Bean Huo [Thu, 5 May 2022 13:47:06 +0000 (15:47 +0200)]
scsi: ufs: ufshpb: Add handing of device reset regions in HPB device mode
In UFS HPB Spec JESD220-3A,
"5.8. Active and inactive information upon power cycle
...
When the device is powered off by the host, the device may restore L2P map
data upon power up or build from the host's HPB READ command. In case
device powered up and lost HPB information, device can signal to the host
through HPB Sense data, by setting HPB Operation as '2' which will inform
the host that device reset HPB information."
Therefore, for HPB device control mode, if the UFS device is reset via the
RST_N pin, the active region information in the device will be reset. If
the host side receives this notification from the device side, it is
recommended to inactivate all active regions in the host's HPB cache.
Link: https://lore.kernel.org/r/20220505134707.35929-6-huobean@gmail.com Reviewed-by: Keoseong Park <keosung.park@samsung.com> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Signed-off-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Bean Huo [Thu, 5 May 2022 13:47:05 +0000 (15:47 +0200)]
scsi: ufs: ufshpb: Change sysfs node hpb_stats/rb_* prefix to start with rcmd_*
According to the documentation of the sysfs nodes rb_noti_cnt,
rb_active_cnt and rb_inactive_cnt, these are all related to HPB
recommendation in UPIU response packet. 'rcmd' (recommendation) should be
the correct abbreviation.
Change the sysfs documentation about these sysfs nodes to highlight what
they mean under different HPB control modes.
Bean Huo [Thu, 5 May 2022 13:47:04 +0000 (15:47 +0200)]
scsi: ufs: ufshpb: Clean up the handler when device resets HPB information
"When the device is powered off by the host, the device may restore L2P map
data upon power up or build from the host's HPB READ command. In case
device powered up and lost HPB information, device can signal to the host
through HPB Sense data, by setting HPB Operation as '2' which will inform
the host that device reset HPB information."
Clean up the handler and make the intent of this handler more readable, no
functional change.
Link: https://lore.kernel.org/r/20220505134707.35929-4-huobean@gmail.com Reviewed-by: Keoseong Park <keosung.park@samsung.com> Reviewed-by: Daejun Park <daejun7.park@samsung.com> Signed-off-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: qcom: Enable RPM_AUTOSUSPEND for runtime PM
In order to allow the block devices to enter autosuspend mode during
runtime, thereby allowing the ufshcd host driver to also runtime suspend,
let's make use of the RPM_AUTOSUSPEND flag.
Without this flag, userspace needs to enable the autosuspend feature of
the block devices through sysfs.
scsi: ufs: core: Remove redundant wmb() in ufshcd_send_command()
The wmb() inside ufshcd_send_command() is added to make sure that the
doorbell is committed immediately. This leads to couple of expectations:
1. The doorbell write should complete before the function return.
2. The doorbell write should not cross the function boundary.
2nd expectation is fullfilled by the Linux memory model as there is a
guarantee that the critical section won't cross the unlock (release)
operation.
1st expectation is not really needed here as there is no following read/
write that depends on the doorbell to be complete implicitly. Even if the
doorbell write is in a CPUs Write Buffer (WB), wmb() won't flush it. And
there is no real need of a WB flush here as well.
So let's get rid of the wmb() that seems redundant.
Link: https://lore.kernel.org/r/20220504084212.11605-5-manivannan.sadhasivam@linaro.org Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: qcom: Add a readl() to make sure ref_clk gets enabled
In ufs_qcom_dev_ref_clk_ctrl(), it was noted that the ref_clk needs to be
stable for at least 1us. Even though there is wmb() to make sure the write
gets "completed", there is no guarantee that the write actually reached the
UFS device. There is a good chance that the write could be stored in a
Write Buffer (WB). In that case, even though the CPU waits for 1us, the
ref_clk might not be stable for that period.
So lets do a readl() to make sure that the previous write has reached the
UFS device before udelay().
Also, the wmb() after writel_relaxed() is not really needed. Both writel()
and readl() are ordered on all architectures and the CPU won't speculate
instructions after readl() due to the in-built control dependency with read
value on weakly ordered architectures. So it can be safely removed.
