[SCSI] mptfusion: Fix for device offline while doing aggressive HBA reset

[Resend patch as per Bernd Schubert comment ]

Issue:

Device goes offline while doing aggressive HBA reset
along with IO using some utility.

Root cause:

FW goes into bad state due to aggressive reset. Softreset does not
help to recover FW. And also aggressive reset open up the window for
Error handling thread to kicked off at the same time HBA will be in
constant RESET loop as part of aggressive reset test case can lead
Device to goes offline.

Changes:

1. Added extra check as below inside eh_timed_out call back as below.
   if(ioc->ioc_reset_in_progress) Rc = EH_TIMER_RESET

2. Removed " DOORBELL_ACTIVE" check for SAS controller from task
   management context.  Since SAS controller uses high priority queue
   for task management. This check is not required for SAS controller.

3. Moved SoftReset call to HardReset from Task Mgmt context.

Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>

1 files changed, 26 insertions, 3 deletions

diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 517621fa8bca..e9c6a6047a00 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -6474,8 +6474,19 @@ mpt_config(MPT_ADAPTER *ioc, CONFIGPARMS *pCfg)
 			pReq->Action, ioc->mptbase_cmds.status, timeleft));
 		if (ioc->mptbase_cmds.status & MPT_MGMT_STATUS_DID_IOCRESET)
 			goto out;
-		if (!timeleft)
+		if (!timeleft) {
+			spin_lock_irqsave(&ioc->taskmgmt_lock, flags);
+			if (ioc->ioc_reset_in_progress) {
+				spin_unlock_irqrestore(&ioc->taskmgmt_lock,
+					flags);
+				printk(MYIOC_s_INFO_FMT "%s: host reset in"
+					" progress mpt_config timed out.!!\n",
+					__func__, ioc->name);
+				return -EFAULT;
+			}
+			spin_unlock_irqrestore(&ioc->taskmgmt_lock, flags);
 			issue_hard_reset = 1;
+		}
 		goto out;
 	}
 
@@ -7189,7 +7200,18 @@ mpt_HardResetHandler(MPT_ADAPTER *ioc, int sleepFlag)
 	spin_lock_irqsave(&ioc->taskmgmt_lock, flags);
 	if (ioc->ioc_reset_in_progress) {
 		spin_unlock_irqrestore(&ioc->taskmgmt_lock, flags);
-		return 0;
+		ioc->wait_on_reset_completion = 1;
+		do {
+			ssleep(1);
+		} while (ioc->ioc_reset_in_progress == 1);
+		ioc->wait_on_reset_completion = 0;
+		return ioc->reset_status;
+	}
+	if (ioc->wait_on_reset_completion) {
+		spin_unlock_irqrestore(&ioc->taskmgmt_lock, flags);
+		rc = 0;
+		time_count = jiffies;
+		goto exit;
 	}
 	ioc->ioc_reset_in_progress = 1;
 	if (ioc->alt_ioc)
@@ -7226,6 +7248,7 @@ mpt_HardResetHandler(MPT_ADAPTER *ioc, int sleepFlag)
 	ioc->ioc_reset_in_progress = 0;
 	ioc->taskmgmt_quiesce_io = 0;
 	ioc->taskmgmt_in_progress = 0;
+	ioc->reset_status = rc;
 	if (ioc->alt_ioc) {
 		ioc->alt_ioc->ioc_reset_in_progress = 0;
 		ioc->alt_ioc->taskmgmt_quiesce_io = 0;
@@ -7241,7 +7264,7 @@ mpt_HardResetHandler(MPT_ADAPTER *ioc, int sleepFlag)
 					ioc->alt_ioc, MPT_IOC_POST_RESET);
 		}
 	}
-
+exit:
 	dtmprintk(ioc,
 	    printk(MYIOC_s_DEBUG_FMT
 		"HardResetHandler: completed (%d seconds): %s\n", ioc->name,