[SCSI] mpt2sas: Do not retry a timed out direct IO for warpdrive

When an I/O request to a WarpDrive is timed out by SML and if the
I/O request to the WarpDrive is sent as direct I/O then the aborted
direct I/O will be retried as normal Volume I/O and which results
in failure of Target Reset and results in host reset.

The fix is to not retry a failed IO to volume when the original
IO was sent as direct IO with an ioc status
MPI2_IOCSTATUS_SCSI_TASK_TERMINATED.

Signed-off-by: Nagalakshmi Nandigama <nagalakshmi.nandigama@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
This commit is contained in:
nagalakshmi.nandigama@lsi.com 2011-12-01 07:53:02 +05:30 committed by James Bottomley
parent 30c43282f3
commit 4da7af9494

View File

@ -4436,11 +4436,14 @@ _scsih_io_done(struct MPT2SAS_ADAPTER *ioc, u16 smid, u8 msix_index, u32 reply)
scmd->result = DID_NO_CONNECT << 16;
goto out;
}
ioc_status = le16_to_cpu(mpi_reply->IOCStatus);
/*
* WARPDRIVE: If direct_io is set then it is directIO,
* the failed direct I/O should be redirected to volume
*/
if (_scsih_scsi_direct_io_get(ioc, smid)) {
if (_scsih_scsi_direct_io_get(ioc, smid) &&
((ioc_status & MPI2_IOCSTATUS_MASK)
!= MPI2_IOCSTATUS_SCSI_TASK_TERMINATED)) {
spin_lock_irqsave(&ioc->scsi_lookup_lock, flags);
ioc->scsi_lookup[smid - 1].scmd = scmd;
spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
@ -4474,7 +4477,6 @@ _scsih_io_done(struct MPT2SAS_ADAPTER *ioc, u16 smid, u8 msix_index, u32 reply)
xfer_cnt = le32_to_cpu(mpi_reply->TransferCount);
scsi_set_resid(scmd, scsi_bufflen(scmd) - xfer_cnt);
ioc_status = le16_to_cpu(mpi_reply->IOCStatus);
if (ioc_status & MPI2_IOCSTATUS_FLAG_LOG_INFO_AVAILABLE)
log_info = le32_to_cpu(mpi_reply->IOCLogInfo);
else