A SCSI LLD may start cleaning up host resources as soon as
scsi_remove_host() returns. These host resources may be needed by
the LLD in an implementation of one of the eh_* functions. So if
one of the eh_* functions is in progress when scsi_remove_host()
is invoked, wait until the eh_* function has finished. Also, do
not invoke any of the eh_* functions after scsi_remove_host() has
started.

Signed-off-by: Bart Van Assche <bvanass...@acm.org>
Cc: Hannes Reinecke <h...@suse.de>
Cc: Mike Christie <micha...@cs.wisc.edu>
Cc: Tejun Heo <t...@kernel.org>
---
 drivers/scsi/hosts.c      |    2 +-
 drivers/scsi/scsi_error.c |  114 ++++++++++++++++++++++++++++++++++-----------
 include/scsi/scsi_host.h  |    1 +
 3 files changed, 89 insertions(+), 28 deletions(-)

diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 7bd944e..477b8d6 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -155,7 +155,7 @@ static bool __scsi_remove_host_done(struct Scsi_Host *shost)
 {
        lockdep_assert_held(shost->host_lock);
 
-       return list_empty(&shost->__devices);
+       return list_empty(&shost->__devices) && !shost->eh_active;
 }
 
 /* Test whether scsi_remove_host() may finish, and if so, wake it up. */
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index c1b05a8..cab2ac3 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -536,8 +536,52 @@ static void scsi_eh_done(struct scsi_cmnd *scmd)
 }
 
 /**
+ * scsi_begin_eh - start host-related error handling
+ *
+ * Must be called before invoking any of the scsi_host_template.eh_* functions
+ * to avoid that scsi_remove_host() returns while one of these callback
+ * functions is in progress.
+ *
+ * Returns true if invoking the eh_* function is allowed and false if not.
+ * If this function returns true then scsi_end_eh() must be called eventually.
+ *
+ * Note: scsi_send_eh_cmnd() calls do not have to be protected by a
+ * scsi_begin_eh() / scsi_end_eh() pair since these operate on an unfinished
+ * block layer request. Since scsi_remove_host() waits until all SCSI devices
+ * have been removed and since blk_cleanup_queue() is invoked during SCSI
+ * device removal scsi_remove_host() won't finish while a scsi_send_eh_cmnd()
+ * call is in progress.
+ */
+static bool scsi_begin_eh(struct Scsi_Host *host)
+{
+       bool res;
+
+       spin_lock_irq(host->host_lock);
+       res = scsi_host_scan_allowed(host);
+       if (res) {
+               WARN_ON_ONCE(host->eh_active < 0 || host->eh_active > 1);
+               host->eh_active++;
+       }
+       spin_unlock_irq(host->host_lock);
+
+       return res;
+}
+
+/**
+ * scsi_end_eh - finish host-related error handling
+ */
+static void scsi_end_eh(struct Scsi_Host *host)
+{
+       spin_lock_irq(host->host_lock);
+       host->eh_active--;
+       WARN_ON_ONCE(host->eh_active < 0 || host->eh_active > 1);
+       __scsi_check_remove_host_done(host);
+       spin_unlock_irq(host->host_lock);
+}
+
+/**
  * scsi_try_host_reset - ask host adapter to reset itself
- * @scmd:      SCSI cmd to send hsot reset.
+ * @scmd:      SCSI cmd to send host reset.
  */
 static int scsi_try_host_reset(struct scsi_cmnd *scmd)
 {
@@ -552,14 +596,17 @@ static int scsi_try_host_reset(struct scsi_cmnd *scmd)
        if (!hostt->eh_host_reset_handler)
                return FAILED;
 
