fixes: d4f7a16aeca6f ("scsi: qla2xxx: Remove ASYNC GIDPN switch command")
Cc: [email protected] #4.19
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
For eh_bus_reset, driver is supposed to reset the link. Current option to
reset the link is applicable to Loop only. This patch updates current FW
option with the one that is applicable to all topologies.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
When loop was made down explicitly due to cable pull, then for N2N toplogy,
if FAWWPN BIT is enabled by user, then it would restore some default
(garbage) value for Physical port WWPN, so this show garbage WWPN for the
port. Fix is, to restore physical port WWPN, if it is fabric
configuration. When loop is explicitly made down, and FAWWPN feature is
enabled, then driver need to restore original flashed WWPN.
Signed-off-by: Sawan Chandak <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch fixes memory leak by releasing DMA memory in case CT request and
response allocation fails.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch fixes SRB allocation flag from GFP_KERNEL to GFP_ATOMIC, to
prevent sleeping in IRQ context
Signed-off-by: Giridhar Malavali <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch flushes del_work and free_work while sending NACK response for
PRLI
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch allocates DMA memory to prevent NULL pointer access for ct_sns
request while sending switch commands.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
On Transmit respond in target mode, if the chip is already reset or the
session is already deleted, then advance the command to the free step.
There is no need to abort the command, because the chip has already flushed
it.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
On session cleanup, either an implicit LOGO or an implicit PRLO is used to
flush IOs. If the flush command hit Queue Full condition, then it is
dropped. This patch adds retry code to prevent command drop.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Change default ZIO threshold to an optimized value.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch provide call back functions to stop the chip and resume the chip
if the PCI lower level driver wants to perform function level reset/FLR.
Before the FLR, the chip will be stopped to turn off all DMA activity.
After the FLR, the chip is reset to resume previous operation state.
Signed-off-by: Quinn Tran <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch fixes the issue where Dell-EMC Target will fail to discover LUNs
if domain and area of port ID is not same as adapter's.
Signed-off-by: Himanshu Madhani <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
No users left.
Signed-off-by: Christoph Hellwig <[email protected]>
Acked-by: Jens Axboe <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
No more need in a blk-mq world where the scsi command and request are
allocated together.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jens Axboe <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
No real need for bidi support once the OSD code is gone.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jens Axboe <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Now that all the users are gone the SCSI OSD library can be removed as
well.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Jens Axboe <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
We can just stash away the second request in struct bsg_job instead of
using the block layer req->next_rq field, allowing for the eventual removal
of the latter.
Signed-off-by: Christoph Hellwig <[email protected]>
Acked-by: Jens Axboe <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Updated driver version to 27.102.00.00 from 27.101.00.00.
Signed-off-by: Suganath Prabu S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add Atlas PCIe Switch Management Port device PNPID,
Vendor Id: 0x1000
device Id: 0x00B2
This device is based on MPI 2.6 spec and it exposes one SES device to
accept management commands for the PCIe switch.
Signed-off-by: Suganath Prabu S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Added device ID for NVMe Switch Adapter (Ambrosia).
VID: 0x1000
DID: 0x02B1
Signed-off-by: Suganath Prabu S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
MPI Endpoint is a PCIe switch based on MPI2.
Renaming device ID macro from MPI2_MFGPAGE_DEVID_SAS2308_MPI_EP to
MPI2_MFGPAGE_DEVID_SWITCH_MPI_EP.
Signed-off-by: Suganath Prabu S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch adds support for the new DEVICE_LIST DCMD.
Driver currently sends two separate DCMDs for getting the list of PDs and
LDs that are exposed to host. The new DCMD provides a single interface to
get a list of both PDs and LDs that are exposed to the host. Based on the
list of target IDs that are returned by this DCMD, driver will add the
devices (PD/LD) to SML. Driver will check for FW support for this new DCMD
and based on the support will either send the new DCMD or will fall back to
the earlier method of sending two separate DCMDs for PD and LD list.
Signed-off-by: Kashyap Desai <[email protected]>
Signed-off-by: Shivasharan S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
In preparation of adding support for the DEVICE_LIST DCMD, this patch
refactors the code in the AEN event handling path. Add new function to
update the PD and LD list in driver. Move the code to scan PD and VD
channels into separate function.
Signed-off-by: Shivasharan S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
During FW init, combine the code to get the PD and LD list from FW into a
single function. This patch is in preparation for adding support for
HOST_DEVICE_LIST DCMD.
Signed-off-by: Shivasharan S <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Presently when an error is encountered during probe of the cxlflash
adapter, a deadlock is seen with cpu thread stuck inside
cxlflash_remove(). Below is the trace of the deadlock as logged by
khungtaskd:
cxlflash 0006:00:00.0: cxlflash_probe: init_afu failed rc=-16
INFO: task kworker/80:1:890 blocked for more than 120 seconds.
Not tainted 5.0.0-rc4-capi2-kexec+ #2
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/80:1 D 0 890 2 0x00000808
Workqueue: events work_for_cpu_fn
Call Trace:
0x4d72136320 (unreliable)
__switch_to+0x2cc/0x460
__schedule+0x2bc/0xac0
schedule+0x40/0xb0
cxlflash_remove+0xec/0x640 [cxlflash]
cxlflash_probe+0x370/0x8f0 [cxlflash]
local_pci_probe+0x6c/0x140
work_for_cpu_fn+0x38/0x60
process_one_work+0x260/0x530
worker_thread+0x280/0x5d0
kthread+0x1a8/0x1b0
ret_from_kernel_thread+0x5c/0x80
INFO: task systemd-udevd:5160 blocked for more than 120 seconds.
