[PATCH] target: Add WRITE_VERIFY_16 Support

[PATCH] target: Add WRITE_VERIFY_16 Support

[Bryant G. Ly]

For clients who want to utilize more than 2GB you need
WRITE AND VERIFY 16 support.

Reported-by: Michael Cyr <mikecyr@linux.vnet.ibm.com>
Signed-off-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
---
 drivers/target/target_core_sbc.c | 1 +
 include/scsi/scsi_proto.h        | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/target/target_core_sbc.c b/drivers/target/target_core_sbc.c
index 04f616b..e26acf2 100644
--- a/drivers/target/target_core_sbc.c
+++ b/drivers/target/target_core_sbc.c
@@ -918,6 +918,7 @@ sbc_parse_cdb(struct se_cmd *cmd, struct sbc_ops *ops)
 		cmd->se_cmd_flags |= SCF_SCSI_DATA_CDB;
 		cmd->execute_cmd = sbc_execute_rw;
 		break;
+	case WRITE_VERIFY_16:
 	case WRITE_16:
 		sectors = transport_get_sectors_16(cdb);
 		cmd->t_task_lba = transport_lba_64(cdb);
diff --git a/include/scsi/scsi_proto.h b/include/scsi/scsi_proto.h
index d1defd1..b39c321 100644
--- a/include/scsi/scsi_proto.h
+++ b/include/scsi/scsi_proto.h
@@ -112,6 +112,7 @@
 #define WRITE_16              0x8a
 #define READ_ATTRIBUTE        0x8c
 #define WRITE_ATTRIBUTE	      0x8d
+#define WRITE_VERIFY_16	      0x8e
 #define VERIFY_16	      0x8f
 #define SYNCHRONIZE_CACHE_16  0x91
 #define WRITE_SAME_16	      0x93
-- 
2.5.4 (Apple Git-61)

Re: [PATCH] target: Add WRITE_VERIFY_16 Support

[Bart Van Assche]

On Fri, 2017-01-27 at 14:49 -0600, Bryant G. Ly wrote:
> For clients who want to utilize more than 2GB you need
> WRITE AND VERIFY 16 support.

Hello Bryant,

This patch introduces a bug for WRITE AND VERIFY(16) parsing that already
exists for WRITE AND VERIFY(10) parsing, namely that the value of the BYTCHK
field is ignored. Is it OK for you to wait a few weeks with adding WRITE AND
VERIFY(16) support and to come back to this after I have posted my WRITE AND
VERIFY(10) fix?

Bart.

Re: [PATCH] target: Add WRITE_VERIFY_16 Support

[Bryant G. Ly]

> On Fri, 2017-01-27 at 14:49 -0600, Bryant G. Ly wrote:
>> For clients who want to utilize more than 2GB you need
>> WRITE AND VERIFY 16 support.
> Hello Bryant,
>
> This patch introduces a bug for WRITE AND VERIFY(16) parsing that already
> exists for WRITE AND VERIFY(10) parsing, namely that the value of the BYTCHK
> field is ignored. Is it OK for you to wait a few weeks with adding WRITE AND
> VERIFY(16) support and to come back to this after I have posted my WRITE AND
> VERIFY(10) fix?
>
> Bart.

Sounds good, thanks. I'll have to look at your patch sets that fixes that bug.