Link: https://lore.kernel.org/r/20220504084212.11605-4-manivannan.sadhasivam@linaro.org Fixes: cd40d3be913c ("scsi: ufs-qcom: add QUniPro hardware support and power optimizations") Cc: stable@vger.kernel.org Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: qcom: Simplify handling of devm_phy_get()
There is no need to call devm_phy_get() if ACPI is used, so skip it. The
host->generic_phy pointer should already be NULL due to the kzalloc(), so
no need to set it NULL again.
While at it, also remove the comment that has no relationship with
devm_phy_get().
Link: https://lore.kernel.org/r/20220504084212.11605-3-manivannan.sadhasivam@linaro.org Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Reviewed-by: Andrew Halaney <ahalaney@redhat.com> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: qcom: Fix acquiring the optional reset control line
On Qcom UFS platforms, the reset control line seems to be optional (for
SoCs like MSM8996 and probably for others too). The current logic tries to
mimic the devm_reset_control_get_optional() API but it also continues the
probe if there is an error with the declared reset line in DT/ACPI.
In an ideal case, if the reset line is not declared in DT/ACPI, the probe
should continue. But if there is problem in acquiring the declared reset
line (like EPROBE_DEFER) it should fail and return the appropriate error
code.
Link: https://lore.kernel.org/r/20220504084212.11605-2-manivannan.sadhasivam@linaro.org Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Reviewed-by: Andrew Halaney <ahalaney@redhat.com> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Xiang Chen [Fri, 6 May 2022 12:25:39 +0000 (20:25 +0800)]
scsi: hisi_sas: Undo RPM resume for failed notify phy event for v3 HW
If we fail to notify the phy up event then undo the RPM resume, as the phy
up notify event handling pairs with that RPM resume.
Link: https://lore.kernel.org/r/1651839939-101188-1-git-send-email-john.garry@huawei.com Reported-by: Yihang Li <liyihang6@hisilicon.com> Tested-by: Yihang Li <liyihang6@hisilicon.com> Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> Signed-off-by: John Garry <john.garry@huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add support for management applications to send an MPI3 Encapsulated NVMe
passthru command to the NVMe devices attached to an Avenger controller.
Since the NVMe drives are exposed as SCSI devices by the controller, the
standard NVMe applications cannot be used to interact with the drives and
the command sets supported are also limited by the controller firmware.
Special handling is required for MPI3 Encapsulated NVMe passthru commands
for PRP/SGL setup in the commands.
Implement driver support for management applications to enable persistent
event log (PEL) notifications. Upon receipt of events, the driver will
increment a sysfs variable named event_counter. The management application
will poll for event_counter value changes and signal the application about
events.
There are certain bsg commands which need to be completed by the driver
without involving firmware. These requests are termed driver commands. Add
support for these.
Link: https://lore.kernel.org/r/20220429211641.642010-3-sumit.saxena@broadcom.com
Reported by: Stephen Rothwell <sfr@canb.auug.org.au> Reported-by: kernel test robot <lkp@intel.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Signed-off-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Enze Li [Wed, 27 Apr 2022 02:56:47 +0000 (10:56 +0800)]
scsi: sr: Add memory allocation failure handling for get_capabilities()
The function get_capabilities() has the possibility of failing to allocate
the transfer buffer but it does not currently handle this. This may lead to
exceptions when accessing the buffer.
Xiaoguang Wang [Thu, 21 Apr 2022 02:37:35 +0000 (10:37 +0800)]
scsi: target: tcmu: Fix possible data corruption
When tcmu_vma_fault() gets a page successfully, before the current context
completes page fault procedure, find_free_blocks() may run and call
unmap_mapping_range() to unmap the page. Assume that when
find_free_blocks() initially completes and the previous page fault
procedure starts to run again and completes, then one truncated page has
been mapped to userspace. But note that tcmu_vma_fault() has gotten a
refcount for the page so any other subsystem won't be able to use the page
unless the userspace address is unmapped later.
If another command subsequently runs and needs to extend dbi_thresh it may
reuse the corresponding slot for the previous page in data_bitmap. Then
though we'll allocate new page for this slot in data_area, no page fault
will happen because we have a valid map and the real request's data will be
lost.
