send readcapacity10 when readcapacity16 failed
PATCH 1/1 send readcapacity10 when readcapacity16 failed
When play with Dell MD3000 target, for sure it is a TYPE_DISK, but readcapacity16 would fail. Then we find that readcapacity10 succeeded. It looks like the target just support readcapacity10 even through it is a TYPE_DISK or have some TYPE_ROM characteristics.
This patch can give a chance to send readcapacity16 when readcapacity10 failed. This patch is not harmful to original pathes
--- block/iscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
block/iscsi.c +2 -1
@@ -1243,8 +1243,9 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) iscsilun->lbprz = !!rc16->lbprz; iscsilun->use_16_for_rw = (rc16->returned_lba >0xffffffff ); } +break ; } -break ; +//fall through to try readcapacity10 instead case TYPE_ROM : task =iscsi_readcapacity10_sync (iscsilun->iscsi, iscsilun->lun,0 ,0 );if (task !=NULL && task->status ==SCSI_STATUS_GOOD ) {
Signature · 1 line
2.6.2
Quoted 22 lines · Zhu Lingshan
When play with Dell MD3000 target, for sure it is a
TYPE_DISK, but readcapacity16 would fail. Then we find that readcapacity10 succeeded. It looks like the target just support readcapacity10 even through it is aTYPE_DISKor have someTYPE_ROMcharacteristics.This patch can give a chance to send readcapacity16 when readcapacity10 failed. This patch is not harmful to original pathes
Signed-off-by Zhu Lingshan <[email protected]>--- block/iscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)block/iscsi.c +0
-0
block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) · line 1251
Quoted 10 lines
block/iscsi.c +2
-1@@ -1243,8 +1243,9 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) iscsilun->lbprz = !!rc16->lbprz; iscsilun->use_16_for_rw = (rc16->returned_lba >0xffffffff ); } +break ; } -break ; +//fall through to try readcapacity10 instead case TYPE_ROM : task =iscsi_readcapacity10_sync (iscsilun->iscsi, iscsilun->lun,0 ,0 );if (task !=NULL && task->status ==SCSI_STATUS_GOOD ) {
For the uninitiated, why does readcapacity16 fail?
My gut feeling is that this is a hack, because:
- Either readcapacity16 should work, or
- We shouldn't be choosing 10/16 based on the target type to begin with
but I don't know much about iSCSI, so maybe You, Paolo or Peter could fill me in.
--js
MMC devices:
READ CAPACITY 10 support is mandatory.
No support for READ CAPACITY 16
SBC devices:
READ CAPACITY 10 is mandatory
READ CAPACITY 16 support is only required when you have thin provisioning or protection information (or if the device is >2^32 blocks) Almost all, but apparently not all, SBC devices support both.
For SBC devices you probably want to start with RC16 and only fallback to RC10 if you get INVALID_OPCODE.
You start with RC16 since this is the way to discover if you have thin provisioning or special protection information.
For MMC devices you could try the "try RC16 first and fallback to RC10" but as probably almost no MMC devices support RC16 it makes little sense to do so.
Quoted 51 lines · John Snow
block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun
Quoted 21 lines · Zhu Lingshan
When play with Dell MD3000 target, for sure it is a
TYPE_DISK, but readcapacity16 would fail. Then we find that readcapacity10 succeeded. It looks like the target just support readcapacity10 even through it is aTYPE_DISKor have someTYPE_ROMcharacteristics.This patch can give a chance to send readcapacity16 when readcapacity10 failed. This patch is not harmful to original pathes
Signed-off-by Zhu Lingshan <[email protected]>--- block/iscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)block/iscsi.c +0
-0@@ -1243,8 +1243,9 @@ static void iscsi_readcapacity_sync(IscsiLun*iscsilun, Error **errp)
block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun · line 1244
Quoted 2 lines
block/iscsi.c +0
-0iscsilun->lbprz = !!rc16->lbprz; iscsilun->use_16_for_rw = (rc16->returned_lba >
0xffffffff);block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
Quoted 7 lines
block/iscsi.c +2
-1} +break ; } -break ; +//fall through to try readcapacity10 instead case TYPE_ROM : task = iscsi_readcapacity10_sync(iscsilun->iscsi,
iscsilun->lun, 0, 0);block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) · line 1251
Quoted 2 lines
block/iscsi.c +0
-0if (task !=NULL && task->status ==SCSI_STATUS_GOOD ) {For the uninitiated, why does readcapacity16 fail?
My gut feeling is that this is a hack, because:
- Either readcapacity16 should work, or
- We shouldn't be choosing 10/16 based on the target type to begin withbut I don't know much about iSCSI, so maybe You, Paolo or Peter could fill me in.
--js
Quoted 11 lines · ronnie sahlberg
MMC devices:
READ CAPACITY 10 support is mandatory.
No support for READ CAPACITY 16SBC devices:
READ CAPACITY 10 is mandatory
READ CAPACITY 16 support is only required when you have thin provisioning or protection information (or if the device is >2^32 blocks) Almost all, but apparently not all, SBC devices support both.
Quoted 10 lines
For SBC devices you probably want to start with RC16 and only fallback to RC10 if you get
INVALID_OPCODE.
You start with RC16 since this is the way to discover if you have thin provisioning or special protection information.For MMC devices you could try the "try RC16 first and fallback to RC10" but as probably almost no MMC devices support RC16 it makes little sense to do so.
Ronnie: Thanks for the explanation!
