Quinn Tran [Tue, 10 Aug 2021 04:37:13 +0000 (21:37 -0700)]
scsi: qla2xxx: Fix port type info
Over time, fcport->port_type became a flag field. The flags within this
field were not defined properly. This caused external tools to read wrong
info.
Link: https://lore.kernel.org/r/20210810043720.1137-8-njavali@marvell.com Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Signed-off-by: Quinn Tran <qutran@marvell.com> Signed-off-by: Nilesh Javali <njavali@marvell.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Shai Malin [Wed, 4 Aug 2021 22:14:12 +0000 (01:14 +0300)]
scsi: qedi: Add support for fastpath doorbell recovery
Driver fastpath employs doorbells to indicate to the device that work is
available. Each doorbell translates to a message sent to the device over
PCI. These messages are queued by the doorbell queue HW block, and handled
by the HW.
If a sufficient amount of CPU cores are sending messages at a sufficient
rate, the queue can overflow, and messages can be dropped. There are many
entities in the driver which can send doorbell messages. When overflow
happens, a fatal HW attention is indicated, and the Doorbell HW block stops
accepting new doorbell messages until recovery procedure is done.
When overflow occurs, all doorbells are dropped. Since doorbells are
aggregatives, if more doorbells are sent nothing has to be done. But if
the "last" doorbell is dropped, the doorbelling entity doesn’t know this
happened, and may wait forever for the device to perform the action. The
doorbell recovery mechanism addresses just that - it sends the last
doorbell of every entity.
[mkp: fix missing brackets reported by Guenter Roeck]
Link: https://lore.kernel.org/r/20210804221412.5048-1-smalin@marvell.com Co-developed-by: Manish Rangankar <mrangankar@marvell.com> Signed-off-by: Manish Rangankar <mrangankar@marvell.com> Signed-off-by: Shai Malin <smalin@marvell.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:55 +0000 (16:03 -0700)]
scsi: core: Remove the request member from struct scsi_cmnd
Since all scsi_cmnd.request users are gone, remove the request pointer
from struct scsi_cmnd.
Link: https://lore.kernel.org/r/20210809230355.8186-53-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:40 +0000 (16:03 -0700)]
scsi: qla1280: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. Remove the unused CMD_REQUEST() macro. This patch does not change
any functionality.
Bart Van Assche [Mon, 9 Aug 2021 23:03:15 +0000 (16:03 -0700)]
scsi: NCR5380: Use sc_data_direction instead of rq_data_dir()
This patch prepares for the removal of the request pointer from struct
scsi_cmnd and does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-13-bvanassche@acm.org Cc: Michael Schmitz <schmitzmic@gmail.com> Suggested-by: Finn Thain <fthain@linux-m68k.org> Acked-by: Finn Thain <fthain@linux-m68k.org> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:13 +0000 (16:03 -0700)]
scsi: zfcp: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-11-bvanassche@acm.org Acked-by: Benjamin Block <bblock@linux.ibm.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:11 +0000 (16:03 -0700)]
scsi: RDMA/iser: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-9-bvanassche@acm.org Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:10 +0000 (16:03 -0700)]
scsi: ata: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-8-bvanassche@acm.org Cc: Jens Axboe <axboe@kernel.dk> Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:09 +0000 (16:03 -0700)]
scsi: scsi_transport_spi: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-7-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:08 +0000 (16:03 -0700)]
scsi: scsi_transport_fc: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-6-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:07 +0000 (16:03 -0700)]
scsi: sr: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-5-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:06 +0000 (16:03 -0700)]
scsi: sd: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-4-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:05 +0000 (16:03 -0700)]
scsi: core: Use scsi_cmd_to_rq() instead of scsi_cmnd.request
Prepare for removal of the request pointer by using scsi_cmd_to_rq()
instead. Cast away constness where necessary when passing a SCSI command
pointer to scsi_cmd_to_rq(). This patch does not change any functionality.
Link: https://lore.kernel.org/r/20210809230355.8186-3-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bart Van Assche [Mon, 9 Aug 2021 23:03:04 +0000 (16:03 -0700)]
scsi: core: Introduce the scsi_cmd_to_rq() function
The 'request' member of struct scsi_cmnd is superfluous. The struct request
and struct scsi_cmnd data structures are adjacent and hence the request
pointer can be derived easily from a scsi_cmnd pointer. Introduce a helper
function that performs that conversion in a type-safe way. This patch is
the first step towards removing the request member from struct
scsi_cmnd. Making that change has the following advantages:
- This is a performance optimization since adding an offset to a pointer
takes less time than dereferencing a pointer.
- struct scsi_cmnd becomes smaller.
