All of lore.kernel.org
 help / color / mirror / Atom feed
From: kys@exchange.microsoft.com
To: linux-kernel@vger.kernel.org, devel@linuxdriverproject.org,
	ohering@suse.com, linux-scsi@vger.kernel.org, apw@canonical.com,
	vkuznets@redhat.com, jasowang@redhat.com,
	martin.petersen@oracle.com, hare@suse.de,
	James.Bottomley@HansenPartnership.com
Cc: Long Li <longli@microsoft.com>,
	"K. Y. Srinivasan" <kys@microsoft.com>, <stable@vger.kernel.org>
Subject: [PATCH 6/6] storvsc: properly set residual data length on errors
Date: Wed, 14 Dec 2016 18:46:03 -0800	[thread overview]
Message-ID: <1481769963-2069-6-git-send-email-kys@exchange.microsoft.com> (raw)
In-Reply-To: <1481769963-2069-1-git-send-email-kys@exchange.microsoft.com>

From: Long Li <longli@microsoft.com>

On I/O errors, the Windows driver doesn't set data_transfer_length
on error conditions other than SRB_STATUS_DATA_OVERRUN.
In these cases we need to set data_transfer_length to 0,
indicating there is no data transferred. On SRB_STATUS_DATA_OVERRUN,
data_transfer_length is set by the Windows driver to the actual data transferred.

Reported-by: Shiva Krishna <Shiva.Krishna@nimblestorage.com>
Signed-off-by: Long Li <longli@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Cc: <stable@vger.kernel.org>
---
 drivers/scsi/storvsc_drv.c |   16 +++++++++++++---
 1 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c
index 3c92dc2..888e16e 100644
--- a/drivers/scsi/storvsc_drv.c
+++ b/drivers/scsi/storvsc_drv.c
@@ -377,6 +377,7 @@ enum storvsc_request_type {
 #define SRB_STATUS_SUCCESS	0x01
 #define SRB_STATUS_ABORTED	0x02
 #define SRB_STATUS_ERROR	0x04
+#define SRB_STATUS_DATA_OVERRUN	0x12
 
 #define SRB_STATUS(status) \
 	(status & ~(SRB_STATUS_AUTOSENSE_VALID | SRB_STATUS_QUEUE_FROZEN))
@@ -996,6 +997,7 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 	struct scsi_cmnd *scmnd = cmd_request->cmd;
 	struct scsi_sense_hdr sense_hdr;
 	struct vmscsi_request *vm_srb;
+	u32 data_transfer_length;
 	struct Scsi_Host *host;
 	u32 payload_sz = cmd_request->payload_sz;
 	void *payload = cmd_request->payload;
@@ -1003,6 +1005,7 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 	host = stor_dev->host;
 
 	vm_srb = &cmd_request->vstor_packet.vm_srb;
+	data_transfer_length = vm_srb->data_transfer_length;
 
 	scmnd->result = vm_srb->scsi_status;
 
@@ -1016,13 +1019,20 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 					     &sense_hdr);
 	}
 
-	if (vm_srb->srb_status != SRB_STATUS_SUCCESS)
+	if (vm_srb->srb_status != SRB_STATUS_SUCCESS) {
 		storvsc_handle_error(vm_srb, scmnd, host, sense_hdr.asc,
 					 sense_hdr.ascq);
+		/*
+		 * The Windows driver set data_transfer_length on
+		 * SRB_STATUS_DATA_OVERRUN. On other errors, this value
+		 * is untouched.  In these cases we set it to 0.
+		 */
+		if (vm_srb->srb_status != SRB_STATUS_DATA_OVERRUN)
+			data_transfer_length = 0;
+	}
 
 	scsi_set_resid(scmnd,
-		cmd_request->payload->range.len -
-		vm_srb->data_transfer_length);
+		cmd_request->payload->range.len - data_transfer_length);
 
 	scmnd->scsi_done(scmnd);
 
-- 
1.7.4.1

WARNING: multiple messages have this Message-ID (diff)
From: kys@exchange.microsoft.com
To: linux-kernel@vger.kernel.org, devel@linuxdriverproject.org,
	ohering@suse.com, linux-scsi@vger.kernel.org, apw@canonical.com,
	vkuznets@redhat.com, jasowang@redhat.com,
	martin.petersen@oracle.com, hare@suse.de,
	James.Bottomley@HansenPartnership.com
Cc: stable@vger.kernel.org
Subject: [PATCH 6/6] storvsc: properly set residual data length on errors
Date: Wed, 14 Dec 2016 18:46:03 -0800	[thread overview]
Message-ID: <1481769963-2069-6-git-send-email-kys@exchange.microsoft.com> (raw)
In-Reply-To: <1481769963-2069-1-git-send-email-kys@exchange.microsoft.com>

From: Long Li <longli@microsoft.com>

On I/O errors, the Windows driver doesn't set data_transfer_length
on error conditions other than SRB_STATUS_DATA_OVERRUN.
In these cases we need to set data_transfer_length to 0,
indicating there is no data transferred. On SRB_STATUS_DATA_OVERRUN,
data_transfer_length is set by the Windows driver to the actual data transferred.

