| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This has ses have the SCSI midlayer retry scsi_execute_cmd() errors instead
of driving them itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has read_capacity_10() have the SCSI midlayer retry errors instead of
driving them itself.
There are 2 behavior changes with this patch:
1. There is one behavior change where we no longer retry when
scsi_execute_cmd() returns < 0, but we should be ok. We don't need to
retry for failures like the queue being removed, and for the case where
there are no tags/reqs since the block layer waits/retries for us. For
possible memory allocation failures from blk_rq_map_kern() we use
GFP_NOIO, so retrying will probably not help.
2. For the specific UAs we checked for and retried, we would get
READ_CAPACITY_RETRIES_ON_RESET retries plus whatever retries were left
from the main loop's retries. Each UA now gets
READ_CAPACITY_RETRIES_ON_RESET retries, and the other errors get up to
3 retries. This is most likely ok, because
READ_CAPACITY_RETRIES_ON_RESET is already 10 and is not based on
anything specific like a spec or device, so the extra 3 we got from the
main loop was probably just an accident and is not going to help.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
It's common to get a UA when doing PR commands. It could be due to a target
restarting, transport level relogin or other PR commands like a release
causing it. The upper layers don't get the sense and in some cases have no
idea if it's a SCSI device, so this has the sd layer retry.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has scsi_report_lun_scan() have the SCSI midlayer retry errors instead
of driving them itself.
There is one behavior change where we no longer retry when
scsi_execute_cmd() returns < 0, but we should be ok. We don't need to retry
for failures like the queue being removed, and for the case where there are
no tags/reqs the block layer waits/retries for us. For possible memory
allocation failures from blk_rq_map_kern() we use GFP_NOIO, so retrying
will probably not help.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has scsi_mode_sense() have the SCSI midlayer retry UAs instead of
driving them itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has ch_do_scsi() have the SCSI midlayer retry UAs instead of driving
them itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
unit_attention is not used so remove it.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Martin Wilck <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has sd_sync_cache() have the SCSI midlayer retry errors instead of
driving them itself.
There is one behavior change where we no longer retry when
scsi_execute_cmd() returns < 0, but we should be ok. We don't need to retry
for failures like the queue being removed, and for the case where there are
no tags/reqs the block layer waits/retries for us. For possible memory
allocation failures from blk_rq_map_kern() we use GFP_NOIO, so retrying
will probably not help.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has spi_execute() have the SCSI midlayer retry UAs instead of driving
them.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has rdac have the SCSI midlayer retry errors instead of driving them
itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has hp_sw have the SCSI midlayer retry scsi_execute_cmd() errors
instead of driving them itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This simplifies sd_spinup_disk() so the SCSI midlayer retries errors for
it. Note that we retried every UA except Medium Not Present and also if
scsi_status_is_good() returned failed which would happen for all check
conditions. In this patch we use SCMD_FAILURE_STAT_ANY which will trigger
for the same conditions as when scsi_status_is_good() returns false and
there is status. This will cover all CCs including UAs so there is no
explicit failures array entry for UAs except for Medium Not Present which
we don't want to retry.
There is one behavior change where we no longer retry when
scsi_execute_cmd() returns < 0, but we should be ok. We don't need to retry
for failures like the queue being removed, and for the case where there are
no tags/reqs the block layer waits/retries for us. For possible memory
allocation failures from blk_rq_map_kern() we use GFP_NOIO, so retrying
will probably not help.
We do not handle the outside loop's retries because we want to sleep
between tries and we don't support that yet.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
We currently reuse the cmd buffer for the TUR and START_STOP commands
which requires us to reset the buffer when retrying. This has us use
separate buffers for the 2 commands so we can make them const and I think
it makes it easier to handle for retries but does not add too much extra to
the stack use.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Martin Wilck <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Description from: Martin Wilck <[email protected]>:
The SCSI mid layer doesn't retry commands after DID_TIME_OUT (see
scsi_noretry_cmd()). Packet loss in the fabric can cause spurious timeouts
during SCSI device probing, causing device probing to fail. This has been
observed in FCoE uplink failover tests, for example.
This patch fixes the issue by retrying the INQUIRY.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Martin Wilck <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
This has scsi_probe_lun() ask the SCSI midlayer to retry UAs instead of
driving them itself.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
For passthrough we don't retry any error which we get a check condition
for. This results in a lot of callers driving their own retries for all
UAs, specific UAs, NOT_READY, specific sense values or any type of failure.
This adds the core code to allow passthrough users to specify what errors
they want the SCSI midlayer to retry for them. We can then convert users to
drop a lot of their sense parsing and retry handling.
Signed-off-by: Mike Christie <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: John Garry <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Per filesystems/sysfs.rst, show() should only use sysfs_emit() or
sysfs_emit_at() when formatting the value to be returned to user space.
coccinelle complains that there are still a couple of functions that use
snprintf(). Convert them to sysfs_emit().