Filesystem implementations will also run into this issue but they usually
lock the page when vm_operations_struct->fault gets a page and unlock the
page after finish_fault() completes. For truncate filesystems lock pages in
truncate_inode_pages() to protect against racing wrt. page faults.
To fix this possible data corruption scenario we can apply a method similar
to the filesystems. For pages that are to be freed, tcmu_blocks_release()
locks and unlocks. Make tcmu_vma_fault() also lock found page under
cmdr_lock. At the same time, since tcmu_vma_fault() gets an extra page
refcount, tcmu_blocks_release() won't free pages if pages are in page fault
procedure, which means it is safe to call tcmu_blocks_release() before
unmap_mapping_range().
With these changes tcmu_blocks_release() will wait for all page faults to
be completed before calling unmap_mapping_range(). And later, if
unmap_mapping_range() is called, it will ensure stale mappings are removed.
Prior patch added a call to lpfc_sli_prep_wqe() prior to
lpfc_sli_issue_iocb(). This call should not have been added as prep_wqe is
called within the issue_iocb routine. So it's called twice now.
scsi: sd: Reorganize DIF/DIX code to avoid calling revalidate twice
During device discovery we ended up calling revalidate twice and thus
requested the same parameters multiple times. This was originally necessary
due to the request_queue and gendisk needing to be instantiated to
configure the block integrity profile.
Since this dependency no longer exists, reorganize the integrity probing
code so it can be run once at the end of discovery and drop the superfluous
revalidate call. Postponing the registration step involves splitting
sd_read_protection() into two functions, one to read the device protection
type and one to configure the mode of operation.
As part of this cleanup, make the printing code a bit less verbose.
scsi: sd: Optimal I/O size should be a multiple of reported granularity
Commit dd13fc0f321e ("scsi: sd: Optimal I/O size should be a multiple of
physical block size") validated the reported optimal I/O size against the
physical block size to overcome problems with devices reporting nonsensical
transfer sizes.
However, some devices claim conformity to older SCSI versions that predate
the physical block size being reported. Other devices do not report a
physical block size at all. We need to be able to validate the optimal I/O
size on those devices as well.
Many devices report an OPTIMAL TRANSFER LENGTH GRANULARITY in the same VPD
page as the OPTIMAL TRANSFER LENGTH. Use this value to validate the optimal
I/O size. Also check that the reported granularity is a multiple of the
physical block size, if supported.
Since the ATA Information VPD is now cached at device discovery time it is
no longer necessary to request this page when we configure WRITE SAME.
Instead use the cached information to determine if this disk sits behind a
SCSI-ATA translation layer.
scsi: core: Do not truncate INQUIRY data on modern devices
Low-level device drivers have had the ability to limit the size of an
INQUIRY for many years. This made sense for a wide variety of legacy
devices. However, we are unnecessarily truncating the INQUIRY response for
many modern devices. This prevents us from consulting fields beyond the
first 36 bytes.
If a device reports that it supports a larger INQUIRY response, and the
device also reports that it implements SPC-4 or newer, allow the larger
INQUIRY to proceed.
The SCSI disk driver consults VPD pages b0 (Block Limits), b1 (Block Device
Characteristics), and b2 (Logical Block Provisioning). Instead of having
sd.c request these pages every revalidate cycle, cache them along with the
other commonly used VPDs.
Link: https://lore.kernel.org/r/20220302053559.32147-6-martin.petersen@oracle.com Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: core: Pick suitable allocation length in scsi_report_opcode()
Some devices hang when a buffer size larger than expected is passed in the
ALLOCATION LENGTH field. For REPORT SUPPORTED OPERATION CODES we currently
only request a single command descriptor at a time and therefore the actual
size of the command is known ahead of time. Limit the ALLOCATION LENGTH to
the header size plus the command length of the opcode we are asking about.
Link: https://lore.kernel.org/r/20220302053559.32147-5-martin.petersen@oracle.com Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: core: Query VPD size before getting full page
We currently default to 255 bytes when fetching VPD pages during discovery.
