| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Create function smp_ata_check_ready_type() for LLDDs to wait for SATA
devices to come up after a link reset.
Signed-off-by: Jie Zhan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Clean up the code by making use of the newly introduced
ata_port_is_frozen() helper function.
Signed-off-by: Niklas Cassel <[email protected]>
Reviewed-by: John Garry <[email protected]>
Acked-by: Martin K. Petersen <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
|
|
The SAS address comparison of asd_sas_port and expander phy is open
coded. Replace it with sas_phy_match_port_addr().
Signed-off-by: Jason Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Damien Le Moal <[email protected]>
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The SAS address comparison of expander phys is open coded. Replace it with
sas_phy_addr_match().
Signed-off-by: Jason Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Damien Le Moal <[email protected]>
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The SAS address comparison of domain device and expander phy is open
coded. Replace it with sas_phy_match_dev_addr().
Signed-off-by: Jason Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Damien Le Moal <[email protected]>
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
LLDDs are all implementing their own attached phy ID finding code. Factor
it out to libsas.
Signed-off-by: Jason Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Jack Wang <[email protected]>
Reviewed-by: Damien Le Moal <[email protected]>
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
SAS address comparison is widely used in libsas. However they are all
opencoded and to avoid the line spill over 80 columns, are mostly split
into multi-lines. Introduce some helpers to prepare for some refactoring.
Signed-off-by: Jason Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Damien Le Moal <[email protected]>
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Johannes Thumshirn <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
In sas_ata_task_done(), for commands which complete with error we set the
SATA dev FIS status field with ATA_ERR. In ata_eh_analyze_tf() this would
be interpreted as a HSM error. Set ATA_DRDY, which will lead libata to
judge as a device error, which is a safer bet.
Signed-off-by: John Garry <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Niklas Cassel <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
We have no users outside libsas any longer, so make sas_alloc_task(),
sas_alloc_slow_task(), and sas_free_task() private.
Signed-off-by: John Garry <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Tested-by: Niklas Cassel <[email protected]> # pm80xx
Reviewed-by: Jason Yan <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Similar to how AHCI handles NCQ errors in ahci_error_intr() ->
ata_port_abort() -> ata_do_link_abort(), add an NCQ error handler for LLDDs
to call to initiate a link abort.
This will mark all outstanding QCs as failed and kick-off EH.
Note:
A "force reset" argument is added for drivers which require the ATA error
handling to always reset the device.
A driver may require this feature for when SATA device per-SCSI cmnd
resources are only released during reset for ATA EH. As such, we need an
option to force reset to be done, regardless of what any EH autopsy
decides.
The SATA device FIS fields are set to indicate a device error from
ata_eh_analyze_tf().
Suggested-by: Damien Le Moal <[email protected]>
Suggested-by: Niklas Cassel <[email protected]>
Signed-off-by: John Garry <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Tested-by: Niklas Cassel <[email protected]> # pm80xx
Reviewed-by: Jason Yan <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Pull SCSI updates from James Bottomley:
"Updates to the usual drivers (qla2xxx, lpfc, ufs, hisi_sas, mpi3mr,
mpt3sas, target). The biggest change (from my biased viewpoint) being
that the mpi3mr now attached to the SAS transport class, making it the
first fusion type device to do so.
Beyond the usual bug fixing and security class reworks, there aren't a
huge number of core changes"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (141 commits)
scsi: iscsi: iscsi_tcp: Fix null-ptr-deref while calling getpeername()
scsi: mpi3mr: Remove unnecessary cast
scsi: stex: Properly zero out the passthrough command structure
scsi: mpi3mr: Update driver version to 8.2.0.3.0
scsi: mpi3mr: Fix scheduling while atomic type bug
scsi: mpi3mr: Scan the devices during resume time
scsi: mpi3mr: Free enclosure objects during driver unload
scsi: mpi3mr: Handle 0xF003 Fault Code
scsi: mpi3mr: Graceful handling of surprise removal of PCIe HBA
scsi: mpi3mr: Schedule IRQ kthreads only on non-RT kernels
scsi: mpi3mr: Support new power management framework
scsi: mpi3mr: Update mpi3 header files
scsi: mpt3sas: Revert "scsi: mpt3sas: Fix ioc->base_readl() use"
scsi: mpt3sas: Revert "scsi: mpt3sas: Fix writel() use"
scsi: wd33c93: Remove dead code related to the long-gone config WD33C93_PIO
scsi: core: Add I/O timeout count for SCSI device
scsi: qedf: Populate sysfs attributes for vport
scsi: pm8001: Replace one-element array with flexible-array member
scsi: 3w-xxxx: Replace one-element array with flexible-array member
scsi: hptiop: Replace one-element array with flexible-array member in struct hpt_iop_request_ioctl_command()
...