> ./drivers/scsi/pm8001/pm8001_ctl.c:883:8-16: WARNING: please use sysfs_emit
No functional change intended
CC: Jack Wang <[email protected]>
CC: James E.J. Bottomley <[email protected]>
CC: Martin K. Petersen <[email protected]>
CC: [email protected]
Signed-off-by: Li Zhijian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Jack Wang <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Per filesystems/sysfs.rst, show() should only use sysfs_emit() or
sysfs_emit_at() when formatting the value to be returned to user space.
coccinelle complains that there are still a couple of functions that use
snprintf(). Convert them to sysfs_emit().
> ./drivers/scsi/isci/init.c:140:8-16: WARNING: please use sysfs_emit
No functional change intended
CC: Artur Paszkiewicz <[email protected]>
CC: James E.J. Bottomley <[email protected]>
CC: Martin K. Petersen <[email protected]>
CC: [email protected]
Signed-off-by: Li Zhijian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Artur Paszkiewicz <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Per filesystems/sysfs.rst, show() should only use sysfs_emit() or
sysfs_emit_at() when formatting the value to be returned to user space.
coccinelle complains that there are still a couple of functions that use
snprintf(). Convert them to sysfs_emit().
> ./drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:3619:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:3625:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:3633:8-16: WARNING: please use sysfs_emit
No functional change intended
CC: Michael Cyr <[email protected]>
CC: James E.J. Bottomley <[email protected]>
CC: Martin K. Petersen <[email protected]>
CC: [email protected]
CC: [email protected]
Signed-off-by: Li Zhijian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Tyrel Datwyler <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Per filesystems/sysfs.rst, show() should only use sysfs_emit() or
sysfs_emit_at() when formatting the value to be returned to user space.
coccinelle complains that there are still a couple of functions that use
snprintf(). Convert them to sysfs_emit().
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3483:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3493:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3503:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3513:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3522:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/ibmvscsi/ibmvfc.c:3530:8-16: WARNING: please use sysfs_emit
No functional change intended
CC: Tyrel Datwyler <[email protected]>
CC: Michael Ellerman <[email protected]>
CC: Nicholas Piggin <[email protected]>
CC: Christophe Leroy <[email protected]>
CC: Aneesh Kumar K.V <[email protected]>
CC: Naveen N. Rao <[email protected]>
CC: James E.J. Bottomley <[email protected]>
CC: Martin K. Petersen <[email protected]>
CC: [email protected]
CC: [email protected]
Signed-off-by: Li Zhijian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Tyrel Datwyler <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Per filesystems/sysfs.rst, show() should only use sysfs_emit() or
sysfs_emit_at() when formatting the value to be returned to user space.
coccinelle complains that there are still a couple of functions that use
snprintf(). Convert them to sysfs_emit().
> ./drivers/scsi/fnic/fnic_attrs.c:17:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/fnic/fnic_attrs.c:23:8-16: WARNING: please use sysfs_emit
> ./drivers/scsi/fnic/fnic_attrs.c:31:8-16: WARNING: please use sysfs_emit
No functional change intended
CC: Satish Kharat <[email protected]>
CC: Sesidhar Baddela <[email protected]>
CC: Karan Tilak Kumar <[email protected]>
CC: James E.J. Bottomley <[email protected]>
CC: Martin K. Petersen <[email protected]>
CC: [email protected]
Signed-off-by: Li Zhijian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Karan Tilak Kumar <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
There is a general misunderstanding amongst engineers that {v}snprintf()
returns the length of the data *actually* encoded into the destination
array. However, as per the C99 standard {v}snprintf() really returns
the length of the data that *would have been* written if there were
enough space for it. This misunderstanding has led to buffer-overruns
in the past. It's generally considered safer to use the {v}scnprintf()
variants in their place (or even sprintf() in simple cases). So let's
do that.
Link: https://lwn.net/Articles/69419/
Link: https://github.com/KSPP/linux/issues/105
Cc: Adaptec OEM Raid Solutions <[email protected]>
Cc: PMC-Sierra, Inc <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
sysfs_emit()
Since snprintf() has the documented, but still rather strange trait of
returning the length of the data that *would have been* written to the
array if space were available, rather than the arguably more useful
length of data *actually* written, it is usually considered wise to use
something else instead in order to avoid confusion.
In the case of sysfs call-backs, new wrappers exist that do just that.
[mkp: removed unrelated whitespace cleanups]
Link: https://lwn.net/Articles/69419/
Link: https://github.com/KSPP/linux/issues/105
Cc: Richard Hirst <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
sysfs_emit()
Since snprintf() has the documented, but still rather strange trait of
returning the length of the data that *would have been* written to the
array if space were available, rather than the arguably more useful
length of data *actually* written, it is usually considered wise to use
something else instead in order to avoid confusion.
In the case of sysfs call-backs, new wrappers exist that do just that.
Link: https://lwn.net/Articles/69419/
Link: https://github.com/KSPP/linux/issues/105
Cc: Adam Radford <[email protected]>
Cc: Joel Jacobson <[email protected]>
Cc: de Melo <[email protected]>
Cc: Andre Hedrick <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
sysfs_emit()
Since snprintf() has the documented, but still rather strange trait of
returning the length of the data that *would have been* written to the
array if space were available, rather than the arguably more useful
length of data *actually* written, it is usually considered wise to use
something else instead in order to avoid confusion.
In the case of sysfs call-backs, new wrappers exist that do just that.
