Commit 92635801 authored by Wei Fang's avatar Wei Fang Committed by Ben Hutchings

scsi: fix race between simultaneous decrements of ->host_failed

commit 72d8c36e upstream.

sas_ata_strategy_handler() adds the works of the ata error handler to
system_unbound_wq. This workqueue asynchronously runs work items, so the
ata error handler will be performed concurrently on different CPUs. In
this case, ->host_failed will be decreased simultaneously in
scsi_eh_finish_cmd() on different CPUs, and become abnormal.

It will lead to permanently inequality between ->host_failed and
->host_busy, and scsi error handler thread won't start running. IO
errors after that won't be handled.

Since all scmds must have been handled in the strategy handler, just
remove the decrement in scsi_eh_finish_cmd() and zero ->host_busy after
the strategy handler to fix this race.

Fixes: 50824d6c ("[SCSI] libsas: async ata-eh")
Signed-off-by: default avatarWei Fang <fangwei1@huawei.com>
Reviewed-by: default avatarJames Bottomley <jejb@linux.vnet.ibm.com>
Signed-off-by: default avatarMartin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: default avatarBen Hutchings <ben@decadent.org.uk>
parent 6947850e
...@@ -263,19 +263,23 @@ scmd->allowed. ...@@ -263,19 +263,23 @@ scmd->allowed.
3. scmd recovered 3. scmd recovered
ACTION: scsi_eh_finish_cmd() is invoked to EH-finish scmd ACTION: scsi_eh_finish_cmd() is invoked to EH-finish scmd
- shost->host_failed--
- clear scmd->eh_eflags - clear scmd->eh_eflags
- scsi_setup_cmd_retry() - scsi_setup_cmd_retry()
- move from local eh_work_q to local eh_done_q - move from local eh_work_q to local eh_done_q
LOCKING: none LOCKING: none
CONCURRENCY: at most one thread per separate eh_work_q to
keep queue manipulation lockless
4. EH completes 4. EH completes
ACTION: scsi_eh_flush_done_q() retries scmds or notifies upper ACTION: scsi_eh_flush_done_q() retries scmds or notifies upper
layer of failure. layer of failure. May be called concurrently but must have
a no more than one thread per separate eh_work_q to
manipulate the queue locklessly
- scmd is removed from eh_done_q and scmd->eh_entry is cleared - scmd is removed from eh_done_q and scmd->eh_entry is cleared
- if retry is necessary, scmd is requeued using - if retry is necessary, scmd is requeued using
scsi_queue_insert() scsi_queue_insert()
- otherwise, scsi_finish_command() is invoked for scmd - otherwise, scsi_finish_command() is invoked for scmd
- zero shost->host_failed
LOCKING: queue or finish function performs appropriate locking LOCKING: queue or finish function performs appropriate locking
......
...@@ -605,7 +605,7 @@ void ata_scsi_error(struct Scsi_Host *host) ...@@ -605,7 +605,7 @@ void ata_scsi_error(struct Scsi_Host *host)
ata_scsi_port_error_handler(host, ap); ata_scsi_port_error_handler(host, ap);
/* finish or retry handled scmd's and clean up */ /* finish or retry handled scmd's and clean up */
WARN_ON(host->host_failed || !list_empty(&eh_work_q)); WARN_ON(!list_empty(&eh_work_q));
DPRINTK("EXIT\n"); DPRINTK("EXIT\n");
} }
......
...@@ -1115,7 +1115,6 @@ static int scsi_eh_action(struct scsi_cmnd *scmd, int rtn) ...@@ -1115,7 +1115,6 @@ static int scsi_eh_action(struct scsi_cmnd *scmd, int rtn)
*/ */
void scsi_eh_finish_cmd(struct scsi_cmnd *scmd, struct list_head *done_q) void scsi_eh_finish_cmd(struct scsi_cmnd *scmd, struct list_head *done_q)
{ {
scmd->device->host->host_failed--;
scmd->eh_eflags = 0; scmd->eh_eflags = 0;
list_move_tail(&scmd->eh_entry, done_q); list_move_tail(&scmd->eh_entry, done_q);
} }
...@@ -2198,6 +2197,9 @@ int scsi_error_handler(void *data) ...@@ -2198,6 +2197,9 @@ int scsi_error_handler(void *data)
else else
scsi_unjam_host(shost); scsi_unjam_host(shost);
/* All scmds have been handled */
shost->host_failed = 0;
/* /*
* Note - if the above fails completely, the action is to take * Note - if the above fails completely, the action is to take
* individual devices offline and flush the queue of any * individual devices offline and flush the queue of any
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment