[3.5.y.z extended stable] Patch "iscsi-target: Fix-up all zero data-length CDBs with R/W_BIT set" has been added to staging queue

[Luis Henriques]

This is a note to let you know that I have just added a patch titled

    iscsi-target: Fix-up all zero data-length CDBs with R/W_BIT set

to the linux-3.5.y-queue branch of the 3.5.y.z extended stable tree 
which can be found at:

 http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.5.y-queue

If you, or anyone else, feels it should not be added to this tree, please 
reply to this email.

For more information about the 3.5.y.z tree, see
https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable

Thanks.
-Luis

------

>From b7f04162e8fe1a782167f5d00a29c376071ace42 Mon Sep 17 00:00:00 2001
From: Nicholas Bellinger <nab at linux-iscsi.org>
Date: Mon, 25 Nov 2013 14:53:57 -0800
Subject: iscsi-target: Fix-up all zero data-length CDBs with R/W_BIT set

commit 4454b66cb67f14c33cd70ddcf0ff4985b26324b7 upstream.

This patch changes special case handling for ISCSI_OP_SCSI_CMD
where an initiator sends a zero length Expected Data Transfer
Length (EDTL), but still sets the WRITE and/or READ flag bits
when no payload transfer is requested.

Many, many moons ago two special cases where added for an ancient
version of ESX that has long since been fixed, so instead of adding
a new special case for the reported bug with a Broadcom 57800 NIC,
go ahead and always strip off the incorrect WRITE + READ flag bits.

Also, avoid sending a reject here, as RFC-3720 does mandate this
case be handled without protocol error.

Reported-by: Witold Bazakbal <865perl at wp.pl>
Tested-by: Witold Bazakbal <865perl at wp.pl>
Signed-off-by: Nicholas Bellinger <nab at linux-iscsi.org>
[ luis: backported to 3.5: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 drivers/target/iscsi/iscsi_target.c | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 3e7ac0c..955392e 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -803,24 +803,22 @@ static int iscsit_handle_scsi_cmd(
 	if (((hdr->flags & ISCSI_FLAG_CMD_READ) ||
 	     (hdr->flags & ISCSI_FLAG_CMD_WRITE)) && !hdr->data_length) {
 		/*
-		 * Vmware ESX v3.0 uses a modified Cisco Initiator (v3.4.2)
-		 * that adds support for RESERVE/RELEASE.  There is a bug
-		 * add with this new functionality that sets R/W bits when
-		 * neither CDB carries any READ or WRITE datapayloads.
+		 * From RFC-3720 Section 10.3.1:
+		 *
+		 * "Either or both of R and W MAY be 1 when either the
+		 *  Expected Data Transfer Length and/or Bidirectional Read
+		 *  Expected Data Transfer Length are 0"
+		 *
+		 * For this case, go ahead and clear the unnecssary bits
+		 * to avoid any confusion with ->data_direction.
 		 */
-		if ((hdr->cdb[0] == 0x16) || (hdr->cdb[0] == 0x17)) {
-			hdr->flags &= ~ISCSI_FLAG_CMD_READ;
-			hdr->flags &= ~ISCSI_FLAG_CMD_WRITE;
-			goto done;
-		}
+		hdr->flags &= ~ISCSI_FLAG_CMD_READ;
+		hdr->flags &= ~ISCSI_FLAG_CMD_WRITE;

-		pr_err("ISCSI_FLAG_CMD_READ or ISCSI_FLAG_CMD_WRITE"
+		pr_warn("ISCSI_FLAG_CMD_READ or ISCSI_FLAG_CMD_WRITE"
 			" set when Expected Data Transfer Length is 0 for"
-			" CDB: 0x%02x. Bad iSCSI Initiator.\n", hdr->cdb[0]);
-		return iscsit_add_reject(ISCSI_REASON_BOOKMARK_INVALID, 1,
-				buf, conn);
+			" CDB: 0x%02x, Fixing up flags\n", hdr->cdb[0]);
 	}
-done:

 	if (!(hdr->flags & ISCSI_FLAG_CMD_READ) &&
 	    !(hdr->flags & ISCSI_FLAG_CMD_WRITE) && (hdr->data_length != 0)) {
--
1.8.3.2