However, we have had a few devices that are known to wedge if the requested
buffer exceeds a certain size. See commit e23ccf036d5f ("[SCSI] sd: Reduce
buffer size for vpd request") which works around one example of this
problem in the SCSI disk driver.
With commit a2200cbc64bf ("scsi: core: Add sysfs attributes for VPD pages
0h and 89h") we now risk triggering the same issue in the generic midlayer
code.
The problem with the ATA VPD page in particular is that the SCSI portion of
the page is trailed by 512 bytes of verbatim ATA Identify Device
information. However, not all controllers actually provide the additional
512 bytes and will lock up if one asks for more than the 64 bytes
containing the SCSI protocol fields.
Instead of picking a new, somewhat arbitrary, number of bytes for the VPD
buffer size, start fetching the 4-byte header for each page. The header
contains the size of the page as far as the device is concerned. We can use
the reported size to specify the correct allocation length when
subsequently fetching the full page.
The header validation is done by a new helper function scsi_get_vpd_size()
and both scsi_get_vpd_page() and scsi_get_vpd_buf() now rely on this to
query the page size.
In addition, scsi_get_vpd_page() is simplified to mirror the logic in
scsi_get_vpd_page(). This involves removing the Supported VPD Pages lookup
prior to attempting to query a page. There does not appear any evidence,
even in the oldest SCSI specs, that this step is required. We already rely
on scsi_get_vpd_page() throughout the stack and this function never
consulted the Supported VPD Pages. Since this has not caused any problems
it should be safe to remove the precondition from scsi_get_vpd_page().
Instrumented runs also revealed that the Supported VPD Pages lookup had
little effect since the device page index often was larger than the
supplied buffer size. As a result, inquiries frequently bypassed the index
check and went through the "If we ran off the end of the buffer, give us
the benefit of the doubt" code path which assumed the page was present
despite not being listed. The revised code takes both the page size
reported by the device as well as the size of the buffer provided by the
scsi_get_vpd_page() caller into account.
Link: https://lore.kernel.org/r/20220302053559.32147-3-martin.petersen@oracle.com Fixes: a2200cbc64bf ("scsi: core: Add sysfs attributes for VPD pages 0h and 89h") Reported-by: Maciej W. Rozycki <macro@orcam.me.uk> Tested-by: Maciej W. Rozycki <macro@orcam.me.uk> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: mpt3sas: Use cached ATA Information VPD page
We now cache VPD page 0x89 (ATA Information) so there is no need to request
it from the hardware. Make mpt3sas use the cached page.
Link: https://lore.kernel.org/r/20220302053559.32147-2-martin.petersen@oracle.com Cc: Sreekanth Reddy <sreekanth.reddy@broadcom.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
James Smart [Tue, 26 Apr 2022 18:14:19 +0000 (11:14 -0700)]
scsi: lpfc: Fix resource leak in lpfc_sli4_send_seq_to_ulp()
If no handler is found in lpfc_complete_unsol_iocb() to match the rctl of a
received frame, the frame is dropped and resources are leaked.
Fix by returning resources when discarding an unhandled frame type. Update
lpfc_fc_frame_check() handling of NOP basic link service.
Link: https://lore.kernel.org/r/20220426181419.9154-1-jsmart2021@gmail.com Co-developed-by: Justin Tee <justin.tee@broadcom.com> Signed-off-by: Justin Tee <justin.tee@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
James Smart [Tue, 26 Apr 2022 18:13:15 +0000 (11:13 -0700)]
scsi: lpfc: Remove unnecessary null ndlp check in lpfc_sli_prep_wqe()
Smatch had the following warning:
drivers/scsi/lpfc/lpfc_sli.c:22305 lpfc_sli_prep_wqe() error: we previously assumed 'ndlp' could be null (see line 22298)
Remove the unnecessary null check.
Link: https://lore.kernel.org/r/20220426181315.8990-1-jsmart2021@gmail.com Fixes: f70b5c0741d4 ("scsi: lpfc: Fix field overload in lpfc_iocbq data structure") Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Minghao Chi [Wed, 20 Apr 2022 09:03:52 +0000 (09:03 +0000)]
scsi: ufs: Use pm_runtime_resume_and_get() instead of pm_runtime_get_sync()
Using pm_runtime_resume_and_get() to replace pm_runtime_get_sync() and
pm_runtime_put_noidle(). This change is just to simplify the code, no
actual functional changes.
scsi: megaraid: Fix error check return value of register_chrdev()
If major equals 0, register_chrdev() returns an error code when it fails.
