Skip to contentSkip to conversation

storvsc: workaround for virtual DVD SCSI version

PATCH Applied Stephen Hemminger 5 participants 5 messages Mar 7, 2017 – Mar 8, 2017

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 storvsc: workaround for virtual DVD SCSI version

Hyper-V host emulation of SCSI for virtual DVD device reports SCSI version 0 (UNKNOWN) but is still capable of supporting REPORTLUN.

Without this patch, a GEN2 Linux guest on Hyper-V will not boot 4.11 successfully with virtual DVD ROM device. What happens is that the SCSI scan process falls back to doing sequential probing by INQUIRY. But the storvsc driver has a previous workaround that masks/blocks all errors reports from INQUIRY (or MODE_SENSE) commands. This workaround causes the scan to then populate a full set of bogus LUN's on the target and then sends kernel spinning off into a death spiral doing block reads on the non-existent LUNs.

By setting the correct blacklist flags, the target with the DVD device is scanned with REPORTLUN and that works correctly.

Patch needs to go in current 4.11, it is safe but not necessary in older kernels.

---
 drivers/scsi/storvsc_drv.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

PS: The error handling does need to be fixed (have patches pending) but that is interrelated with hotplug and can wait.

drivers/scsi/storvsc_drv.c +17 -10
index 638e5f427c90..19973e874830 100644
@@ -400,8 +400,6 @@ MODULE_PARM_DESC(storvsc_vcpus_per_sub_channel, "Ratio of VCPUs to subchannels")
  */
 static int storvsc_timeout = 180;
 
-static int msft_blist_flags = BLIST_TRY_VPD_PAGES;
-
 #if IS_ENABLED(CONFIG_SCSI_FC_ATTRS)
 static struct scsi_transport_template *fc_transport_template;
 #endif
@@ -1383,6 +1381,22 @@ static int storvsc_do_io(struct hv_device *device,
 	return ret;
 }
 
+static int storvsc_device_alloc(struct scsi_device *sdevice)
+{
+	/*
+	 * Set blist flag to permit the reading of the VPD pages even when
+	 * the target may claim SPC-2 compliance. MSFT targets currently
+	 * claim SPC-2 compliance while they implement post SPC-2 features.
+	 * With this flag we can correctly handle WRITE_SAME_16 issues.
+	 *
+	 * Hypervisor reports SCSI_UNKNOWN type for DVD ROM device but
+	 * still supports REPORT LUN.
+	 */
+	sdevice->sdev_bflags = BLIST_REPORTLUN2 | BLIST_TRY_VPD_PAGES;
+
+	return 0;
+}
+
 static int storvsc_device_configure(struct scsi_device *sdevice)
 {
 
@@ -1396,14 +1410,6 @@ static int storvsc_device_configure(struct scsi_device *sdevice)
 	sdevice->no_write_same = 1;
 
 	/*
-	 * Add blist flags to permit the reading of the VPD pages even when
-	 * the target may claim SPC-2 compliance. MSFT targets currently
-	 * claim SPC-2 compliance while they implement post SPC-2 features.
-	 * With this patch we can correctly handle WRITE_SAME_16 issues.
-	 */
-	sdevice->sdev_bflags |= msft_blist_flags;
-
-	/*
 	 * If the host is WIN8 or WIN8 R2, claim conformance to SPC-3
 	 * if the device is a MSFT virtual device.  If the host is
 	 * WIN10 or newer, allow write_same.
@@ -1661,6 +1667,7 @@ static struct scsi_host_template scsi_driver = {
 	.eh_host_reset_handler =	storvsc_host_reset_handler,
 	.proc_name =		"storvsc_host",
 	.eh_timed_out =		storvsc_eh_timed_out,
+	.slave_alloc =		storvsc_device_alloc,
 	.slave_configure =	storvsc_device_configure,
 	.cmd_per_lun =		255,
 	.this_id =		-1,
Signature · 1 line
2.11.0
Reviewed-by on v1 1/1
#1 lore raw
Quoted 25 lines

-----Original Message-----
From: Stephen Hemminger [mailto:[email protected]]
Sent: Tuesday, March 7, 2017 9:16 AM
To: KY Srinivasan <[email protected]>; Haiyang Zhang <[email protected]>; Long Li <[email protected]>; [email protected]; [email protected]; [email protected]
Cc: [email protected]; [email protected]; [email protected]; Stephen Hemminger <[email protected]>
Subject: [PATCH] storvsc: workaround for virtual DVD SCSI version

Hyper-V host emulation of SCSI for virtual DVD device reports SCSI version 0 (UNKNOWN) but is still capable of supporting REPORTLUN.

Without this patch, a GEN2 Linux guest on Hyper-V will not boot 4.11 successfully with virtual DVD ROM device. What happens is that the SCSI scan process falls back to doing sequential probing by INQUIRY. But the storvsc driver has a previous workaround that masks/blocks all errors reports from INQUIRY (or MODE_SENSE) commands. This workaround causes the scan to then populate a full set of bogus LUN's on the target and then sends kernel spinning off into a death spiral doing block reads on the non-existent LUNs.

Quoted 7 lines

By setting the correct blacklist flags, the target with the DVD device is scanned with REPORTLUN and that works correctly.

Patch needs to go in current 4.11, it is safe but not necessary in older kernels.