|
|
The function __ata_change_queue_depth() uses the helper
ata_scsi_find_dev() to get the ata_device structure of a scsi device and
set that device maximum queue depth. However, when the ata device is
managed by libsas, ata_scsi_find_dev() returns NULL, turning
__ata_change_queue_depth() into a nop, which prevents the user from
setting the maximum queue depth of ATA devices used with libsas based
HBAs.
Fix this by renaming __ata_change_queue_depth() to
ata_change_queue_depth() and adding a pointer to the ata_device
structure of the target device as argument. This pointer is provided by
ata_scsi_change_queue_depth() using ata_scsi_find_dev() in the case of
a libata managed device and by sas_change_queue_depth() using
sas_to_ata_dev() in the case of a libsas managed ata device.
Signed-off-by: Damien Le Moal <[email protected]>
Tested-by: John Garry <[email protected]>
|
|
When executing SMP task failed, the smp_execute_task_sg() calls del_timer()
to delete "slow_task->timer". However, if the timer handler
sas_task_internal_timedout() is running, the del_timer() in
smp_execute_task_sg() will not stop it and a UAF will happen. The process
is shown below:
(thread 1) | (thread 2)
smp_execute_task_sg() | sas_task_internal_timedout()
... |
del_timer() |
... | ...
sas_free_task(task) |
kfree(task->slow_task) //FREE|
| task->slow_task->... //USE
Fix by calling del_timer_sync() in smp_execute_task_sg(), which makes sure
the timer handler have finished before the "task->slow_task" is
deallocated.
Link: https://lore.kernel.org/r/[email protected]
Fixes: 2908d778ab3e ("[SCSI] aic94xx: new driver")
Reviewed-by: Jason Yan <[email protected]>
Signed-off-by: Duoming Zhou <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Currently if a phy reset or enable phy is issued via sysfs when controller
is suspended, those operations will be ignored as SAS_HA_REGISTERED is
cleared. If RPM is enabled then we may aggressively suspend automatically.
In this case it may be difficult to enable or reset a phy via sysfs, so
resume the host in these scenarios.
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Similarly to sas report general and discovery responses, define the
structure struct smp_rps_resp to handle SATA PHY report responses using a
structure with a size that is exactly equal to the sas defined response
size.
With this change, struct smp_resp becomes unused and is removed.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
When compiling with gcc 12, several warnings are thrown by gcc when
compiling drivers/scsi/libsas/sas_expander.c, e.g.:
In function ‘sas_get_ex_change_count’,
inlined from ‘sas_find_bcast_dev’ at
drivers/scsi/libsas/sas_expander.c:1816:8:
drivers/scsi/libsas/sas_expander.c:1781:20: warning: array subscript
‘struct smp_resp[0]’ is partly outside array bounds of ‘unsigned
char[32]’ [-Warray-bounds]
1781 | if (rg_resp->result != SMP_RESP_FUNC_ACC) {
| ~~~~~~~^~~~~~~~
This is due to the use of the struct smp_resp to aggregate all possible
response types using a union but allocating a response buffer with a size
exactly equal to the size of the response type needed. This leads to access
to fields of struct smp_resp from an allocated memory area that is smaller
than the size of struct smp_resp.
Fix this by defining struct smp_rg_resp for sas report general responses.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
When compiling with gcc 12, several warnings are thrown by gcc when
compiling drivers/scsi/libsas/sas_expander.c, e.g.:
In function ‘sas_get_phy_change_count’,
inlined from ‘sas_find_bcast_phy.constprop’ at
drivers/scsi/libsas/sas_expander.c:1737:9:
drivers/scsi/libsas/sas_expander.c:1697:39: warning: array subscript
‘struct smp_resp[0]’ is partly outside array bounds of ‘unsigned
char[56]’ [-Warray-bounds]
1697 | *pcc = disc_resp->disc.change_count;
| ~~~~~~~~~~~~~~~^~~~~~~~~~~~~
This is due to the use of the struct smp_resp to aggregate all possible
response types using a union but allocating a response buffer with a size
exactly equal to the size of the response type needed. This leads to access
to fields of struct smp_resp from an allocated memory area that is smaller
than the size of struct smp_resp.