This function dynamically allocates a major and returns its number on
success, so we should use "< 0" to check it instead of "!".
scsi: dc395x: Fix a missing check on list iterator
The bug is here:
p->target_id, p->target_lun);
The list iterator 'p' will point to a bogus position containing HEAD if the
list is empty or no element is found. This case must be checked before any
use of the iterator, otherwise it will lead to an invalid memory access.
To fix this bug, add a check. Use a new variable 'iter' as the list
iterator, and use the original variable 'p' as a dedicated pointer to point
to the found element.
scsi: qedf: Remove an unneeded NULL check on list iterator
The list iterator 'fcport' is always non-NULL so it doesn't need to be
checked. Thus just remove the unnecessary NULL check. Also remove the
unnecessary initializer because the list iterator is always initialized.
And adjust the position of blank lines.
Kiwoong Kim [Thu, 31 Mar 2022 01:24:05 +0000 (10:24 +0900)]
scsi: ufs: core: Exclude UECxx from SFR dump list
Some devices may return invalid or zeroed data during an UIC error
condition. In addition, reading these SFRs will clear them. This means the
subsequent error handling will not be able to see them and therefore no
error handling will be scheduled.
John Garry [Wed, 30 Mar 2022 11:38:35 +0000 (19:38 +0800)]
scsi: core: Refine how we set tag_set NUMA node
For SCSI hosts which enable host_tagset the NUMA node returned from
blk_mq_hw_queue_to_node() is NUMA_NO_NODE always. Then, since in
scsi_mq_setup_tags() the default we choose for the tag_set NUMA node is
NUMA_NO_NODE, we always evaluate the NUMA node as NUMA_NO_NODE in functions
like blk_mq_alloc_rq_map().
The reason we get NUMA_NO_NODE from blk_mq_hw_queue_to_node() is that the
hctx_idx passed is BLK_MQ_NO_HCTX_IDX - so we can't match against a (HW)
queue mapping index.
Improve this by defaulting the tag_set NUMA node to the same NUMA node of
the SCSI host DMA dev.
The following warning showed up when compiling with W=1.
drivers/message/fusion/mptctl.c: In function ‘mptctl_hp_hostinfo’:
drivers/message/fusion/mptctl.c:2337:8: warning: variable ‘retval’ set but not used [-Wunused-but-set-variable]
int retval;
scsi: aacraid: Fix undefined behavior due to shift overflowing the constant
Fix:
drivers/scsi/aacraid/commsup.c: In function ‘aac_handle_sa_aif’:
drivers/scsi/aacraid/commsup.c:1983:2: error: case label does not reduce to an integer constant
case SA_AIF_BPCFG_CHANGE:
^~~~
See https://lore.kernel.org/r/YkwQ6%2BtIH8GQpuct@zn.tnic for the gory
details as to why it triggers with older gccs only.
Link: https://lore.kernel.org/r/20220405151517.29753-2-bp@alien8.de Cc: Adaptec OEM Raid Solutions <aacraid@microsemi.com> Cc: "James E.J. Bottomley" <jejb@linux.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Reviewed-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Borislav Petkov <bp@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Damien Le Moal [Thu, 21 Apr 2022 18:30:23 +0000 (11:30 -0700)]
scsi: scsi_debug: Add gap zone support
Add the 'zone_cap_mb' kernel module parameter. This parameter defines the
zone capacity. The zone capacity must be less than or equal to the zone
size.
Report that sequential write zones and gap zones are paired in the Zoned
Block Device Characteristics VPD page (page B6h).