Quoted 72 lines
---
 drivers/scsi/storvsc_drv.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

PS: The error handling does need to be fixed (have patches pending) but that is interrelated with hotplug and can wait.

drivers/scsi/storvsc_drv.c +0 -0
index 638e5f427c90..19973e874830 100644
@@ -400,8 +400,6 @@

MODULE_PARM_DESC(storvsc_vcpus_per_sub_channel, "Ratio of VCPUs to subchannels")
*/
static int storvsc_timeout = 180;

drivers/scsi/storvsc_drv.c +16 -10
-static int msft_blist_flags = BLIST_TRY_VPD_PAGES;
-
 #if IS_ENABLED(CONFIG_SCSI_FC_ATTRS)
 static struct scsi_transport_template *fc_transport_template;
 #endif
@@ -1383,6 +1381,22 @@ static int storvsc_do_io(struct hv_device *device,
 	return ret;
 }
 
+static int storvsc_device_alloc(struct scsi_device *sdevice)
+{
+	/*
+	 * Set blist flag to permit the reading of the VPD pages even when
+	 * the target may claim SPC-2 compliance. MSFT targets currently
+	 * claim SPC-2 compliance while they implement post SPC-2 features.
+	 * With this flag we can correctly handle WRITE_SAME_16 issues.
+	 *
+	 * Hypervisor reports SCSI_UNKNOWN type for DVD ROM device but
+	 * still supports REPORT LUN.
+	 */
+	sdevice->sdev_bflags = BLIST_REPORTLUN2 |
 BLIST_TRY_VPD_PAGES;
+
+	return 0;
+}
+
 static int storvsc_device_configure(struct scsi_device *sdevice)
 {
@@ -1396,14 +1410,6 @@ static int storvsc_device_configure(struct
 scsi_device *sdevice)
 	sdevice->no_write_same = 1;
 
 	/*
-	 * Add blist flags to permit the reading of the VPD pages even when
-	 * the target may claim SPC-2 compliance. MSFT targets currently
-	 * claim SPC-2 compliance while they implement post SPC-2 features.
-	 * With this patch we can correctly handle WRITE_SAME_16 issues.
-	 */
-	sdevice->sdev_bflags |= msft_blist_flags;
-
-	/*
 	 * If the host is WIN8 or WIN8 R2, claim conformance to SPC-3
 	 * if the device is a MSFT virtual device.  If the host is
drivers/scsi/storvsc_drv.c +1 -1
 	 * WIN10 or newer, allow write_same.
@@ -1661,6 +1667,7 @@ static struct scsi_host_template scsi_driver = {
 	.eh_host_reset_handler =	storvsc_host_reset_handler,
 	.proc_name =		"storvsc_host",
 	.eh_timed_out =		storvsc_eh_timed_out,
+	.slave_alloc =		storvsc_device_alloc,
 	.slave_configure =	storvsc_device_configure,
 	.cmd_per_lun =		255,
 	.this_id =		-1,
--

2.11.0

#3 lore raw
In reply to Stephen Hemminger
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Stephen" == Stephen Hemminger <[email protected]> writes:

Stephen,

Stephen> Hyper-V host emulation of SCSI for virtual DVD device reports Stephen> SCSI version 0 (UNKNOWN) but is still capable of supporting Stephen> REPORTLUN.

Stephen> Without this patch, a GEN2 Linux guest on Hyper-V will not boot Stephen> 4.11 successfully with virtual DVD ROM device. What happens is Stephen> that the SCSI scan process falls back to doing sequential Stephen> probing by INQUIRY. But the storvsc driver has a previous Stephen> workaround that masks/blocks all errors reports from INQUIRY Stephen> (or MODE_SENSE) commands. This workaround causes the scan to Stephen> then populate a full set of bogus LUN's on the target and then Stephen> sends kernel spinning off into a death spiral doing block reads Stephen> on the non-existent LUNs.

Stephen> By setting the correct blacklist flags, the target with the DVD Stephen> device is scanned with REPORTLUN and that works correctly.

Applied to 4.11/scsi-fixes. Thank you!

Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering
Reviewed-by on v1 1/1
#4 lore raw
In reply to Stephen Hemminger
Quoted 18 lines · Stephen Hemminger

Hyper-V host emulation of SCSI for virtual DVD device reports SCSI version 0 (UNKNOWN) but is still capable of supporting REPORTLUN.

Without this patch, a GEN2 Linux guest on Hyper-V will not boot 4.11 successfully with virtual DVD ROM device. What happens is that the SCSI scan process falls back to doing sequential probing by INQUIRY. But the storvsc driver has a previous workaround that masks/blocks all errors reports from INQUIRY (or MODE_SENSE) commands. This workaround causes the scan to then populate a full set of bogus LUN's on the target and then sends kernel spinning off into a death spiral doing block reads on the non-existent LUNs.

By setting the correct blacklist flags, the target with the DVD device is scanned with REPORTLUN and that works correctly.

Patch needs to go in current 4.11, it is safe but not necessary in older kernels.

Quoted 8 lines
---
 drivers/scsi/storvsc_drv.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

PS: The error handling does need to be fixed (have patches pending) but that is interrelated with hotplug and can wait.

Cheers,

Hannes

Signature · 5 lines
Dr. Hannes Reinecke		   Teamlead Storage & Networking
[email protected]			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)