Link: https://lwn.net/Articles/69419/
Link: https://github.com/KSPP/linux/issues/105
Cc: Adam Radford <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
sysfs_emit()
Since snprintf() has the documented, but still rather strange trait of
returning the length of the data that *would have been* written to the
array if space were available, rather than the arguably more useful
length of data *actually* written, it is usually considered wise to use
something else instead in order to avoid confusion.
In the case of sysfs call-backs, new wrappers exist that do just that.
Link: https://lwn.net/Articles/69419/
Link: https://github.com/KSPP/linux/issues/105
Cc: Adam Radford <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Update driver version to 48.100.00.00.
Signed-off-by: Ranjan Kumar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Add a new IOCTL command MPT3ENABLEDIAGSBRRELOAD. As a part of firmware
update operation, applications use this IOCTL command to set the SBR reload
bit in the Host Diagnostic register. This permits HBA firmware to be
updated without powercycling the system.
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Signed-off-by: Ranjan Kumar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
chenxiang <[email protected]> says:
This series contains some fixes and cleanups including:
- Fix a deadlock issue related to automatic debugfs;
- Remove redundant checks for automatic debugfs;
- Check whether debugfs is enabled before removing or releasing it;
- Remove hisi_hba->timer for v3 hw;
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
hisi_hba->timer is not used for v3 hw but there are two places that some
operations related to hisi_hba->timer are called by v3 hw:
- Deleting the timer in function hisi_sas_v3_hw() which is only for v3 hw;
- Deleting the timer in function hisi_sas_controller_reset_prepare() which
is common for v1/v2/v3 hw.
We can remove the timer in the first case, but for the second scenario we
need to remove it only for v3 hw, so check hw->sht which is NULL only for
v3 hw before deleting hisi_hba->timer.
Signed-off-by: Xiang Chen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
hisi_sas debugfs remove should be executed only when debugfs is enabled.
Check whether debugfs is enabled and then remove it only if enabled.
Signed-off-by: Yihang Li <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
In commit 63f0733d07ce ("scsi: hisi_sas: Allocate DFX memory during dump
trigger"), the memory allocation time of the DFX is changed from device
initialization to dump occurs, so .debugfs_itct is not a valid address and
do not need to check.
The parameter hisi_sas_debugfs_enable is enough to check whether automatic
debugfs dump is triggered, so remove redunant checks.
Fixes: 63f0733d07ce ("scsi: hisi_sas: Allocate DFX memory during dump trigger")
Signed-off-by: Yihang Li <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
If we issue a disabling PHY command, the device attached with it will go
offline, if a 2 bit ECC error occurs at the same time, a hung task may be
found:
[ 4613.652388] INFO: task kworker/u256:0:165233 blocked for more than 120 seconds.
[ 4613.666297] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 4613.674809] task:kworker/u256:0 state:D stack: 0 pid:165233 ppid: 2 flags:0x00000208
[ 4613.683959] Workqueue: 0000:74:02.0_disco_q sas_revalidate_domain [libsas]
[ 4613.691518] Call trace:
[ 4613.694678] __switch_to+0xf8/0x17c
[ 4613.698872] __schedule+0x660/0xee0
[ 4613.703063] schedule+0xac/0x240
[ 4613.706994] schedule_timeout+0x500/0x610
[ 4613.711705] __down+0x128/0x36c
[ 4613.715548] down+0x240/0x2d0
[ 4613.719221] hisi_sas_internal_abort_timeout+0x1bc/0x260 [hisi_sas_main]
[ 4613.726618] sas_execute_internal_abort+0x144/0x310 [libsas]
[ 4613.732976] sas_execute_internal_abort_dev+0x44/0x60 [libsas]
[ 4613.739504] hisi_sas_internal_task_abort_dev.isra.0+0xbc/0x1b0 [hisi_sas_main]
[ 4613.747499] hisi_sas_dev_gone+0x174/0x250 [hisi_sas_main]
[ 4613.753682] sas_notify_lldd_dev_gone+0xec/0x2e0 [libsas]
[ 4613.759781] sas_unregister_common_dev+0x4c/0x7a0 [libsas]
[ 4613.765962] sas_destruct_devices+0xb8/0x120 [libsas]
[ 4613.771709] sas_do_revalidate_domain.constprop.0+0x1b8/0x31c [libsas]
[ 4613.778930] sas_revalidate_domain+0x60/0xa4 [libsas]
[ 4613.784716] process_one_work+0x248/0x950
[ 4613.789424] worker_thread+0x318/0x934
[ 4613.793878] kthread+0x190/0x200
[ 4613.797810] ret_from_fork+0x10/0x18
[ 4613.802121] INFO: task kworker/u256:4:316722 blocked for more than 120 seconds.