The deadlock occurs as cxlflash_remove() is called from cxlflash_probe()
without setting 'cxlflash_cfg->state' to STATE_PROBED and the probe thread
starts to wait on 'cxlflash_cfg->reset_waitq'. Since the device was never
successfully probed the 'cxlflash_cfg->state' never changes from
STATE_PROBING hence the deadlock occurs.
We fix this deadlock by setting the variable 'cxlflash_cfg->state' to
STATE_PROBED in case an error occurs during cxlflash_probe() and just
before calling cxlflash_remove().
Cc: [email protected]
Fixes: c21e0bbfc485("cxlflash: Base support for IBM CXL Flash Adapter")
Signed-off-by: Vaibhav Jain <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
There are several issues with badly indented statements. Fix these
and clean up the formatting.
Signed-off-by: Colin Ian King <[email protected]>
Acked-by: Mahesh Rajashekhara <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This reverts commit bbc0f8bd88abefb0f27998f40a073634a3a2db89.
It added a warning whose intent was to check whether the rport was still
linked into the peer list. It doesn't work as intended and gives false
positive warnings for two reasons:
1) If the rport is never linked into the peer list it will not be
considered empty since the list_head is never initialized.
2) If the rport is deleted from the peer list using list_del_rcu(), then
the list_head is in an undefined state and it is not considered empty.
Signed-off-by: Ross Lagerwall <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Commit bf5054569653 ("block: Introduce blk_revalidate_disk_zones()")
inadvertently broke the message output of sd_zbc_print_zones() because the
zone information initialization of the scsi disk structure was moved to the
second scan run while sd_zbc_print_zones() is called on the first
scan. This leads to the following incorrect message to be printed for any
ZBC or ZAC zoned disks.
"...[sdX] 4294967295 zones of 0 logical blocks + 1 runt zone"
Fix this by initializing sdkp zone size and number of zones early on the
first scan. This does not impact the execution of
blk_revalidate_zones(). This functions is still called only once the block
device capacity is set on the second revalidate run on boot, or if the disk
zone configuration changed (i.e. the disk changed).
Fixes: bf5054569653 ("block: Introduce blk_revalidate_disk_zones()")
Cc: [email protected]
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
There is no need to call twice string_get_size() when the capacity messages
are not going to be printed. Reverse the message output condition to return
early and avoid executing string_get_size() when it is not necessary.
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
uic errors are currently printed out of time order.
Make things more readable by printing logs in time order, and printing
"No record" if history is empty.
Signed-off-by: Stanley Chu <[email protected]>
Reviewed-by: Avri Altman <[email protected]>
Reviewed-by: Bean Huo <[email protected]>
Tested-by: Bean Huo <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Due to the patch that makes TMF handling synchronous the
write_pending_status() callback function is no longer called. Hence remove
it.
Acked-by: Felipe Balbi <[email protected]>
Reviewed-by: Sagi Grimberg <[email protected]>
Reviewed-by: Andy Grover <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Bryant G. Ly <[email protected]>
Cc: Nicholas Bellinger <[email protected]>
Cc: Mike Christie <[email protected]>
Cc: Himanshu Madhani <[email protected]>
Cc: Quinn Tran <[email protected]>
Cc: Saurav Kashyap <[email protected]>
Cc: Michael S. Tsirkin <[email protected]>
Cc: Juergen Gross <[email protected]>
Signed-off-by: Bart Van Assche <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Five minor bug fixes.
The libfc one is a tiny memory leak, the zfcp one is an incorrect user
visible parameter and the rest are on error legs or obscure features"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: 53c700: pass correct "dev" to dma_alloc_attrs()
scsi: bnx2fc: Fix error handling in probe()
scsi: scsi_debug: fix write_same with virtual_gb problem
scsi: libfc: free skb when receiving invalid flogi resp
scsi: zfcp: fix sysfs block queue limit output for max_segment_size
|
|
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/scsi/csiostor/csio_attr.c: In function 'csio_fcoe_free_vnp':
drivers/scsi/csiostor/csio_attr.c:500:21: warning:
variable 'pln' set but not used [-Wunused-but-set-variable]
Signed-off-by: YueHaibing <[email protected]>
Acked-by: Varun Prakash <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This patch adds new BIT detection to enable FC-NVMe feature in the driver.
[mkp: fixed Giridhar's SoB]
Signed-off-by: Giridhar Malavali <[email protected]>
Signed-off-by: Himanshu Madhani <[email protected]>
Reviewed-by: Ewan D. Milne <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The aic94xx driver is currently failing to load with errors like
sysfs: cannot create duplicate filename '/devices/pci0000:00/0000:00:03.0/0000:02:00.3/0000:07:02.0/revision'
Because the PCI code had recently added a file named 'revision' to every
PCI device. Fix this by renaming the aic94xx revision file to
aic_revision. This is safe to do for us because as far as I can tell,
there's nothing in userspace relying on the current aic94xx revision file
so it can be renamed without breaking anything.
Fixes: 702ed3be1b1b (PCI: Create revision file in sysfs)
Cc: [email protected]
Signed-off-by: James Bottomley <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This call must have been missed when I reworked the debugfs feature for
upstreaming, so add it back.
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
A NULL-pointer dereference was introduced for TMF SSP commands from the
upstreaming reworking.
Fix this by relocating the scsi_get_prot_op() callsite.
Fixes: d6a9000b81be ("scsi: hisi_sas: Add support for DIF feature for v2 hw")
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Sparse detected some problems in the driver, so tidy them up.
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Some sizes we allocate for debugfs structure are incorrect, so fix them.
Signed-off-by: Luo Jiaxing <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|