This patch has been tested as follows:
modprobe scsi_debug delay=0 sector_size=512 dev_size_mb=128 zbc=host-managed zone_nr_conv=16 zone_size_mb=4 zone_cap_mb=3
modprobe brd rd_nr=1 rd_size=$((1<<20))
mkfs.f2fs -m /dev/ram0 -c /dev/${scsi_debug_dev}
mount /dev/ram0 /mnt
# Run a fio job that uses /mnt
Link: https://lore.kernel.org/r/20220421183023.3462291-10-bvanassche@acm.org Cc: Douglas Gilbert <dgilbert@interlog.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
[ bvanassche: Switched to reporting a constant zone starting LBA granularity ] Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Damien Le Moal [Thu, 21 Apr 2022 18:30:22 +0000 (11:30 -0700)]
scsi: scsi_debug: Rename zone type constants
Rename the scsi_debug zone type constants to prevent a conflict with the
ZBC_ZONE_TYPE_GAP constant from include/scsi/scsi_proto.h.
Link: https://lore.kernel.org/r/20220421183023.3462291-9-bvanassche@acm.org Cc: Douglas Gilbert <dgilbert@interlog.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
[ bvanassche: Extracted these changes from a larger patch ] Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/20220421183023.3462291-8-bvanassche@acm.org Cc: Douglas Gilbert <dgilbert@interlog.com> Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Damien Le Moal [Thu, 21 Apr 2022 18:30:20 +0000 (11:30 -0700)]
scsi: sd: sd_zbc: Hide gap zones
ZBC-2 allows host-managed disks to report gap zones. This allow zoned disks
to report an offset between data zone starts that is a power of two even if
the number of logical blocks with data per zone is not a power of two.
Another new feature in ZBC-2 is support for constant zone starting LBA
offsets. For zoned disks that report a constant zone starting LBA offset,
hide the gap zones from the block layer. Report the offset between data
zone starts as zone size and report the number of logical blocks with data
per zone as the zone capacity.
Link: https://lore.kernel.org/r/20220421183023.3462291-7-bvanassche@acm.org Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
[ bvanassche: Reworked this patch ] Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Damien Le Moal [Thu, 21 Apr 2022 18:30:19 +0000 (11:30 -0700)]
scsi: sd: sd_zbc: Return early in sd_zbc_check_zoned_characteristics()
Return early in sd_zbc_check_zoned_characteristics() for host-aware
disks. This patch does not change any functionality but makes a later patch
easier to read.
Link: https://lore.kernel.org/r/20220421183023.3462291-6-bvanassche@acm.org Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
[ bvanassche: extracted this change from a larger patch ] Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Deriving the meaning of the nr_zones, rev_nr_zones, zone_blocks and
rev_zone_blocks member variables requires careful analysis of the source
code. Make the meaning of these member variables easier to understand by
introducing struct zoned_disk_info.
Link: https://lore.kernel.org/r/20220421183023.3462291-5-bvanassche@acm.org Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Damien Le Moal [Thu, 21 Apr 2022 18:30:17 +0000 (11:30 -0700)]
scsi: sd: sd_zbc: Use logical blocks as unit when querying zones
When querying zones, track the position in logical blocks instead of in
sectors. This change slightly simplifies sd_zbc_report_zones().
Link: https://lore.kernel.org/r/20220421183023.3462291-4-bvanassche@acm.org Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
[ bvanassche: extracted this change from a larger patch ] Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Hence this patch that verifies that the zone size is a power of two.
Link: https://lore.kernel.org/r/20220421183023.3462291-3-bvanassche@acm.org Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add several kernel-doc headers. Declare input arrays const. Specify the
array size in function declarations.
Link: https://lore.kernel.org/r/20220421183023.3462291-2-bvanassche@acm.org Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Douglas Gilbert <dgilbert@interlog.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Move the definition of this data structure since it is only used in a
single source file.
Link: https://lore.kernel.org/r/20220419225811.4127248-28-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
Split the ufshcd.h header file into a header file that defines the
interface used by UFS drivers and another header file with declarations and
data structures only used by the UFS core.
Follow the convention that is used elsewhere in the Linux kernel source
code and only include those headers of which the declarations are used
directly.
Link: https://lore.kernel.org/r/20220419225811.4127248-25-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Remove unnecessary ufshcd-crypto.h include directives
ufshcd-crypto.h declares functions that must only be called by the UFS
core. Hence remove the #include "ufshcd-crypto.h" directive from UFS
drivers.