Fix this by defining struct smp_disc_resp for sas discovery operations.
Since this structure and the generic struct smp_resp are identical for
the little endian and big endian archs, move the definition of these
structures at the end of include/scsi/sas.h to avoid repeating their
definition.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Create function sas_ata_wait_after_reset() from sas_ata_hard_reset() as
some LLDDs may want to check for a remote ATA phy is up after reset.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Reviewed-by: Xiang Chen <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Use the common libsas internal abort functionality.
In addition, this driver has special handling for internal abort timeouts -
specifically whether to reset the controller in that instance, so extend
the API for that.
Timeout is now increased to 20 * Hz from 6 * Hz.
We also retry for failure now, but this should not make a difference.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Acked-by: Jack Wang <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add support for a "device" variant of internal abort, which will abort all
pending I/Os for a specific device.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Acked-by: Jack Wang <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The internal abort feature is common to hisi_sas and pm8001 HBAs, and the
driver support is similar also, so add a common handler.
Two modes of operation will be supported:
- single: Abort a single tagged command
- device: Abort all commands associated with a specific domain device
A new protocol is added, SAS_PROTOCOL_INTERNAL_ABORT, so the common queue
command API may be re-used.
Only add "single" support as a first step.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Acked-by: Jack Wang <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Sparse throws a warning about context imbalance ("different lock contexts
for basic block") in sas_form_port() as it gets confused with the fact that
a port is locked within one of the two search loops and unlocked afterward
outside of the search loops once the phy is added to the port. Since this
code is not easy to follow, improve it by factoring out the code adding the
phy to the port once the port is locked into the helper function
sas_form_port_add_phy(). This helper can then be called directly within the
port search loops, avoiding confusion and clearing the sparse warning.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Function queue_work() returns a bool, so use a bool to hold this value for
the return code from callers, which should make the code a tiny bit more
clear.
Also take this opportunity to condense the code of the those callers, such
as sas_queue_work(), as suggested by Damien.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Damien Le Moal <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Nobody checks the return codes, so make them return void. Indeed, if the
LLDD cannot send an event, nothing much can be done in the LLDD about it.
Also remove prototype for sas_notify_phy_event() in sas_internal.h, which
should not be there.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Xiang Chen <[email protected]>
Reviewed-by: Damien Le Moal <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
To detect if a command is NCQ, there is no need to test all possible NCQ
command codes. Instead, use ata_is_ncq() to test the command protocol.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
To detect for the DMA_NONE (no data transfer) DMA direction,
sas_ata_qc_issue() tests if the command protocol is ATA_PROT_NODATA. This
test does not include the ATA_CMD_NCQ_NON_DATA command as this command
protocol is defined as ATA_PROT_NCQ_NODATA (equal to ATA_PROT_FLAG_NCQ) and
not as ATA_PROT_NODATA.
To include both NCQ and non-NCQ commands when testing for the DMA_NONE DMA
direction, use "!ata_is_data()".
Link: https://lore.kernel.org/r/[email protected]
Fixes: 176ddd89171d ("scsi: libsas: Reset num_scatter if libata marks qc as NODATA")
Cc: [email protected]
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Jack Wang <[email protected]>
Signed-off-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a function to execute an ATA command using the TMF code, and use in the
hisi_sas driver. That driver needs to be able to issue the command on a
specific phy, so add an interface for that.
With that, hisi_sas_exec_internal_tmf_task() may be deleted.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a generic implementation of abort task TMF handler, and use in LLDDs.
With that, some LLDDs custom TMF functions can now be deleted.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a generic implementation of query task TMF handler, and use in LLDDs.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a generic implementation of LU reset TMF handler, and use in LLDDs.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a generic implementation of clear task set TMF handler, and use in
LLDDs.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a generic implementation of abort task set TMF handler, and use in
LLDDs.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The hisi_sas and pm8001 TMF handlers have some special processing for when
the TMF is aborted, so add a callback and fill it in for those drivers.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The pm8001 TMF handler has some special processing when the TMF completes,
so add a callback and fill it in for the pm8001 driver.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a function to issue an SSP TMF.
Add a temp prototype to keep make W=1 happy.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Drivers using libsas have to issue their own TMFs, and the code to do this
is duplicated between drivers.
