2014-02-07 16:27:18

by Michele Baldessari

Subject: [PATCH] Fix the arguments order of a few iscsi_conn_printk() calls

A few places had the opcode and itt arguments reversed.
Fix the order.

Reported-by: Josef Moellers <[email protected]>
Signed-off-by: Michele Baldessari <[email protected]>
---
drivers/scsi/libiscsi.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 4046241..cd330c0 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -1013,13 +1013,13 @@ static int iscsi_handle_reject(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
iscsi_conn_printk(KERN_ERR, conn,
"pdu (op 0x%x itt 0x%x) rejected "
"due to DataDigest error.\n",
- rejected_pdu.itt, opcode);
+ opcode, rejected_pdu.itt);
break;
case ISCSI_REASON_IMM_CMD_REJECT:
iscsi_conn_printk(KERN_ERR, conn,
"pdu (op 0x%x itt 0x%x) rejected. Too many "
"immediate commands.\n",
- rejected_pdu.itt, opcode);
+ opcode, rejected_pdu.itt);
/*
* We only send one TMF at a time so if the target could not
* handle it, then it should get fixed (RFC mandates that
@@ -1059,8 +1059,8 @@ static int iscsi_handle_reject(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
default:
iscsi_conn_printk(KERN_ERR, conn,
"pdu (op 0x%x itt 0x%x) rejected. Reason "
- "code 0x%x\n", rejected_pdu.itt,
- rejected_pdu.opcode, reject->reason);
+ "code 0x%x\n", rejected_pdu.opcode,
+ rejected_pdu.itt, reject->reason);
break;
}
return rc;
--
1.8.5.3

2014-02-07 21:11:57

by Mike Christie

Subject: Re: [PATCH] Fix the arguments order of a few iscsi_conn_printk() calls

On 02/07/2014 10:18 AM, Michele Baldessari wrote:
> A few places had the opcode and itt arguments reversed.
> Fix the order.
>
> Reported-by: Josef Moellers <[email protected]>
> Signed-off-by: Michele Baldessari <[email protected]>


Thanks. I just acked a patch to do the same:
https://lkml.org/lkml/2014/1/8/754

Have you been hitting that code and with what target and what type of
command?

2014-02-07 21:55:28

by Michele Baldessari

Subject: Re: [PATCH] Fix the arguments order of a few iscsi_conn_printk() calls

Hi Mike,

On Fri, Feb 07, 2014 at 03:11:32PM -0600, Mike Christie wrote:
> On 02/07/2014 10:18 AM, Michele Baldessari wrote:
> > A few places had the opcode and itt arguments reversed.
> > Fix the order.
> >
> > Reported-by: Josef Moellers <[email protected]>
> > Signed-off-by: Michele Baldessari <[email protected]>
>
>
> Thanks. I just acked a patch to do the same:
> https://lkml.org/lkml/2014/1/8/754

ah ok. Too many mailing lists and I did not see it in git master yet ;)

> Have you been hitting that code and with what target and what type of
> command?

I was made aware of it by a colleague who was looking at code. So no
real-life situation here.

regards,
Michele
--
Michele Baldessari <[email protected]>
C2A5 9DA3 9961 4FFB E01B D0BC DDD4 DCCB 7515 5C6D