[PATCH -2nd repost 1/4] SCSI: aacraid, fix memory leak

[PATCH -2nd repost 1/4] SCSI: aacraid, fix memory leak

[Jiri Slaby]

Stanse found a memory leak on one fail path in aac_send_raw_srb. Add kfree
there.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/aacraid/commctrl.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/scsi/aacraid/commctrl.c b/drivers/scsi/aacraid/commctrl.c
index 0391d75..5c9a8be 100644
--- a/drivers/scsi/aacraid/commctrl.c
+++ b/drivers/scsi/aacraid/commctrl.c
@@ -649,6 +649,7 @@ static int aac_send_raw_srb(struct aac_dev* dev, void __user * arg)
 				     AAC_OPT_NEW_COMM) ?
 				      (dev->scsi_host_ptr->max_sectors << 9) :
 				      65536) {
+					kfree (usg);
 					rcode = -EINVAL;
 					goto cleanup;
 				}
-- 
1.6.4.2

[PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

[Jiri Slaby]

Stanse found that c3cn is poked many times around in
cxgb3i_conn_pdu_ready, there is no need to check if it is NULL.

Remove the test.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/cxgb3i/cxgb3i_pdu.c |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
index 7091050..64bbc28 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
@@ -461,10 +461,8 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
 		skb = skb_peek(&c3cn->receive_queue);
 	}
 	read_unlock(&c3cn->callback_lock);
-	if (c3cn) {
-		c3cn->copied_seq += read;
-		cxgb3i_c3cn_rx_credits(c3cn, read);
-	}
+	c3cn->copied_seq += read;
+	cxgb3i_c3cn_rx_credits(c3cn, read);
 	conn->rxdata_octets += read;
 
 	if (err) {
-- 
1.6.4.2

Re: [PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

[Mike Christie]

Adding Karen and removing adaptec since this is chelsio's driver.

Jiri Slaby wrote:
> Stanse found that c3cn is poked many times around in
> cxgb3i_conn_pdu_ready, there is no need to check if it is NULL.
> 
> Remove the test.
> 
> Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
> Cc: James E.J. Bottomley <James.Bottomley@suse.de>
> ---
>  drivers/scsi/cxgb3i/cxgb3i_pdu.c |    6 ++----
>  1 files changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
> index 7091050..64bbc28 100644
> --- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
> +++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
> @@ -461,10 +461,8 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
>  		skb = skb_peek(&c3cn->receive_queue);
>  	}
>  	read_unlock(&c3cn->callback_lock);
> -	if (c3cn) {
> -		c3cn->copied_seq += read;
> -		cxgb3i_c3cn_rx_credits(c3cn, read);
> -	}
> +	c3cn->copied_seq += read;
> +	cxgb3i_c3cn_rx_credits(c3cn, read);
>  	conn->rxdata_octets += read;
>  
>  	if (err) {

Looks ok to me. The null check was useless since we access c3cn all over 
the function and would have oopsed before we got there.

Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>

RE: [PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

[Karen Xie]

Yes, the check is not necessary.

Thanks,
Karen

-----Original Message-----
From: linux-scsi-owner@vger.kernel.org
[mailto:linux-scsi-owner@vger.kernel.org] On Behalf Of Jiri Slaby
Sent: Wednesday, November 04, 2009 8:15 AM
To: James.Bottomley@suse.de
Cc: aacraid@adaptec.com; james.smart@emulex.com;
linux-scsi@vger.kernel.org; linux-kernel@vger.kernel.org; Jiri Slaby
Subject: [PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

Stanse found that c3cn is poked many times around in
cxgb3i_conn_pdu_ready, there is no need to check if it is NULL.