[ 4613.816026] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 4613.824538] task:kworker/u256:4 state:D stack: 0 pid:316722 ppid: 2 flags:0x00000208
[ 4613.833670] Workqueue: 0000:74:02.0 hisi_sas_rst_work_handler [hisi_sas_main]
[ 4613.841491] Call trace:
[ 4613.844647] __switch_to+0xf8/0x17c
[ 4613.848852] __schedule+0x660/0xee0
[ 4613.853052] schedule+0xac/0x240
[ 4613.856984] schedule_timeout+0x500/0x610
[ 4613.861695] __down+0x128/0x36c
[ 4613.865542] down+0x240/0x2d0
[ 4613.869216] hisi_sas_controller_prereset+0x58/0x1fc [hisi_sas_main]
[ 4613.876324] hisi_sas_rst_work_handler+0x40/0x8c [hisi_sas_main]
[ 4613.883019] process_one_work+0x248/0x950
[ 4613.887732] worker_thread+0x318/0x934
[ 4613.892204] kthread+0x190/0x200
[ 4613.896118] ret_from_fork+0x10/0x18
[ 4613.900423] INFO: task kworker/u256:1:348985 blocked for more than 121 seconds.
[ 4613.914341] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 4613.922852] task:kworker/u256:1 state:D stack: 0 pid:348985 ppid: 2 flags:0x00000208
[ 4613.931984] Workqueue: 0000:74:02.0_event_q sas_port_event_worker [libsas]
[ 4613.939549] Call trace:
[ 4613.942702] __switch_to+0xf8/0x17c
[ 4613.946892] __schedule+0x660/0xee0
[ 4613.951083] schedule+0xac/0x240
[ 4613.955015] schedule_timeout+0x500/0x610
[ 4613.959725] wait_for_common+0x200/0x610
[ 4613.964349] wait_for_completion+0x3c/0x5c
[ 4613.969146] flush_workqueue+0x198/0x790
[ 4613.973776] sas_porte_broadcast_rcvd+0x1e8/0x320 [libsas]
[ 4613.979960] sas_port_event_worker+0x54/0xa0 [libsas]
[ 4613.985708] process_one_work+0x248/0x950
[ 4613.990420] worker_thread+0x318/0x934
[ 4613.994868] kthread+0x190/0x200
[ 4613.998800] ret_from_fork+0x10/0x18
This is because when the device goes offline, we obtain the hisi_hba
semaphore and send the ABORT_DEV command to the device. However, the
internal abort timed out due to the 2 bit ECC error and triggers automatic
dump. In addition, since the hisi_hba semaphore has been obtained, the dump
cannot be executed and the controller cannot be reset.
Therefore, the deadlocks occur on the following circular dependencies:
hisi_sas_dev_gone() -> down() -> hisi_sas_internal_task_abort_dev() -> ...
-> hisi_sas_internal_abort_timeout() -> down().
The deadlock is triggered only when the timeout occurs during device goes
offline. To fix this issue, use .rst_ha_timeout to distinguish the scenario
where a device goes offline from other scenarios.
Fixes: 2ff07b5c6fe9 ("scsi: hisi_sas: Directly call register snapshot instead of using workqueue")
Signed-off-by: Yihang Li <[email protected]>
Signed-off-by: Xiang Chen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
No functional modification involved.
drivers/scsi/fnic/fnic_scsi.c:1964 fnic_abort_cmd() warn: inconsistent indenting.
Reported-by: Abaci Robot <[email protected]>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=7930
Signed-off-by: Jiapeng Chong <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Karan Tilak Kumar <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
To ensure that the same ID is not obtained during concurrent execution of
the probe, an ida is used to manage the mrioc's ID.
Signed-off-by: Guixin Liu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Lee Duncan <[email protected]>
Reviewed-by: Martin Wilck <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
strncpy() is deprecated for use on NUL-terminated destination strings [1]
and as such we should prefer more robust and less ambiguous string
interfaces.
We don't need the NUL-padding behavior that strncpy() provides as vscsi is
NUL-allocated in ibmvscsis_probe() which proceeds to call
ibmvscsis_adapter_info():
| vscsi = kzalloc(sizeof(*vscsi), GFP_KERNEL);
ibmvscsis_probe() -> ibmvscsis_handle_crq() -> ibmvscsis_parse_command()
-> ibmvscsis_mad() -> ibmvscsis_process_mad() -> ibmvscsis_adapter_info()
Following the same idea, `partition_name` is defiend as:
| static char partition_name[PARTITION_NAMELEN] = "UNKNOWN";
... which is NUL-padded already, meaning strscpy() is the best option.
Considering the above, a suitable replacement is strscpy() [2] due to the
fact that it guarantees NUL-termination on the destination buffer without
unnecessarily NUL-padding.
However, for cap->name and info let's use strscpy_pad() as they are
allocated via dma_alloc_coherent():
| cap = dma_alloc_coherent(&vscsi->dma_dev->dev, olen, &token,
| GFP_ATOMIC);
&
| info = dma_alloc_coherent(&vscsi->dma_dev->dev, sizeof(*info), &token,
| GFP_ATOMIC);
Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1]
Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2]
Link: https://github.com/KSPP/linux/issues/90
Cc: [email protected]
Signed-off-by: Justin Stitt <[email protected]>
Link: https://lore.kernel.org/r/20231212-strncpy-drivers-scsi-ibmvscsi_tgt-ibmvscsi_tgt-c-v2-1-bdb9a7cd96c8@google.com
Acked-by: Tyrel Datwyler <[email protected]>
Reviewed-by: Kees Cook <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The variable 'retval' is being assigned a value that is not being read
afterwards. The assignment is redundant and can be removed.
