* [PATCH 1/2] zfcp: fix missing zfcp_port reference put on -EBUSY from port_remove [not found] <1558617826-30129-1-git-send-email-maier@linux.ibm.com> @ 2019-05-23 13:23 ` Steffen Maier 2019-05-23 13:23 ` Steffen Maier 2019-05-30 1:53 ` [PATCH 0/2] zfcp fixes for v5.2-rcX Martin K. Petersen 2 siblings, 0 replies; 5+ messages in thread From: Steffen Maier @ 2019-05-23 13:23 UTC (permalink / raw) To: James E.J. Bottomley, Martin K. Petersen Cc: linux-scsi, linux-s390, Benjamin Block, Heiko Carstens, Steffen Maier, stable With this early return due to zfcp_unit child(ren), we don't use the zfcp_port reference from the earlier zfcp_get_port_by_wwpn()any more and need to put it. Signed-off-by: Steffen Maier <maier@linux.ibm.com> Fixes: d99b601b6338 ("[SCSI] zfcp: restore refcount check on port_remove") Cc: <stable@vger.kernel.org> #3.7+ Reviewed-by: Jens Remus <jremus@linux.ibm.com> Reviewed-by: Benjamin Block <bblock@linux.ibm.com> --- drivers/s390/scsi/zfcp_sysfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index b277be6f7611..2d78732b270b 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -261,6 +261,7 @@ static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, if (atomic_read(&port->units) > 0) { retval = -EBUSY; mutex_unlock(&zfcp_sysfs_port_units_mutex); + put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ goto out; } /* port is about to be removed, so no more unit_add */ -- 2.17.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 1/2] zfcp: fix missing zfcp_port reference put on -EBUSY from port_remove @ 2019-05-23 13:23 ` Steffen Maier 0 siblings, 0 replies; 5+ messages in thread From: Steffen Maier @ 2019-05-23 13:23 UTC (permalink / raw) To: James E.J. Bottomley, Martin K. Petersen Cc: linux-scsi, linux-s390, Benjamin Block, Heiko Carstens, Steffen Maier, stable With this early return due to zfcp_unit child(ren), we don't use the zfcp_port reference from the earlier zfcp_get_port_by_wwpn()any more and need to put it. Signed-off-by: Steffen Maier <maier@linux.ibm.com> Fixes: d99b601b6338 ("[SCSI] zfcp: restore refcount check on port_remove") Cc: <stable@vger.kernel.org> #3.7+ Reviewed-by: Jens Remus <jremus@linux.ibm.com> Reviewed-by: Benjamin Block <bblock@linux.ibm.com> --- drivers/s390/scsi/zfcp_sysfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index b277be6f7611..2d78732b270b 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -261,6 +261,7 @@ static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, if (atomic_read(&port->units) > 0) { retval = -EBUSY; mutex_unlock(&zfcp_sysfs_port_units_mutex); + put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ goto out; } /* port is about to be removed, so no more unit_add */ -- 2.17.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 2/2] zfcp: fix to prevent port_remove with pure auto scan LUNs (only sdevs) [not found] <1558617826-30129-1-git-send-email-maier@linux.ibm.com> @ 2019-05-23 13:23 ` Steffen Maier 2019-05-23 13:23 ` Steffen Maier 2019-05-30 1:53 ` [PATCH 0/2] zfcp fixes for v5.2-rcX Martin K. Petersen 2 siblings, 0 replies; 5+ messages in thread From: Steffen Maier @ 2019-05-23 13:23 UTC (permalink / raw) To: James E.J. Bottomley, Martin K. Petersen Cc: linux-scsi, linux-s390, Benjamin Block, Heiko Carstens, Steffen Maier, stable When the user tries to remove a zfcp port via sysfs, we only rejected it if there are zfcp unit children under the port. With purely automatically scanned LUNs there are no zfcp units but only SCSI devices. In such cases, the port_remove erroneously continued. We close the port and this implicitly closes all LUNs under the port. The SCSI devices survive with their private zfcp_scsi_dev still holding a reference to the "removed" zfcp_port (still allocated but invisible