Add a common function to handle TMFs. This will be also used for sending
ATA commands, but use name "tmf" as the purpose is similar between SCSI and
ATA in the usage.
The force_phy_id argument will be used for a hisi_sas HW bug workaround.
We check the sas task status stat field against TMF codes, as according to
the SAS v1.1 spec 9.2.2.5.3, if response_data is in datapres, then the
response code should be a TMF code - see table 128.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Move the SMP task handlers to the core host code as they will be re-used
for executing internal abort and TMF tasks.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
No LLDD sets exec status as SAS_SG_ERR, so remove support.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
As defined in table 126 of the SAS spec 1.1, use an enum for the DATAPRES
field, which makes reading the code easier.
Also change sas_ssp_task_response() to use a switch statement, which is
more suitable (than if-else), as suggested by Christoph.
Link: https://lore.kernel.org/r/[email protected]
Suggested-by: Xiang Chen <[email protected]>
Tested-by: Yihang Li <[email protected]>
Tested-by: Damien Le Moal <[email protected]>
Reviewed-by: Jack Wang <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
LLDD TMF callbacks may return linux or other error codes instead of TMF
codes. This may cause problems in sas_scsi_find_task() ->
.lldd_query_task(), as only TMF codes are handled there. As such, we may
not return a task_disposition type from sas_scsi_find_task(). Function
sas_eh_handle_sas_errors() only handles that type, and will only progress
error handling for those recognised types.
Return TASK_ABORT_FAILED upon exit on the assumption that the command may
still be alive and error handling should be escalated.
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Damien Le Moal <[email protected]>
Signed-off-by: John Garry <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Input parameter work_q is not unused in function sas_ata_eh(), so remove
it.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Task->task_state_flags is already set in function sas_alloc_task(), so
remove duplicated setting.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The callers of function sas_discover_event() do not check its return value.
The function also only ever returns 0, so use void instead.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Processing events such as PORTE_BROADCAST_RCVD may cause dependency issues
for runtime power management support. Such a problem would be that
handling a PORTE_BROADCAST_RCVD event requires that the host is resumed to
send SMP commands. However, in resuming the host, the phyup events
generated from re-enabling the phys are processed in the same workqueue as
the original PORTE_BROADCAST_RCVD event. As such, the host will never
finish resuming (as it waits for the phyup event processing), and then the
PORTE_BROADCAST_RCVD event can't be processed as the SMP commands are
blocked, and so we have a deadlock. Solve this problem by ensuring that
libsas keeps the host active until completely finished phy or port events,
such as PORTE_BYTES_DMAED. As such, we don't have to worry about resuming
the host for processing individual SMP commands in this example.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
During the processing of event PORT_BYTES_DMAED, the driver queues work
DISCE_DISCOVER_DOMAIN and then flushes workqueue ha->disco_q. If a new
phyup event occurs during resuming the controller, the work
PORTE_BYTES_DMAED of new phy occurs before suspended phy's. The work
DISCE_DISCOVER_DOMAIN of new phy requires an active SAS controller (it
needs to resume SAS controller by function scsi_sysfs_add_sdev() and some
other functions such as function add_device_link()). However, the
activation of the SAS controller requires completion of work
PORTE_BYTES_DMAED of suspended phys while it is blocked by new phy's work
on ha->event_q. So there is a deadlock and it is released only after resume
timeout.
To solve the issue, defer works of new phys during suspend and queue those
defer works after SAS controller becomes active.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
In the second part of function __sas_drain_work(), deferred work is queued.
This functionality is required other places so factor it out into the
function sas_queue_deferred_work().
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a flag SAS_HA_RESUMING and use it to indicate the state of resuming the
host controller.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
When sending SMP I/Os to the host we need to ensure that the host is not
suspended and can process the commands. This is a better approach than
replying on the host to resume itself to handle such commands. Use
pm_runtime_get_sync() and pm_runtime_put_sync() calls for the host when
executing SMP I/Os.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
If a new disk is inserted through an expander when the host was suspended,
it will not necessarily be detected as the topology is not re-scanned
during resume. To detect possible changes in topology during suspension,
insert a PORTE_BROADCAST_RCVD event per port when resuming to trigger a
revalidation.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Most places that use asd_sas_port->phy_list in libsas are protected by
spinlock asd_sas_port->phy_list_lock. However, there are still a few places
which miss the lock. Add it in those places.
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|