Reported-by: Shiva Krishna <Shiva.Krishna@nimblestorage.com>
Signed-off-by: Long Li <longli@microsoft.com>
Reviewed-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Cc: <stable@vger.kernel.org>
---
 drivers/scsi/storvsc_drv.c |   16 +++++++++++++---
 1 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c
index 3c92dc2..888e16e 100644
--- a/drivers/scsi/storvsc_drv.c
+++ b/drivers/scsi/storvsc_drv.c
@@ -377,6 +377,7 @@ enum storvsc_request_type {
 #define SRB_STATUS_SUCCESS	0x01
 #define SRB_STATUS_ABORTED	0x02
 #define SRB_STATUS_ERROR	0x04
+#define SRB_STATUS_DATA_OVERRUN	0x12
 
 #define SRB_STATUS(status) \
 	(status & ~(SRB_STATUS_AUTOSENSE_VALID | SRB_STATUS_QUEUE_FROZEN))
@@ -996,6 +997,7 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 	struct scsi_cmnd *scmnd = cmd_request->cmd;
 	struct scsi_sense_hdr sense_hdr;
 	struct vmscsi_request *vm_srb;
+	u32 data_transfer_length;
 	struct Scsi_Host *host;
 	u32 payload_sz = cmd_request->payload_sz;
 	void *payload = cmd_request->payload;
@@ -1003,6 +1005,7 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 	host = stor_dev->host;
 
 	vm_srb = &cmd_request->vstor_packet.vm_srb;
+	data_transfer_length = vm_srb->data_transfer_length;
 
 	scmnd->result = vm_srb->scsi_status;
 
@@ -1016,13 +1019,20 @@ static void storvsc_command_completion(struct storvsc_cmd_request *cmd_request,
 					     &sense_hdr);
 	}
 
-	if (vm_srb->srb_status != SRB_STATUS_SUCCESS)
+	if (vm_srb->srb_status != SRB_STATUS_SUCCESS) {
 		storvsc_handle_error(vm_srb, scmnd, host, sense_hdr.asc,
 					 sense_hdr.ascq);
+		/*
+		 * The Windows driver set data_transfer_length on
+		 * SRB_STATUS_DATA_OVERRUN. On other errors, this value
+		 * is untouched.  In these cases we set it to 0.
+		 */
+		if (vm_srb->srb_status != SRB_STATUS_DATA_OVERRUN)
+			data_transfer_length = 0;
+	}
 
 	scsi_set_resid(scmnd,
-		cmd_request->payload->range.len -
-		vm_srb->data_transfer_length);
+		cmd_request->payload->range.len - data_transfer_length);
 
 	scmnd->scsi_done(scmnd);
 
-- 
1.7.4.1

  parent reply	other threads:[~2016-12-15  0:50 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-15  2:45 [PATCH 0/6] storvsc: Miscellaneous fixes and enhancements kys
2016-12-15  2:45 ` kys
2016-12-15  2:45 ` [PATCH 1/6] storvsc: Enable tracking of queue depth kys
2016-12-15  2:45   ` kys
2016-12-15  2:45   ` [PATCH 2/6] storvsc: Remove the restriction on max segment size kys
2016-12-15  2:45     ` kys
2016-12-15  2:46   ` [PATCH 3/6] storvsc: Enable multi-queue support kys
2016-12-15  2:46   ` [PATCH 4/6] storvsc: use tagged SRB requests if supported by the device kys
2016-12-15  2:46     ` kys
2016-12-15  2:46   ` [PATCH 5/6] storvsc: properly handle SRB_ERROR when sense message is present kys
2016-12-15  2:46     ` kys
2016-12-15  2:46   ` kys [this message]
2016-12-15  2:46     ` [PATCH 6/6] storvsc: properly set residual data length on errors kys
2016-12-20 22:43 ` [PATCH 0/6] storvsc: Miscellaneous fixes and enhancements Martin K. Petersen
2016-12-20 22:43   ` Martin K. Petersen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1481769963-2069-6-git-send-email-kys@exchange.microsoft.com \
    --to=kys@exchange.microsoft.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=apw@canonical.com \
    --cc=devel@linuxdriverproject.org \
    --cc=hare@suse.de \
    --cc=jasowang@redhat.com \
    --cc=kys@microsoft.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=longli@microsoft.com \
    --cc=martin.petersen@oracle.com \
    --cc=ohering@suse.com \
    --cc=stable@vger.kernel.org \
    --cc=vkuznets@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.