in sysfs) [zfcp_get_port_by_wwpn in zfcp_scsi_slave_alloc]. This is not a problem as long as the fc_rport stays blocked. Once (auto) port scan brings back the removed port, we unblock its fc_rport again by design. However, there is no mechanism that would recover (open) the LUNs under the port (no "ersfs_3" without zfcp_unit [zfcp_erp_strategy_followup_success]). Any pending or new I/O to such LUN leads to repeated: Done: NEEDS_RETRY Result: hostbyte=DID_IMM_RETRY driverbyte=DRIVER_OK See also v4.10 commit 6f2ce1c6af37 ("scsi: zfcp: fix rport unblock race with LUN recovery"). Even a manual LUN recovery (echo 0 > /sys/bus/scsi/devices/H:C:T:L/zfcp_failed) does not help, as the LUN links to the old "removed" port which remains to lack ZFCP_STATUS_COMMON_RUNNING [zfcp_erp_required_act]. The only workaround is to first ensure that the fc_rport is blocked (e.g. port_remove again in case it was re-discovered by (auto) port scan), then delete the SCSI devices, and finally re-discover by (auto) port scan. The port scan includes an fc_rport unblock, which in turn triggers a new scan on the scsi target to freshly get new pure auto scan LUNs. Fix this by rejecting port_remove also if there are SCSI devices (even without any zfcp_unit) under this port. Re-use mechanics from v3.7 commit d99b601b6338 ("[SCSI] zfcp: restore refcount check on port_remove"). However, we have to give up zfcp_sysfs_port_units_mutex earlier in unit_add to prevent a deadlock with scsi_host scan taking shost->scan_mutex first and then zfcp_sysfs_port_units_mutex now in our zfcp_scsi_slave_alloc(). Signed-off-by: Steffen Maier <maier@linux.ibm.com> Fixes: b62a8d9b45b9 ("[SCSI] zfcp: Use SCSI device data zfcp scsi dev instead of zfcp unit") Fixes: f8210e34887e ("[SCSI] zfcp: Allow midlayer to scan for LUNs when running in NPIV mode") Cc: <stable@vger.kernel.org> #2.6.37+ Reviewed-by: Benjamin Block <bblock@linux.ibm.com> --- drivers/s390/scsi/zfcp_ext.h | 1 + drivers/s390/scsi/zfcp_scsi.c | 9 ++++++ drivers/s390/scsi/zfcp_sysfs.c | 54 ++++++++++++++++++++++++++++++---- drivers/s390/scsi/zfcp_unit.c | 8 ++++- 4 files changed, 65 insertions(+), 7 deletions(-) diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index c6acca521ffe..31e8a7240fd7 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -167,6 +167,7 @@ extern const struct attribute_group *zfcp_port_attr_groups[]; extern struct mutex zfcp_sysfs_port_units_mutex; extern struct device_attribute *zfcp_sysfs_sdev_attrs[]; extern struct device_attribute *zfcp_sysfs_shost_attrs[]; +bool zfcp_sysfs_port_is_removing(const struct zfcp_port *const port); /* zfcp_unit.c */ extern int zfcp_unit_add(struct zfcp_port *, u64); diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 221d0dfb8493..e9ded2befa0d 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -129,6 +129,15 @@ static int zfcp_scsi_slave_alloc(struct scsi_device *sdev) zfcp_sdev->erp_action.port = port; + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (zfcp_sysfs_port_is_removing(port)) { + /* port is already gone */ + mutex_unlock(&zfcp_sysfs_port_units_mutex); + put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ + return -ENXIO; + } + mutex_unlock(&zfcp_sysfs_port_units_mutex); + unit = zfcp_unit_find(port, zfcp_scsi_dev_lun(sdev)); if (unit) put_device(&unit->dev); diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index 2d78732b270b..af197e2b3e69 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -235,6 +235,53 @@ static ZFCP_DEV_ATTR(adapter, port_rescan, S_IWUSR, NULL, DEFINE_MUTEX(zfcp_sysfs_port_units_mutex); +static void zfcp_sysfs_port_set_removing(struct zfcp_port *const port) +{ + lockdep_assert_held(&zfcp_sysfs_port_units_mutex); + atomic_set(&port->units, -1); +} + +bool zfcp_sysfs_port_is_removing(const