Cleans up clang scan warning:
Although the value stored to 'retval' is used in the enclosing
expression, the value is never actually read from 'retval'
[deadcode.DeadStores]
Signed-off-by: Colin Ian King <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Current code uses the specified ring buffer size (either the default of 128
Kbytes or a module parameter specified value) to encompass the one page
ring buffer header plus the actual ring itself. When the page size is 4K,
carving off one page for the header isn't significant. But when the page
size is 64K on ARM64, only half of the default 128 Kbytes is left for the
actual ring. While this doesn't break anything, the smaller ring size
could be a performance bottleneck.
Fix this by applying the VMBUS_RING_SIZE macro to the specified ring buffer
size. This macro adds a page for the header, and rounds up the size to a
page boundary, using the page size for which the kernel is built. Use this
new size for subsequent ring buffer calculations. For example, on ARM64
with 64K page size and the default ring size, this results in the actual
ring being 128 Kbytes, which is intended.
Cc: [email protected] # 5.15.x
Signed-off-by: Michael Kelley <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Inside scsi_eh_wakeup(), scsi_host_busy() is called & checked with host
lock every time for deciding if error handler kthread needs to be waken up.
This can be too heavy in case of recovery, such as:
- N hardware queues
- queue depth is M for each hardware queue
- each scsi_host_busy() iterates over (N * M) tag/requests
If recovery is triggered in case that all requests are in-flight, each
scsi_eh_wakeup() is strictly serialized, when scsi_eh_wakeup() is called
for the last in-flight request, scsi_host_busy() has been run for (N * M -
1) times, and request has been iterated for (N*M - 1) * (N * M) times.
If both N and M are big enough, hard lockup can be triggered on acquiring
host lock, and it is observed on mpi3mr(128 hw queues, queue depth 8169).
Fix the issue by calling scsi_host_busy() outside the host lock. We don't
need the host lock for getting busy count because host the lock never
covers that.
[mkp: Drop unnecessary 'busy' variables pointed out by Bart]
Cc: Ewan Milne <[email protected]>
Fixes: 6eb045e092ef ("scsi: core: avoid host-wide host_busy counter for scsi_mq")
Signed-off-by: Ming Lei <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Ewan D. Milne <[email protected]>
Reviewed-by: Sathya Prakash Veerichetty <[email protected]>
Tested-by: Sathya Prakash Veerichetty <[email protected]>
Reviewed-by: Bart Van Assche <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Both "model" and "strflags" are passed to "%s" even when one or both are
NULL.
It is safe because vsprintf() would detect the NULL pointer and print
"(null)". But it is a kernel-specific feature and compiler warns about it:
<warning>
In file included from include/linux/kernel.h:19,
from arch/x86/include/asm/percpu.h:27,
from arch/x86/include/asm/current.h:6,
from include/linux/sched.h:12,
from include/linux/blkdev.h:5,
from drivers/scsi/scsi_devinfo.c:3:
drivers/scsi/scsi_devinfo.c: In function 'scsi_dev_info_list_add_str':
>> include/linux/printk.h:434:44: warning: '%s' directive argument is null [-Wformat-overflow=]
434 | #define printk(fmt, ...) printk_index_wrap(_printk, fmt, ##__VA_ARGS__)
| ^
include/linux/printk.h:430:3: note: in definition of macro 'printk_index_wrap'
430 | _p_func(_fmt, ##__VA_ARGS__); \
| ^~~~~~~
drivers/scsi/scsi_devinfo.c:551:4: note: in expansion of macro 'printk'
551 | printk(KERN_ERR "%s: bad dev info string '%s' '%s'"
| ^~~~~~
drivers/scsi/scsi_devinfo.c:552:14: note: format string is defined here
552 | " '%s'\n", __func__, vendor, model,
| ^~
</warning>
Do not rely on the kernel specific behavior and print the message a safe
way.
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Signed-off-by: Petr Mladek <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Bart Van Assche <[email protected]>
Acked-by: Chris Down <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
The runtime suspend timer delay is a const value in scsi_host_template
which a host driver cannot modify at runtime. Move the delay to Scsi_Host
to allow a driver to update it.
Signed-off-by: Peter Wang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Bart Van Assche <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Pull in staged fixes for 6.8.
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Pull SCSI updates from James Bottomley:
"Final round of fixes that came in too late to send in the first
request.
