The vmw_pvscsi driver reports most successful aborts as FAILED to the scsi
error handler. This is do to a misunderstanding of how completion_done() works
and its interaction with a successful wait using wait_for_completion_timeout().
The vmw_pvscsi driver is expecting completion_done() to always return true if
complete() has been called on the completion structure. But completion_done()
returns true after complete() has been called only if no function like
wait_for_completion_timeout() has seen the completion and cleared it as part of
successfully waiting for the completion.
Instead of using completion_done(), vmw_pvscsi should just use the return
value from wait_for_completion_timeout() to know if the wait timed out or not.
Signed-off-by: David Jeffery <djeffery@xxxxxxxxxx>
---
vmw_pvscsi.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
--- a/drivers/scsi/vmw_pvscsi.c
+++ b/drivers/scsi/vmw_pvscsi.c
@@ -793,6 +793,7 @@ static int pvscsi_abort(struct scsi_cmnd
unsigned long flags;
int result = SUCCESS;
DECLARE_COMPLETION_ONSTACK(abort_cmp);
+ int done;
scmd_printk(KERN_DEBUG, cmd, "task abort on host %u, %p\n",
adapter->host->host_no, cmd);
@@ -824,10 +825,10 @@ static int pvscsi_abort(struct scsi_cmnd
pvscsi_abort_cmd(adapter, ctx);
spin_unlock_irqrestore(&adapter->hw_lock, flags);
/* Wait for 2 secs for the completion. */
- wait_for_completion_timeout(&abort_cmp, msecs_to_jiffies(2000));
+ done = wait_for_completion_timeout(&abort_cmp, msecs_to_jiffies(2000));
spin_lock_irqsave(&adapter->hw_lock, flags);
- if (!completion_done(&abort_cmp)) {
+ if (!done) {
/*
* Failed to abort the command, unmark the fact that it
* was requested to be aborted.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