struct zfcp_port *const port) +{ + lockdep_assert_held(&zfcp_sysfs_port_units_mutex); + return atomic_read(&port->units) == -1; +} + +static bool zfcp_sysfs_port_in_use(struct zfcp_port *const port) +{ + struct zfcp_adapter *const adapter = port->adapter; + unsigned long flags; + struct scsi_device *sdev; + bool in_use = true; + + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (atomic_read(&port->units) > 0) + goto unlock_port_units_mutex; /* zfcp_unit(s) under port */ + + spin_lock_irqsave(adapter->scsi_host->host_lock, flags); + __shost_for_each_device(sdev, adapter->scsi_host) { + const struct zfcp_scsi_dev *zsdev = sdev_to_zfcp(sdev); + + if (sdev->sdev_state == SDEV_DEL || + sdev->sdev_state == SDEV_CANCEL) + continue; + if (zsdev->port != port) + continue; + /* alive scsi_device under port of interest */ + goto unlock_host_lock; + } + + /* port is about to be removed, so no more unit_add or slave_alloc */ + zfcp_sysfs_port_set_removing(port); + in_use = false; + +unlock_host_lock: + spin_unlock_irqrestore(adapter->scsi_host->host_lock, flags); +unlock_port_units_mutex: + mutex_unlock(&zfcp_sysfs_port_units_mutex); + return in_use; +} + static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -257,16 +304,11 @@ static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, else retval = 0; - mutex_lock(&zfcp_sysfs_port_units_mutex); - if (atomic_read(&port->units) > 0) { + if (zfcp_sysfs_port_in_use(port)) { retval = -EBUSY; - mutex_unlock(&zfcp_sysfs_port_units_mutex); put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ goto out; } - /* port is about to be removed, so no more unit_add */ - atomic_set(&port->units, -1); - mutex_unlock(&zfcp_sysfs_port_units_mutex); write_lock_irq(&adapter->port_list_lock); list_del(&port->list); diff --git a/drivers/s390/scsi/zfcp_unit.c b/drivers/s390/scsi/zfcp_unit.c index 1bf0a0984a09..e67bf7388cae 100644 --- a/drivers/s390/scsi/zfcp_unit.c +++ b/drivers/s390/scsi/zfcp_unit.c @@ -124,7 +124,7 @@ int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) int retval = 0; mutex_lock(&zfcp_sysfs_port_units_mutex); - if (atomic_read(&port->units) == -1) { + if (zfcp_sysfs_port_is_removing(port)) { /* port is already gone */ retval = -ENODEV; goto out; @@ -168,8 +168,14 @@ int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) write_lock_irq(&port->unit_list_lock); list_add_tail(&unit->list, &port->unit_list); write_unlock_irq(&port->unit_list_lock); + /* + * lock order: shost->scan_mutex before zfcp_sysfs_port_units_mutex + * due to zfcp_unit_scsi_scan() => zfcp_scsi_slave_alloc() + */ + mutex_unlock(&zfcp_sysfs_port_units_mutex); zfcp_unit_scsi_scan(unit); + return retval; out: mutex_unlock(&zfcp_sysfs_port_units_mutex); -- 2.17.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH 2/2] zfcp: fix to prevent port_remove with pure auto scan LUNs (only sdevs) @ 2019-05-23 13:23 ` Steffen Maier 0 siblings, 0 replies; 5+ messages in thread From: Steffen Maier @ 2019-05-23 13:23 UTC (permalink / raw) To: James E.J. Bottomley, Martin K. Petersen Cc: linux-scsi, linux-s390, Benjamin Block, Heiko Carstens, Steffen Maier, stable When the user tries to remove a zfcp port via sysfs, we only rejected it if there are zfcp unit children under the port. With purely automatically scanned LUNs there are no zfcp units but only SCSI devices. In such cases, the port_remove erroneously continued. We close the port and this implicitly closes all LUNs under the port. The SCSI devices survive with their private zfcp_scsi_dev still holding a reference to the "removed" zfcp_port (still allocated but invisible in sysfs) [zfcp_get_port_by_wwpn in zfcp_scsi_slave_alloc]. This is not a problem as long as the fc_rport stays blocked. Once (auto) port scan brings