Link: https://lore.kernel.org/r/20220419225811.4127248-24-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Eric Biggers <ebiggers@google.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 hba->is_sys_suspended if ufs_qcom_resume() succeeds is wrong. That
variable must only be cleared if all actions involved in a resume succeed.
Hence remove the statement that clears hba->is_sys_suspended from
ufs_qcom_resume().
Link: https://lore.kernel.org/r/20220419225811.4127248-23-bvanassche@acm.org Fixes: e5586a66ebb0 ("ufs-qcom: add support for Qualcomm Technologies Inc platforms") Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> 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>
Since the code to modify delay_ms while holding the host lock occurs twice,
introduce a function that performs this action.
Link: https://lore.kernel.org/r/20220419225811.4127248-22-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Remove locking from around single register writes
Single register writes are atomic and hence do not need to be surrounded by
locking. Additionally, MMIO writes are typically posted asynchronously.
Hence, there is no guarantee that these have finished by the time the
spin_unlock*() call has finished. See also the nonposted-mmio property of
the Open Firmware tree. See also pci_iomap().
Link: https://lore.kernel.org/r/20220419225811.4127248-21-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
In the Linux kernel coding style document
(Documentation/process/coding-style.rst) it is recommended to use the type
'bool' and also the values 'true' and 'false'. Hence this patch that
removes the definitions and uses of TRUE and FALSE from the UFS driver.
Link: https://lore.kernel.org/r/20220419225811.4127248-20-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
Since specifying the path in a source file is redundant, remove the paths
from source code comments.
Link: https://lore.kernel.org/r/20220419225811.4127248-19-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Rename sdev_ufs_device into ufs_device_wlun
The new name reflects the role of this member variable better: a WLUN
through which the power mode of the UFS device is controlled.
Link: https://lore.kernel.org/r/20220419225811.4127248-17-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
The current version number is 0.2. That driver version was assigned more
than nine years ago. A version number that is not updated while the driver
is updated is not useful. Hence remove the driver version number from the
UFS driver. See also commit 53e303e34feb ("[SCSI] ufs: Separate PCI code
into glue driver").
Link: https://lore.kernel.org/r/20220419225811.4127248-16-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Avri Altman <avri.altman@wdc.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: Make the config_scaling_param calls type safe
Pass the actual type to config_scaling_param callback as the third argment
instead of a void pointer. Remove a superfluous NULL pointer check from
ufs_qcom_config_scaling_param().
Link: https://lore.kernel.org/r/20220419225811.4127248-15-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> 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>
Make it easier to verify for humans that ufshcd_init_pwr_dev_param()
initializes all structure members. This patch does not change any
functionality.
Link: https://lore.kernel.org/r/20220419225811.4127248-14-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
Commit 6ff7a99a44c2 ("scsi: ufs: Remove pre-defined initial voltage values
of device power") removed the code that uses the UFS_VREG_VCC* constants
and also the code that sets the min_uV and max_uV member variables. Hence
also remove these constants and that member variable.
Link: https://lore.kernel.org/r/20220419225811.4127248-13-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Stanley Chu <stanley.chu@mediatek.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: Invert the return value of ufshcd_is_hba_active()
It is confusing that ufshcd_is_hba_active() returns 'true' if the HBA is
not active. Clear up this confusion by inverting the return value of
ufshcd_is_hba_active(). This patch does not change any functionality.
Link: https://lore.kernel.org/r/20220419225811.4127248-12-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
Declare the quirks array and also its 'model' member const to make it
explicit that these are not modified.
Link: https://lore.kernel.org/r/20220419225811.4127248-11-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Rename struct ufs_dev_fix into ufs_dev_quirk
Since struct ufs_dev_fix contains quirk information, rename it into struct
ufs_dev_quirk.
Link: https://lore.kernel.org/r/20220419225811.4127248-10-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Remove the UFS_FIX() and END_FIX() macros
Since these two macros reduce code readability, remove them.