Link: https://lore.kernel.org/r/20210809230355.8186-2-bvanassche@acm.org Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Ming Lei <ming.lei@redhat.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Avri Altman [Sun, 8 Aug 2021 09:00:24 +0000 (12:00 +0300)]
scsi: ufs: ufshpb: Do not report victim error in HCM
In host control mode, eviction is perceived as an extreme measure. There
are several conditions that both the entering and exiting regions should
meet, so that eviction will take place.
The common case however, is that those conditions are rarely met, so it is
normal that the act of eviction fails. Therefore, do not report an error
in host control mode if eviction fails.
Avri Altman [Sun, 8 Aug 2021 09:00:23 +0000 (12:00 +0300)]
scsi: ufs: ufshpb: Verify that 'num_inflight_map_req' is non-negative
'num_inflight_map_req' should not be negative. It is incremented and
decremented without any protection, allowing it theoretically to be
negative, should some weird unbalanced count occur.
Verify that the those calls are properly serialized.
Avri Altman [Sun, 8 Aug 2021 09:00:22 +0000 (12:00 +0300)]
scsi: ufs: ufshpb: Use a correct max multi chunk
In HPB2.0, if pre_req_min_tr_len < transfer_len < pre_req_max_tr_len, the
driver is expected to send a HPB-WRITE-BUFFER companion to HPB-READ.
The upper bound should fit into a single byte, regardless of bMAX_
DATA_SIZE_FOR_HPB_SINGLE_CMD which being an attribute (u32) can be
significantly larger.
To further illustrate the issue, consider the following scenario:
- SCSI_DEFAULT_MAX_SECTORS is 1024 limiting the I/O chunks to 512KB
- The OEM changes scsi_host_template .max_sectors to be 2048 which allows
for 1MB requests: transfer_len = 256
Avri Altman [Sun, 8 Aug 2021 09:00:21 +0000 (12:00 +0300)]
scsi: ufs: ufshpb: Rewind the read timeout on every read
The purpose of the "cold"-timer is not to hang-on to active regions with no
reads. Therefore the read timeout should be rewound on every read, and not
just when the region is activated.
Adrian Hunter [Fri, 6 Aug 2021 13:04:41 +0000 (16:04 +0300)]
scsi: ufshcd: Fix device links when BOOT WLUN fails to probe
Managed device links are deleted by device_del(). However it is possible to
add a device link to a consumer before device_add(), and then discovering
an error prevents the device from being used. In that case normally
references to the device would be dropped and the device would be deleted.
However the device link holds a reference to the device, so the device link
and device remain indefinitely (unless the supplier is deleted).
For UFSHCD, if a LUN fails to probe (e.g. absent BOOT WLUN), the device
will not have been registered but can still have a device link holding a
reference to the device. The unwanted device link will prevent runtime
suspend indefinitely.
Amend device link removal to accept removal of a link with an unregistered
consumer device (suggested by Rafael), and fix UFSHCD by explicitly
deleting the device link when SCSI destroys the SCSI device.
Link: https://lore.kernel.org/r/a1c9bac8-b560-b662-f0aa-58c7e000cbbd@intel.com Fixes: b294ff3e3449 ("scsi: ufs: core: Enable power management for wlun") Reviewed-by: Rafael J. Wysocki <rafael@kernel.org> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Colin Ian King [Fri, 6 Aug 2021 14:43:01 +0000 (15:43 +0100)]
scsi: ufs: Fix unsigned int compared with less than zero
Variable 'tag' is currently an unsigned int and is being compared to less
than zero, this check is always false. Fix this by making 'tag' an int.
Link: https://lore.kernel.org/r/20210806144301.19864-1-colin.king@canonical.com Fixes: 4728ab4a8e64 ("scsi: ufs: Remove ufshcd_valid_tag()") Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Addresses-Coverity: ("Macro compares unsigned to 0")
Similarly to AHCI, introduce the device sysfs attribute
sas_ncq_prio_supported to advertise if a SATA device supports the NCQ
priority feature. Without this new attribute, the user can only discover if
a SATA device supports NCQ priority by trying to enable the feature use
with the sas_ncq_prio_enable sysfs device attribute, which fails when the
device does not support high prioity commands.
scsi: mpt3sas: Use firmware recommended queue depth
Currently, the mpt3sas driver sets the default queue depth based on the
physical interface of the attached device:
- SAS : 254
- SATA: 32
- NVMe: 128
The IOC firmware provides a recommended queue depth for each device through
SAS IO Unit Page1 for SAS/SATA and PCIe IO Unit Page 1 for NVMe devices.
If the host sets the queue depth greater than the firmware recommended
value, then the IOC places the I/Os above the recommended queue depth in an
internal pending queue. This consumes outstanding host-credit/resources,
thereby leading to potential starvation of other devices.