Remove the test.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/cxgb3i/cxgb3i_pdu.c |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
index 7091050..64bbc28 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
@@ -461,10 +461,8 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
 		skb = skb_peek(&c3cn->receive_queue);
 	}
 	read_unlock(&c3cn->callback_lock);
-	if (c3cn) {
-		c3cn->copied_seq += read;
-		cxgb3i_c3cn_rx_credits(c3cn, read);
-	}
+	c3cn->copied_seq += read;
+	cxgb3i_c3cn_rx_credits(c3cn, read);
 	conn->rxdata_octets += read;
 
 	if (err) {
-- 
1.6.4.2

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: [PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

[Karen Xie]

Yes, the check is not necessary.

Thanks,
Karen

-----Original Message-----
From: linux-scsi-owner@vger.kernel.org
[mailto:linux-scsi-owner@vger.kernel.org] On Behalf Of Jiri Slaby
Sent: Wednesday, November 04, 2009 8:15 AM
To: James.Bottomley@suse.de
Cc: aacraid@adaptec.com; james.smart@emulex.com;
linux-scsi@vger.kernel.org; linux-kernel@vger.kernel.org; Jiri Slaby
Subject: [PATCH -2nd repost 2/4] SCSI: remove unnecessary NULL test

Stanse found that c3cn is poked many times around in
cxgb3i_conn_pdu_ready, there is no need to check if it is NULL.

Remove the test.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/cxgb3i/cxgb3i_pdu.c |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
index 7091050..64bbc28 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_pdu.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_pdu.c
@@ -461,10 +461,8 @@ void cxgb3i_conn_pdu_ready(struct s3_conn *c3cn)
 		skb = skb_peek(&c3cn->receive_queue);
 	}
 	read_unlock(&c3cn->callback_lock);
-	if (c3cn) {
-		c3cn->copied_seq += read;
-		cxgb3i_c3cn_rx_credits(c3cn, read);
-	}
+	c3cn->copied_seq += read;
+	cxgb3i_c3cn_rx_credits(c3cn, read);
 	conn->rxdata_octets += read;
 
 	if (err) {
-- 
1.6.4.2

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH -2nd repost 3/4] SCSI: lpfc, fix memory leak

[Jiri Slaby]

Stanse found a memory leak on one fail path in lpfc_sli4_read_rev.

Add there kfree.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James Smart <james.smart@emulex.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/lpfc/lpfc_sli.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 43cbe33..7750868 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -4211,6 +4211,7 @@ lpfc_sli4_read_rev(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq,
 	if (rc) {
 		dma_free_coherent(&phba->pcidev->dev, dma_size,
 				  dmabuf->virt, dmabuf->phys);
+		kfree(dmabuf);
 		return -EIO;
 	}
 
-- 
1.6.4.2

[PATCH -2nd repost 4/4] SCSI: scsi_lib, fix potential NULL dereference

[Jiri Slaby]

Stanse found a potential NULL dereference in scsi_kill_request.

Instead of triggering BUG() in 'if (unlikely(cmd == NULL))' branch,
the kernel will Oops earlier on cmd dereference.

Move the dereferences after the if.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: James E.J. Bottomley <James.Bottomley@suse.de>
---
 drivers/scsi/scsi_lib.c |    9 ++++++---
 1 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index ed98279..47ee130 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1359,9 +1359,9 @@ static int scsi_lld_busy(struct request_queue *q)
 static void scsi_kill_request(struct request *req, struct request_queue *q)
 {
 	struct scsi_cmnd *cmd = req->special;
-	struct scsi_device *sdev = cmd->device;
-	struct scsi_target *starget = scsi_target(sdev);
-	struct Scsi_Host *shost = sdev->host;
+	struct scsi_device *sdev;
+	struct scsi_target *starget;
+	struct Scsi_Host *shost;
 
 	blk_start_request(req);
 
@@ -1371,6 +1371,9 @@ static void scsi_kill_request(struct request *req, struct request_queue *q)
 		BUG();
 	}
 
+	sdev = cmd->device;
+	starget = scsi_target(sdev);
+	shost = sdev->host;
 	scsi_init_cmd_errh(cmd);
 	cmd->result = DID_NO_CONNECT << 16;
 	atomic_inc(&cmd->device->iorequest_cnt);
-- 
1.6.4.2