It's nine bug fixes and one version update (because of a bug fix) and
one set of PCI ID additions. There's one bug fix in the core which is
really a one liner (except that an additional sdev pointer was added
for convenience) and the rest are in drivers"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: target: core: Add TMF to tmr_list handling
scsi: core: Kick the requeue list after inserting when flushing
scsi: fnic: unlock on error path in fnic_queuecommand()
scsi: fcoe: Fix unsigned comparison with zero in store_ctlr_mode()
scsi: mpi3mr: Fix mpi3mr_fw.c kernel-doc warnings
scsi: smartpqi: Bump driver version to 2.1.26-030
scsi: smartpqi: Fix logical volume rescan race condition
scsi: smartpqi: Add new controller PCI IDs
scsi: ufs: qcom: Remove unnecessary goto statement from ufs_qcom_config_esi()
scsi: ufs: core: Remove the ufshcd_hba_exit() call from ufshcd_async_scan()
scsi: ufs: core: Simplify power management during async scan
|
|
Pull virtio updates from Michael Tsirkin:
- vdpa/mlx5: support for resumable vqs
- virtio_scsi: mq_poll support
- 3virtio_pmem: support SHMEM_REGION
- virtio_balloon: stay awake while adjusting balloon
- virtio: support for no-reset virtio PCI PM
- Fixes, cleanups
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
vdpa/mlx5: Add mkey leak detection
vdpa/mlx5: Introduce reference counting to mrs
vdpa/mlx5: Use vq suspend/resume during .set_map
vdpa/mlx5: Mark vq state for modification in hw vq
vdpa/mlx5: Mark vq addrs for modification in hw vq
vdpa/mlx5: Introduce per vq and device resume
vdpa/mlx5: Allow modifying multiple vq fields in one modify command
vdpa/mlx5: Expose resumable vq capability
vdpa: Block vq property changes in DRIVER_OK
vdpa: Track device suspended state
scsi: virtio_scsi: Add mq_poll support
virtio_pmem: support feature SHMEM_REGION
virtio_balloon: stay awake while adjusting balloon
vdpa: Remove usage of the deprecated ida_simple_xx() API
virtio: Add support for no-reset virtio PCI PM
virtio_net: fix missing dma unmap for resize
vhost-vdpa: account iommu allocations
vdpa: Fix an error handling path in eni_vdpa_probe()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing updates from Steven Rostedt:
- Allow kernel trace instance creation to specify what events are
created
Inside the kernel, a subsystem may create a tracing instance that it
can use to send events to user space. This sub-system may not care
about the thousands of events that exist in eventfs. Allow the
sub-system to specify what sub-systems of events it cares about, and
only those events are exposed to this instance.
- Allow the ring buffer to be broken up into bigger sub-buffers than
just the architecture page size.
A new tracefs file called "buffer_subbuf_size_kb" is created. The
user can now specify a minimum size the sub-buffer may be in
kilobytes. Note, that the implementation currently make the
sub-buffer size a power of 2 pages (1, 2, 4, 8, 16, ...) but the user
only writes in kilobyte size, and the sub-buffer will be updated to
the next size that it will can accommodate it. If the user writes in
10, it will change the size to be 4 pages on x86 (16K), as that is
the next available size that can hold 10K pages.
- Update the debug output when a corrupt time is detected in the ring
buffer. If the ring buffer detects inconsistent timestamps, there's a
debug config options that will dump the contents of the meta data of
the sub-buffer that is used for debugging. Add some more information
to this dump that helps with debugging.
- Add more timestamp debugging checks (only triggers when the config is
enabled)
- Increase the trace_seq iterator to 2 page sizes.
- Allow strings written into tracefs_marker to be larger. Up to just
under 2 page sizes (based on what trace_seq can hold).
- Increase the trace_maker_raw write to be as big as a sub-buffer can
hold.
- Remove 32 bit time stamp logic, now that the rb_time_cmpxchg() has
been removed.
- More selftests were added.
- Some code clean ups as well.
* tag 'trace-v6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (29 commits)
ring-buffer: Remove stale comment from ring_buffer_size()
tracing histograms: Simplify parse_actions() function
tracing/selftests: Remove exec permissions from trace_marker.tc test
ring-buffer: Use subbuf_order for buffer page masking
tracing: Update subbuffer with kilobytes not page order
ringbuffer/selftest: Add basic selftest to test changing subbuf order
ring-buffer: Add documentation on the buffer_subbuf_order file
ring-buffer: Just update the subbuffers when changing their allocation order
ring-buffer: Keep the same size when updating the order
tracing: Stop the tracing while changing the ring buffer subbuf size
tracing: Update snapshot order along with main buffer order
ring-buffer: Make sure the spare sub buffer used for reads has same size
ring-buffer: Do no swap cpu buffers if order is different
ring-buffer: Clear pages on error in ring_buffer_subbuf_order_set() failure
ring-buffer: Read and write to ring buffers with custom sub buffer size
ring-buffer: Set new size of the ring buffer sub page
ring-buffer: Add interface for configuring trace sub buffer size
ring-buffer: Page size per ring buffer
ring-buffer: Have ring_buffer_print_page_header() be able to access ring_buffer_iter