back the removed port, we unblock its fc_rport again by design. However, there is no mechanism that would recover (open) the LUNs under the port (no "ersfs_3" without zfcp_unit [zfcp_erp_strategy_followup_success]). Any pending or new I/O to such LUN leads to repeated: Done: NEEDS_RETRY Result: hostbyte=DID_IMM_RETRY driverbyte=DRIVER_OK See also v4.10 commit 6f2ce1c6af37 ("scsi: zfcp: fix rport unblock race with LUN recovery"). Even a manual LUN recovery (echo 0 > /sys/bus/scsi/devices/H:C:T:L/zfcp_failed) does not help, as the LUN links to the old "removed" port which remains to lack ZFCP_STATUS_COMMON_RUNNING [zfcp_erp_required_act]. The only workaround is to first ensure that the fc_rport is blocked (e.g. port_remove again in case it was re-discovered by (auto) port scan), then delete the SCSI devices, and finally re-discover by (auto) port scan. The port scan includes an fc_rport unblock, which in turn triggers a new scan on the scsi target to freshly get new pure auto scan LUNs. Fix this by rejecting port_remove also if there are SCSI devices (even without any zfcp_unit) under this port. Re-use mechanics from v3.7 commit d99b601b6338 ("[SCSI] zfcp: restore refcount check on port_remove"). However, we have to give up zfcp_sysfs_port_units_mutex earlier in unit_add to prevent a deadlock with scsi_host scan taking shost->scan_mutex first and then zfcp_sysfs_port_units_mutex now in our zfcp_scsi_slave_alloc(). Signed-off-by: Steffen Maier <maier@linux.ibm.com> Fixes: b62a8d9b45b9 ("[SCSI] zfcp: Use SCSI device data zfcp scsi dev instead of zfcp unit") Fixes: f8210e34887e ("[SCSI] zfcp: Allow midlayer to scan for LUNs when running in NPIV mode") Cc: <stable@vger.kernel.org> #2.6.37+ Reviewed-by: Benjamin Block <bblock@linux.ibm.com> --- drivers/s390/scsi/zfcp_ext.h | 1 + drivers/s390/scsi/zfcp_scsi.c | 9 ++++++ drivers/s390/scsi/zfcp_sysfs.c | 54 ++++++++++++++++++++++++++++++---- drivers/s390/scsi/zfcp_unit.c | 8 ++++- 4 files changed, 65 insertions(+), 7 deletions(-) diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index c6acca521ffe..31e8a7240fd7 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -167,6 +167,7 @@ extern const struct attribute_group *zfcp_port_attr_groups[]; extern struct mutex zfcp_sysfs_port_units_mutex; extern struct device_attribute *zfcp_sysfs_sdev_attrs[]; extern struct device_attribute *zfcp_sysfs_shost_attrs[]; +bool zfcp_sysfs_port_is_removing(const struct zfcp_port *const port); /* zfcp_unit.c */ extern int zfcp_unit_add(struct zfcp_port *, u64); diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c index 221d0dfb8493..e9ded2befa0d 100644 --- a/drivers/s390/scsi/zfcp_scsi.c +++ b/drivers/s390/scsi/zfcp_scsi.c @@ -129,6 +129,15 @@ static int zfcp_scsi_slave_alloc(struct scsi_device *sdev) zfcp_sdev->erp_action.port = port; + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (zfcp_sysfs_port_is_removing(port)) { + /* port is already gone */ + mutex_unlock(&zfcp_sysfs_port_units_mutex); + put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ + return -ENXIO; + } + mutex_unlock(&zfcp_sysfs_port_units_mutex); + unit = zfcp_unit_find(port, zfcp_scsi_dev_lun(sdev)); if (unit) put_device(&unit->dev); diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index 2d78732b270b..af197e2b3e69 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -235,6 +235,53 @@ static ZFCP_DEV_ATTR(adapter, port_rescan, S_IWUSR, NULL, DEFINE_MUTEX(zfcp_sysfs_port_units_mutex); +static void zfcp_sysfs_port_set_removing(struct zfcp_port *const port) +{ + lockdep_assert_held(&zfcp_sysfs_port_units_mutex); + atomic_set(&port->units, -1); +} + +bool zfcp_sysfs_port_is_removing(const struct zfcp_port *const port) +{ + lockdep_assert_held(&zfcp_sysfs_port_units_mutex); + return