To avoid this, use the device depth recommended by the IOC firmware.
Enable the driver to work in non-IRQ mode, i.e. there will not be any MSI-X
vectors associated with queues dedicated to polling. The IOC hardware is
single submission queue and multiple reply queue. However, using the shared
host tagset support it is possible to simulate multiple hardware queues.
When poll_queues are enabled through the module parameter, the driver will
allocate extra reply queues without an MSI-X association. All I/O
completion on these queues will be done through the iopoll interface.
For Micron UFS devices the L2P entry need to be byteswapped before sending
an HPB READ command to the UFS device. Add the quirk
UFS_DEVICE_QUIRK_SWAP_L2P_ENTRY_FOR_HPB_READ to address this.
Colin Ian King [Wed, 4 Aug 2021 13:13:44 +0000 (14:13 +0100)]
scsi: qla2xxx: Remove redundant initialization of variable num_cnt
The variable num_cnt is being initialized with a value that is never read,
it is being updated later on. The assignment is redundant and can be
removed.
Link: https://lore.kernel.org/r/20210804131344.112635-1-colin.king@canonical.com Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Addresses-Coverity: ("Unused value")
David Disseldorp [Wed, 28 Jul 2021 11:53:53 +0000 (13:53 +0200)]
scsi: target: sbp: Drop incorrect ASC/ASCQ usage
The se_cmd scsi_asc and scsi_ascq members are only used for tracking ALUA
SCSI sense detail between target_core_alua and translate_sense_reason(), so
they're effectively always zero here.
Link: https://lore.kernel.org/r/20210728115353.2396-3-ddiss@suse.de Cc: Chris Boot <bootc@bootc.net> Reviewed-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: David Disseldorp <ddiss@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
David Disseldorp [Wed, 28 Jul 2021 11:53:52 +0000 (13:53 +0200)]
scsi: target: core: Avoid using lun_tg_pt_gp after unlock
core_alua_state_lba_dependent() currently uses lun->lun_tg_pt_gp without
holding the lun_tg_pt_gp_lock. The lock is taken in the caller, so obtain
the needed tg_pt_gp_id there instead.
Link: https://lore.kernel.org/r/20210728115353.2396-2-ddiss@suse.de Cc: Hannes Reinecke <hare@suse.de> Reviewed-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: David Disseldorp <ddiss@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bodo Stroesser [Tue, 13 Jul 2021 17:50:21 +0000 (19:50 +0200)]
scsi: target: tcmu: Add new feature KEEP_BUF
When running command pipelining for WRITE direction commands (e.g. tape
device write), userspace sends cmd completion to cmd ring before processing
write data. In that case userspace has to copy data before sending
completion, because cmd completion also implicitly releases the data buffer
in data area.
The new feature KEEP_BUF allows userspace to optionally keep the buffer
after completion by setting new bit TCMU_UFLAG_KEEP_BUF in
tcmu_cmd_entry_hdr->uflags. In that case buffer has to be released
explicitly by writing the cmd_id to new action item free_kept_buf.
All kept buffers are released during reset_ring and if userspace closes uio
device (tcmu_release).
scsi: ufs: Retry aborted SCSI commands instead of completing these successfully
Neither SAM nor the UFS standard require that the UFS controller fills in
the completion status of commands that have been aborted (LUN RESET aborts
pending commands). Hence do not rely on the completion status provided by
the UFS controller for aborted commands but instead ask the SCSI core to
retry SCSI commands that have been aborted.
Link: https://lore.kernel.org/r/20210722033439.26550-18-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: Synchronize SCSI and UFS error handling
Use the SCSI error handler instead of a custom error handling strategy.
This change reduces the number of potential races in the UFS drivers since
the UFS error handler and the SCSI error handler no longer run
concurrently.
Link: https://lore.kernel.org/r/20210722033439.26550-17-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Clearing a unit attention synchronously from inside the UFS error handler
may trigger the following deadlock:
- ufshcd_err_handler() calls ufshcd_err_handling_unprepare() and the
latter function calls ufshcd_clear_ua_wluns().
- ufshcd_clear_ua_wluns() submits a REQUEST SENSE command and that command
activates the SCSI error handler.
- The SCSI error handler calls ufshcd_host_reset_and_restore().
- ufshcd_host_reset_and_restore() executes the following code:
ufshcd_schedule_eh_work(hba); flush_work(&hba->eh_work);
This sequence results in a deadlock (circular wait). Fix this by requesting
sense data asynchronously.