ring-buffer: Check if absolute timestamp goes backwards
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci
Pull pci updates from Bjorn Helgaas:
"Enumeration:
- Reserve ECAM so we don't assign it to PCI BARs; this works around
bugs where BIOS included ECAM in a PNP0A03 host bridge window,
didn't reserve it via a PNP0C02 motherboard device, and didn't
allocate space for SR-IOV VF BARs (Bjorn Helgaas)
- Add MMCONFIG/ECAM debug logging (Bjorn Helgaas)
- Rename 'MMCONFIG' to 'ECAM' to match spec usage (Bjorn Helgaas)
- Log device type (Root Port, Switch Port, etc) during enumeration
(Bjorn Helgaas)
- Log bridges before downstream devices so the dmesg order is more
logical (Bjorn Helgaas)
- Log resource names (BAR 0, VF BAR 0, bridge window, etc)
consistently instead of a mix of names and "reg 0x10" (Puranjay
Mohan, Bjorn Helgaas)
- Fix 64GT/s effective data rate calculation to use 1b/1b encoding
rather than the 8b/10b or 128b/130b used by lower rates (Ilpo
Järvinen)
- Use PCI_HEADER_TYPE_* instead of literals in x86, powerpc, SCSI
lpfc (Ilpo Järvinen)
- Clean up open-coded PCIBIOS return code mangling (Ilpo Järvinen)
Resource management:
- Restructure pci_dev_for_each_resource() to avoid computing the
address of an out-of-bounds array element (the bounds check was
performed later so the element was never actually *read*, but it's
nicer to avoid even computing an out-of-bounds address) (Andy
Shevchenko)
Driver binding:
- Convert pci-host-common.c platform .remove() callback to
.remove_new() returning 'void' since it's not useful to return
error codes here (Uwe Kleine-König)
- Convert exynos, keystone, kirin from .remove() to .remove_new(),
which returns void instead of int (Uwe Kleine-König)
- Drop unused struct pci_driver.node member (Mathias Krause)
Virtualization:
- Add ACS quirk for more Zhaoxin Root Ports (LeoLiuoc)
Error handling:
- Log AER errors as "Correctable" (not "Corrected") or
"Uncorrectable" to match spec terminology (Bjorn Helgaas)
- Decode Requester ID when no error info found instead of printing
the raw hex value (Bjorn Helgaas)
Endpoint framework:
- Use a unique test pattern for each BAR in the pci_endpoint_test to
make it easier to debug address translation issues (Niklas Cassel)
Broadcom STB PCIe controller driver:
- Add DT property "brcm,clkreq-mode" and driver support for different
CLKREQ# modes to make ASPM L1.x states possible (Jim Quinlan)
Freescale Layerscape PCIe controller driver:
- Add suspend/resume support for Layerscape LS1043a and LS1021a,
including software-managed PME_Turn_Off and transitions between L0,
L2/L3_Ready Link states (Frank Li)
MediaTek PCIe controller driver:
- Clear MSI interrupt status before handler to avoid missing MSIs
that occur after the handler (qizhong cheng)
MediaTek PCIe Gen3 controller driver:
- Update mediatek-gen3 translation window setup to handle MMIO space
that is not a power of two in size (Jianjun Wang)
Qualcomm PCIe controller driver:
- Increase qcom iommu-map maxItems to accommodate SDX55 (five
entries) and SDM845 (sixteen entries) (Krzysztof Kozlowski)
- Describe qcom,pcie-sc8180x clocks and resets accurately (Krzysztof
Kozlowski)
- Describe qcom,pcie-sm8150 clocks and resets accurately (Krzysztof
Kozlowski)
- Correct the qcom "reset-name" property, previously incorrectly
called "reset-names" (Krzysztof Kozlowski)
- Document qcom,pcie-sm8650, based on qcom,pcie-sm8550 (Neil
Armstrong)
Renesas R-Car PCIe controller driver:
- Replace of_device.h with explicit of.h include to untangle header
usage (Rob Herring)
- Add DT and driver support for optional miniPCIe 1.5v and 3.3v
regulators on KingFisher (Wolfram Sang)
SiFive FU740 PCIe controller driver:
- Convert fu740 CONFIG_PCIE_FU740 dependency from SOC_SIFIVE to
ARCH_SIFIVE (Conor Dooley)
Synopsys DesignWare PCIe controller driver:
- Align iATU mapping for endpoint MSI-X (Niklas Cassel)
- Drop "host_" prefix from struct dw_pcie_host_ops members (Yoshihiro
Shimoda)
- Drop "ep_" prefix from struct dw_pcie_ep_ops members (Yoshihiro
Shimoda)
- Rename struct dw_pcie_ep_ops.func_conf_select() to
.get_dbi_offset() to be more descriptive (Yoshihiro Shimoda)
- Add Endpoint DBI accessors to encapsulate offset lookups (Yoshihiro
Shimoda)
TI J721E PCIe driver:
- Add j721e DT and driver support for 'num-lanes' for devices that
support x1, x2, or x4 Links (Matt Ranostay)
- Add j721e DT compatible strings and driver support for j784s4 (Matt
Ranostay)
- Make TI J721E Kconfig depend on ARCH_K3 since the hardware is
specific to those TI SoC parts (Peter Robinson)
TI Keystone PCIe controller driver:
- Hold power management references to all PHYs while enabling them to
avoid a race when one provides clocks to others (Siddharth
Vadapalli)
Xilinx XDMA PCIe controller driver:
- Remove redundant dev_err(), since platform_get_irq() and
platform_get_irq_byname() already log errors (Yang Li)
- Fix uninitialized symbols in xilinx_pl_dma_pcie_setup_irq()
(Krzysztof Wilczyński)
- Fix xilinx_pl_dma_pcie_init_irq_domain() error return when
irq_domain_add_linear() fails (Harshit Mogalapalli)
MicroSemi Switchtec management driver:
- Do dma_mrpc cleanup during switchtec_pci_remove() to match its devm