-       rtn = hostt->eh_host_reset_handler(scmd);
-
-       if (rtn == SUCCESS) {
-               if (!hostt->skip_settle_delay)
-                       ssleep(HOST_RESET_SETTLE_TIME);
-               spin_lock_irqsave(host->host_lock, flags);
-               scsi_report_bus_reset(host, scmd_channel(scmd));
-               spin_unlock_irqrestore(host->host_lock, flags);
+       rtn = FAST_IO_FAIL;
+       if (scsi_begin_eh(host)) {
+               rtn = hostt->eh_host_reset_handler(scmd);
+               if (rtn == SUCCESS) {
+                       if (!hostt->skip_settle_delay)
+                               ssleep(HOST_RESET_SETTLE_TIME);
+                       spin_lock_irqsave(host->host_lock, flags);
+                       scsi_report_bus_reset(host, scmd_channel(scmd));
+                       spin_unlock_irqrestore(host->host_lock, flags);
+               }
+               scsi_end_eh(host);
        }
 
        return rtn;
@@ -582,14 +629,17 @@ static int scsi_try_bus_reset(struct scsi_cmnd *scmd)
        if (!hostt->eh_bus_reset_handler)
                return FAILED;
 
-       rtn = hostt->eh_bus_reset_handler(scmd);
-
-       if (rtn == SUCCESS) {
-               if (!hostt->skip_settle_delay)
-                       ssleep(BUS_RESET_SETTLE_TIME);
-               spin_lock_irqsave(host->host_lock, flags);
-               scsi_report_bus_reset(host, scmd_channel(scmd));
-               spin_unlock_irqrestore(host->host_lock, flags);
+       rtn = FAST_IO_FAIL;
+       if (scsi_begin_eh(host)) {
+               rtn = hostt->eh_bus_reset_handler(scmd);
+               if (rtn == SUCCESS) {
+                       if (!hostt->skip_settle_delay)
+                               ssleep(BUS_RESET_SETTLE_TIME);
+                       spin_lock_irqsave(host->host_lock, flags);
+                       scsi_report_bus_reset(host, scmd_channel(scmd));
+                       spin_unlock_irqrestore(host->host_lock, flags);
+               }
+               scsi_end_eh(host);
        }
 
        return rtn;
@@ -621,12 +671,17 @@ static int scsi_try_target_reset(struct scsi_cmnd *scmd)
        if (!hostt->eh_target_reset_handler)
                return FAILED;
 
-       rtn = hostt->eh_target_reset_handler(scmd);
-       if (rtn == SUCCESS) {
-               spin_lock_irqsave(host->host_lock, flags);
-               __starget_for_each_device(scsi_target(scmd->device), NULL,
-                                         __scsi_report_device_reset);
-               spin_unlock_irqrestore(host->host_lock, flags);
+       rtn = FAST_IO_FAIL;
+       if (scsi_begin_eh(host)) {
+               rtn = hostt->eh_target_reset_handler(scmd);
+               if (rtn == SUCCESS) {
+                       spin_lock_irqsave(host->host_lock, flags);
+                       __starget_for_each_device(scsi_target(scmd->device),
+                                                 NULL,
+                                                 __scsi_report_device_reset);
+                       spin_unlock_irqrestore(host->host_lock, flags);
+               }
+               scsi_end_eh(host);
        }
 
        return rtn;
@@ -645,14 +700,19 @@ static int scsi_try_target_reset(struct scsi_cmnd *scmd)
 static int scsi_try_bus_device_reset(struct scsi_cmnd *scmd)
 {
        int rtn;
-       struct scsi_host_template *hostt = scmd->device->host->hostt;
+       struct Scsi_Host *host = scmd->device->host;
+       struct scsi_host_template *hostt = host->hostt;
 
        if (!hostt->eh_device_reset_handler)
                return FAILED;
 
-       rtn = hostt->eh_device_reset_handler(scmd);
-       if (rtn == SUCCESS)
-               __scsi_report_device_reset(scmd->device, NULL);
+       rtn = FAST_IO_FAIL;
+       if (scsi_begin_eh(host)) {
+               rtn = hostt->eh_device_reset_handler(scmd);
+               if (rtn == SUCCESS)
+                       __scsi_report_device_reset(scmd->device, NULL);
+               scsi_end_eh(host);
+       }
        return rtn;
 }
 
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 1b7fd89..5e2fcd2 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -576,6 +576,7 @@ struct Scsi_Host {
        struct task_struct    * ehandler;  /* Error recovery thread. */
        struct completion     * eh_action; /* Wait for specific actions on the
                                              host. */
+       int                     eh_active;
        wait_queue_head_t       host_wait;
        wait_queue_head_t       remove_host;
        struct scsi_host_template *hostt;
-- 
1.7.10.4

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to