Link: https://lore.kernel.org/r/20220419225811.4127248-9-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
ufshcd_lrb.sense_buffer is NULL if ufshcd_lrb.cmd is NULL and
ufshcd_lrb.sense_buffer points at cmd->sense_buffer if ufshcd_lrb.cmd is
set. In other words, the ufshcd_lrb.sense_buffer member is identical to
cmd->sense_buffer. Hence this patch that removes the
ufshcd_lrb.sense_buffer structure member.
Link: https://lore.kernel.org/r/20220419225811.4127248-7-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
ufshcd_lrb.sense_bufflen is set but never read. Hence remove this struct
member.
Link: https://lore.kernel.org/r/20220419225811.4127248-6-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Avri Altman <avri.altman@wdc.com> Reviewed-by: Keoseong Park <keosung.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: Simplify statements that return a boolean
Convert "if (expr) return true; else return false;" into "return expr;" if
either 'expr' is a boolean expression or the return type of the function is
'bool'.
Link: https://lore.kernel.org/r/20220419225811.4127248-5-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Reviewed-by: Keoseong Park <keosung.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>
Remove "? true : false" if the preceding expression yields a boolean or if
the result of the expression is assigned to a boolean since in these two
cases the "? true : false" part is superfluous.
Link: https://lore.kernel.org/r/20220419225811.4127248-4-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.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>
Declare this function static since it is only used inside the ufshcd.c
source file.
Link: https://lore.kernel.org/r/20220419225811.4127248-3-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: Fix a spelling error in a source code comment
Change one occurrence of "adpater" into "adapter".
Link: https://lore.kernel.org/r/20220419225811.4127248-2-bvanassche@acm.org Tested-by: Bean Huo <beanhuo@micron.com> 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>
scsi: ufs: core: Increase fDeviceInit poll frequency
UFS devices are expected to clear fDeviceInit flag in single digit
milliseconds. Current values of 5 to 10 millisecond sleep add to increased
latency during the initialization and resume path. This CL lowers the sleep
range to 500 to 1000 microseconds.
Link: https://lore.kernel.org/r/20220421002429.3136933-1-bvanassche@acm.org Acked-by: Avri Altman <avri.altman@wdc.com> Signed-off-by: Konstantin Vyshetsky <vkon@google.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Dan Carpenter [Thu, 21 Apr 2022 15:03:52 +0000 (18:03 +0300)]
scsi: iscsi: Fix harmless double shift bug
These flags are supposed to be bit numbers. Right now they cause a double
shift bug where we use BIT(BIT(2)) instead of BIT(2). Fortunately, the bit
numbers are small and it's done consistently so it does not cause an issue
at run time.
Link: https://lore.kernel.org/r/YmFyWHf8nrrx+SHa@kili Fixes: bfc38cb7416d ("scsi: iscsi: Merge suspend fields") Reviewed-by: Mike Christie <michael.christie@oracle.com> Reviewed-by: Lee Duncan <lduncan@suse.com> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: target: core: Silence the message about unknown VPD pages
Target does not support some VPD pages and is very verbose about it.
Sometimes initiators don't bother and just keep sending the same request
from time to time, filling up the logs.
The replyPostRegisterIndex array of struct MPT3SAS_ADAPTER stores iomem
resource addresses. Fix its declaration to annotate it with __iomem to
avoid sparse warnings for writel() calls using the stored addresses.
Damien Le Moal [Mon, 7 Mar 2022 23:48:53 +0000 (08:48 +0900)]
scsi: mpt3sas: Fix event callback log_code value handling
In mpt3sas_scsih_event_callback(), fix a sparse warning when testing the
event log code value by replacing the use of a pointer to the address
storing the event log code with a log code local variable. Doing so,
le32_to_cpu() is used when the log code value is assigned, avoiding a
sparse warning.
Damien Le Moal [Mon, 7 Mar 2022 23:48:52 +0000 (08:48 +0900)]
scsi: mpt3sas: Fix ioc->base_readl() use
The functions _base_readl_aero() and _base_readl() used for an adapter
base_readl() method are implemented using a regular readl() call which
internally performs a conversion to CPU endianness (le32_to_cpu()) of
the values read. The users of the ioc base_readl() method should thus
not convert again the values read using le16_to_cpu().
Fixing this removes sparse warnings.
projects / kernel.git / log