ioremapping in switchtec_pci_probe(). Previously the cleanup was
done in stdev_release(), which used stale pointers if stdev->cdev
happened to be open when the PCI device was removed (Daniel
Stodden)
Miscellaneous:
- Convert interrupt terminology from "legacy" to "INTx" to be more
specific and match spec terminology (Damien Le Moal)
- In dw-xdata-pcie, pci_endpoint_test, and vmd, replace usage of
deprecated ida_simple_*() API with ida_alloc() and ida_free()
(Christophe JAILLET)"
* tag 'pci-v6.8-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci: (97 commits)
PCI: Fix kernel-doc issues
PCI: brcmstb: Configure HW CLKREQ# mode appropriate for downstream device
dt-bindings: PCI: brcmstb: Add property "brcm,clkreq-mode"
PCI: mediatek-gen3: Fix translation window size calculation
PCI: mediatek: Clear interrupt status before dispatching handler
PCI: keystone: Fix race condition when initializing PHYs
PCI: xilinx-xdma: Fix error code in xilinx_pl_dma_pcie_init_irq_domain()
PCI: xilinx-xdma: Fix uninitialized symbols in xilinx_pl_dma_pcie_setup_irq()
PCI: rcar-gen4: Fix -Wvoid-pointer-to-enum-cast error
PCI: iproc: Fix -Wvoid-pointer-to-enum-cast warning
PCI: dwc: Add dw_pcie_ep_{read,write}_dbi[2] helpers
PCI: dwc: Rename .func_conf_select to .get_dbi_offset in struct dw_pcie_ep_ops
PCI: dwc: Rename .ep_init to .init in struct dw_pcie_ep_ops
PCI: dwc: Drop host prefix from struct dw_pcie_host_ops members
misc: pci_endpoint_test: Use a unique test pattern for each BAR
PCI: j721e: Make TI J721E depend on ARCH_K3
PCI: j721e: Add TI J784S4 PCIe configuration
PCI/AER: Use explicit register sizes for struct members
PCI/AER: Decode Requester ID when no error info found
PCI/AER: Use 'Correctable' and 'Uncorrectable' spec terms for errors
...
|
|
The variable 'rb' is being assigned a value but it isn't being read
afterwards. The assignment is redundant and so 'rb' can be removed.
Cleans up clang scan build warning:
warning: Although the value stored to 'rb' is used in the enclosing
expression, the value is never actually read from 'rb'[deadcode.DeadStores]
Signed-off-by: Colin Ian King <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
virtqueue_enable_cb() will call virtqueue_poll() which will check if queue
is broken at beginning, so remove the virtqueue_is_broken() call
Signed-off-by: Li RongQing <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Stefan Hajnoczi <[email protected]>
Acked-by: Michael S. Tsirkin <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
Clang static complains that Value stored to 'status' is never read. Return
'status' rather than 'SCI_SUCCESS'.
Fixes: f1f52e75939b ("isci: uplevel request infrastructure")
Signed-off-by: Su Hui <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Artur Paszkiewicz <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|
|
When libata calls ata_link_abort() to abort all ata queued commands, it
calls blk_abort_request() on the SCSI command representing each QC.
This causes scsi_timeout() to be called, which calls scsi_eh_scmd_add() for
each SCSI command.
scsi_eh_scmd_add() sets the SCSI host to state recovery, and then adds the
command to shost->eh_cmd_q.
This will wake up the SCSI EH, and eventually the libata EH strategy
handler will be called, which calls scsi_eh_flush_done_q() to either flush
retry or flush finish each failed command.
The commands that are flush retried by scsi_eh_flush_done_q() are done so
using scsi_queue_insert().
Before commit 8b566edbdbfb ("scsi: core: Only kick the requeue list if
necessary"), __scsi_queue_insert() called blk_mq_requeue_request() with the
second argument set to true, indicating that it should always kick/run the
requeue list after inserting.
After commit 8b566edbdbfb ("scsi: core: Only kick the requeue list if
necessary"), __scsi_queue_insert() does not kick/run the requeue list after
inserting, if the current SCSI host state is recovery (which is the case in
the libata example above).
This optimization is probably fine in most cases, as I can only assume that
most often someone will eventually kick/run the queues.
However, that is not the case for scsi_eh_flush_done_q(), where we can see
that the request gets inserted to the requeue list, but the queue is never
started after the request has been inserted, leading to the block layer
waiting for the completion of command that never gets to run.
Since scsi_eh_flush_done_q() is called by SCSI EH context, the SCSI host
state is most likely always in recovery when this function is called.
Thus, let scsi_eh_flush_done_q() explicitly kick the requeue list after
inserting a flush retry command, so that scsi_eh_flush_done_q() keeps the
same behavior as before commit 8b566edbdbfb ("scsi: core: Only kick the
requeue list if necessary").
Simple reproducer for the libata example above:
$ hdparm -Y /dev/sda
$ echo 1 > /sys/class/scsi_device/0\:0\:0\:0/device/delete
Fixes: 8b566edbdbfb ("scsi: core: Only kick the requeue list if necessary")
Reported-by: Kevin Locke <[email protected]>
Closes: https://lore.kernel.org/linux-scsi/[email protected]/
Signed-off-by: Niklas Cassel <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Bart Van Assche <[email protected]>
Reviewed-by: Damien Le Moal <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
|