Link: https://lore.kernel.org/r/20210722033439.26550-16-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
- Return FAILED instead of SUCCESS if the abort handler notices that a
SCSI command has already been completed. Returning SUCCESS in this case
triggers a use-after-free and may trigger a kernel crash.
- Fix the code for aborting SCSI commands submitted to a WLUN.
The current approach for aborting SCSI commands that have been submitted to
a WLUN and that timed out is as follows:
- Report to the SCSI core that the command has completed successfully.
Let the block layer free any data buffers associated with the command.
- Mark the command as outstanding in 'outstanding_reqs'.
- If the block layer tries to reuse the tag associated with the aborted
command, busy-wait until the tag is freed.
This approach can result in:
- Memory corruption if the controller accesses the data buffer after the
block layer has freed the associated data buffers.
- A race condition if ufshcd_queuecommand() or ufshcd_exec_dev_cmd()
checks the bit that corresponds to an aborted command in
'outstanding_reqs' after it has been cleared and before it is reset.
- High energy consumption if ufshcd_queuecommand() repeatedly returns
SCSI_MLQUEUE_HOST_BUSY.
Fix this by reporting to the SCSI error handler that aborting a SCSI
command failed if the SCSI command was submitted to a WLUN.
Link: https://lore.kernel.org/r/20210722033439.26550-15-bvanassche@acm.org Fixes: 7a7e66c65d41 ("scsi: ufs: Fix a race condition between ufshcd_abort() and eh_work()") Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Use a spinlock to protect hba->outstanding_reqs instead of using atomic
operations to update this member variable.
This patch is a performance improvement because it reduces the number of
atomic operations in the hot path (test_and_clear_bit()) and because it
reduces the lock contention on the SCSI host lock. On my test setup this
patch improves IOPS by about 1%.
Link: https://lore.kernel.org/r/20210722033439.26550-14-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: Optimize serialization of setup_xfer_req() calls
Reduce the number of times the host lock is taken in the hot path.
Additionally, inline ufshcd_vops_setup_xfer_req() because that function is
too short to keep it.
Link: https://lore.kernel.org/r/20210722033439.26550-13-bvanassche@acm.org Fixes: a45f937110fa ("scsi: ufs: Optimize host lock on transfer requests send/compl paths") Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Bean Huo <beanhuo@micron.com> Cc: Asutosh Das <asutoshd@codeaurora.org> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: Revert "Utilize Transfer Request List Completion Notification Register"
Using the UTRLCNR register involves two MMIO accesses in the hot path while
using the doorbell register only involves a single MMIO access. Since MMIO
accesses take time, do not use the UTRLCNR register. The spinlock
contention on the SCSI host lock that is reintroduced by this commit will
be addressed later.
Link: https://lore.kernel.org/r/20210722033439.26550-12-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Tested-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Inline ufshcd_outstanding_req_clear() since it only has one caller and
since its body is only one line long.
Link: https://lore.kernel.org/r/20210722033439.26550-11-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Reviewed-by: Avri Altman <avri.altman@wdc.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
From Documentation/memory-barriers.txt: "Note that, when using writel(), a
prior wmb() is not needed to guarantee that the cache coherent memory
writes have completed before writing to the MMIO region."
In other words, calling wmb() before writel() is not necessary. Hence
remove the wmb() calls that precede a writel() call. Remove the wmb() calls
that precede a ufshcd_send_command() call since the latter function uses
writel(). Remove the wmb() call from ufshcd_wait_for_dev_cmd() since the
following chain of events guarantees that the CPU will see up-to-date LRB
values:
- UFS controller writes to host memory.
- UFS controller posts completion interrupt after the memory writes from
the previous step are visible to the CPU.
- complete(hba->dev_cmd.complete) is called from the UFS interrupt handler.
- The wait_for_completion(hba->dev_cmd.complete) call in
ufshcd_wait_for_dev_cmd() returns.
Link: https://lore.kernel.org/r/20210722033439.26550-10-bvanassche@acm.org Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Can Guo <cang@codeaurora.org> Cc: Asutosh Das <asutoshd@codeaurora.org> Cc: Avri Altman <avri.altman@wdc.com> Tested-by: Avri altman <avri.altman@wdc.com> Reviewed-by: Avri Altman <avri.altman@wdc.com> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: Improve static type checking for the host controller state
Assign a name to the enumeration type for UFS host controller states and
remove the default clause from switch statements on this enumeration type
to make the compiler warn about unhandled enumeration labels.
Link: https://lore.kernel.org/r/20210722033439.26550-9-bvanassche@acm.org Cc: Can Guo <cang@codeaurora.org> Reviewed-by: Avri Altman <avri.altman@wdc.com> Reviewed-by: Keoseong Park <keosung.park@samsung.com> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>