Zhu: In light of this, can the patch be reworked slightly to explicitly check *why* READCAPACITY16 failed and only attempt the READCAPACITY10 as a fallback if it receives INVALID_OPCODE?
If it fails for any other reason it's probably best to report the error and let QEMU decide what to do about it.
I suppose caching a flag that lets us know to go straight to READ_CAPACITY10 is not worthwhile because this command is not likely to be issued very often.
Thanks,
--js
Quoted 56 lines
On Tue, Jan 5, 2016 at 11:42 AM, John Snow <[email protected] <mailto:[email protected]>> wrote:
On 12/28/2015 10:32 PM, Zhu Lingshan wrote: > When play with Dell MD3000 target, for sure it > is a TYPE_DISK, but readcapacity16 would fail. > Then we find that readcapacity10 succeeded. It > looks like the target just support readcapacity10 > even through it is a TYPE_DISK or have some > TYPE_ROM characteristics. > > This patch can give a chance to send > readcapacity16 when readcapacity10 failed. > This patch is not harmful to original pathes > > Signed-off-by: Zhu Lingshan <[email protected] <mailto:[email protected]>> > --- > block/iscsi.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/block/iscsi.c b/block/iscsi.c > index bd1f1bf..c8d167f 100644 > --- a/block/iscsi.c > +++ b/block/iscsi.c > @@ -1243,8 +1243,9 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) > iscsilun->lbprz = !!rc16->lbprz; > iscsilun->use_16_for_rw = (rc16->returned_lba > 0xffffffff); > } > + break; > } > - break; > + //fall through to try readcapacity10 instead > case TYPE_ROM: > task = iscsi_readcapacity10_sync(iscsilun->iscsi, iscsilun->lun, 0, 0); > if (task != NULL && task->status == SCSI_STATUS_GOOD) { > For the uninitiated, why does readcapacity16 fail? My gut feeling is that this is a hack, because: - Either readcapacity16 should work, or - We shouldn't be choosing 10/16 based on the target type to begin with but I don't know much about iSCSI, so maybe You, Paolo or Peter could fill me in. --js
Quoted 8 lines · John Snow
Ronnie: Thanks for the explanation!
Zhu: In light of this, can the patch be reworked slightly to explicitly check *why* READCAPACITY16 failed and only attempt the READCAPACITY10 as a fallback if it receives
INVALID_OPCODE?If it fails for any other reason it's probably best to report the error and let QEMU decide what to do about it.
Any other failure probably would happen for READ CAPACITY(10) as well, so it's okay to ignore it for READ CAPACITY(16).
Zhu's patch matches what Linux does by default, it seems okay. The only change needed is to retry READ CAPACITY(16) if there is a UNIT ATTENTION sense:
+ if (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION
+ && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) {
+ break;
+ }
Paolo
Am 07.01.2016 um 11:07 schrieb Paolo Bonzini:
Quoted 22 lines · John Snow
Quoted 8 lines
Ronnie: Thanks for the explanation!
Zhu: In light of this, can the patch be reworked slightly to explicitly check *why* READCAPACITY16 failed and only attempt the READCAPACITY10 as a fallback if it receives
INVALID_OPCODE?If it fails for any other reason it's probably best to report the error and let QEMU decide what to do about it.
Any other failure probably would happen for READ CAPACITY(10) as well, so it's okay to ignore it for READ CAPACITY(16).
Zhu's patch matches what Linux does by default, it seems okay. The only change needed is to retry READ CAPACITY(16) if there is a UNIT ATTENTION sense:
(quoted diff) +4
-0+ if (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION + && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) { + break; + }
Quoted 1 line
Paolo
Paolo, Ronnie, do you know what Readcapacity(10) returns if the target blocks count is greater than what can be described in 32bit?
Anyway, is there a new version of this patch? I would also like to have a look before it is commited.
Thanks,
Peter
Quoted 7 lines · Peter Lieven
Review
Quoted 4 lines
(quoted diff) +4
-0+ if (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION + && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) { + break; + }Paolo, Ronnie, do you know what Readcapacity(10) returns if the target blocks count is greater than what can be described in 32bit?
Yes, it returns 0xFFFFFFFF.
Quoted 2 lines
Anyway, is there a new version of this patch? I would also like to have a look before it is commited.
See https://github.com/bonzini/qemu/commit/1efcbda37.
Paolo
Paolo Bonzini
AppliedQuoted 22 lines · Zhu Lingshan
When play with Dell MD3000 target, for sure it is a
TYPE_DISK, but readcapacity16 would fail. Then we find that readcapacity10 succeeded. It looks like the target just support readcapacity10 even through it is aTYPE_DISKor have someTYPE_ROMcharacteristics.This patch can give a chance to send readcapacity16 when readcapacity10 failed. This patch is not harmful to original pathes
Signed-off-by Zhu Lingshan <[email protected]>--- block/iscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)block/iscsi.c +0
-0
block/iscsi.c static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) · line 1251
Quoted 10 lines
block/iscsi.c +2
-1@@ -1243,8 +1243,9 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp) iscsilun->lbprz = !!rc16->lbprz; iscsilun->use_16_for_rw = (rc16->returned_lba >0xffffffff ); } +break ; } -break ; +//fall through to try readcapacity10 instead case TYPE_ROM : task =iscsi_readcapacity10_sync (iscsilun->iscsi, iscsilun->lun,0 ,0 );if (task !=NULL && task->status ==SCSI_STATUS_GOOD ) {
Queued, thanks.
Paolo