atomic_read(&port->units) == -1; +} + +static bool zfcp_sysfs_port_in_use(struct zfcp_port *const port) +{ + struct zfcp_adapter *const adapter = port->adapter; + unsigned long flags; + struct scsi_device *sdev; + bool in_use = true; + + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (atomic_read(&port->units) > 0) + goto unlock_port_units_mutex; /* zfcp_unit(s) under port */ + + spin_lock_irqsave(adapter->scsi_host->host_lock, flags); + __shost_for_each_device(sdev, adapter->scsi_host) { + const struct zfcp_scsi_dev *zsdev = sdev_to_zfcp(sdev); + + if (sdev->sdev_state == SDEV_DEL || + sdev->sdev_state == SDEV_CANCEL) + continue; + if (zsdev->port != port) + continue; + /* alive scsi_device under port of interest */ + goto unlock_host_lock; + } + + /* port is about to be removed, so no more unit_add or slave_alloc */ + zfcp_sysfs_port_set_removing(port); + in_use = false; + +unlock_host_lock: + spin_unlock_irqrestore(adapter->scsi_host->host_lock, flags); +unlock_port_units_mutex: + mutex_unlock(&zfcp_sysfs_port_units_mutex); + return in_use; +} + static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -257,16 +304,11 @@ static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, else retval = 0; - mutex_lock(&zfcp_sysfs_port_units_mutex); - if (atomic_read(&port->units) > 0) { + if (zfcp_sysfs_port_in_use(port)) { retval = -EBUSY; - mutex_unlock(&zfcp_sysfs_port_units_mutex); put_device(&port->dev); /* undo zfcp_get_port_by_wwpn() */ goto out; } - /* port is about to be removed, so no more unit_add */ - atomic_set(&port->units, -1); - mutex_unlock(&zfcp_sysfs_port_units_mutex); write_lock_irq(&adapter->port_list_lock); list_del(&port->list); diff --git a/drivers/s390/scsi/zfcp_unit.c b/drivers/s390/scsi/zfcp_unit.c index 1bf0a0984a09..e67bf7388cae 100644 --- a/drivers/s390/scsi/zfcp_unit.c +++ b/drivers/s390/scsi/zfcp_unit.c @@ -124,7 +124,7 @@ int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) int retval = 0; mutex_lock(&zfcp_sysfs_port_units_mutex); - if (atomic_read(&port->units) == -1) { + if (zfcp_sysfs_port_is_removing(port)) { /* port is already gone */ retval = -ENODEV; goto out; @@ -168,8 +168,14 @@ int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) write_lock_irq(&port->unit_list_lock); list_add_tail(&unit->list, &port->unit_list); write_unlock_irq(&port->unit_list_lock); + /* + * lock order: shost->scan_mutex before zfcp_sysfs_port_units_mutex + * due to zfcp_unit_scsi_scan() => zfcp_scsi_slave_alloc() + */ + mutex_unlock(&zfcp_sysfs_port_units_mutex); zfcp_unit_scsi_scan(unit); + return retval; out: mutex_unlock(&zfcp_sysfs_port_units_mutex); -- 2.17.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 0/2] zfcp fixes for v5.2-rcX [not found] <1558617826-30129-1-git-send-email-maier@linux.ibm.com> 2019-05-23 13:23 ` Steffen Maier 2019-05-23 13:23 ` Steffen Maier @ 2019-05-30 1:53 ` Martin K. Petersen 2 siblings, 0 replies; 5+ messages in thread From: Martin K. Petersen @ 2019-05-30 1:53 UTC (permalink / raw) To: linux-s390, linux-scsi Steffen, > here are 2 zfcp bugfixes for v5.2-rcX. Applied to 5.2/scsi-fixes. Thank you! -- Martin K. Petersen Oracle Linux Engineering ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2019-05-30 1:53 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <1558617826-30129-1-git-send-email-maier@linux.ibm.com>
2019-05-23 13:23 ` [PATCH 1/2] zfcp: fix missing zfcp_port reference put on -EBUSY from port_remove Steffen Maier
2019-05-23 13:23 ` Steffen Maier
2019-05-23 13:23 ` [PATCH 2/2] zfcp: fix to prevent port_remove with pure auto scan LUNs (only sdevs) Steffen Maier
2019-05-23 13:23 ` Steffen Maier
2019-05-30 1:53 ` [PATCH 0/2] zfcp fixes for v5.2-rcX Martin K. Petersen
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.