[PATCH] usnix: correctly handle kzalloc return value

[PATCH] usnix: correctly handle kzalloc return value

[Insu Yun]

Since kzalloc returns memory address, not error code,
it should be checked whether it is null or not. 

Signed-off-by: Insu Yun <wuninsu@gmail.com>
---
 drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
index db3588d..a678a66 100644
--- a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
+++ b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
@@ -221,8 +221,8 @@ create_roce_custom_flow(struct usnic_ib_qp_grp *qp_grp,
 
 	/* Create Flow Handle */
 	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
-	if (IS_ERR_OR_NULL(qp_flow)) {
-		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
+	if (!qp_flow) {
+		err = -ENOMEM;
 		goto out_dealloc_flow;
 	}
 	qp_flow->flow = flow;
@@ -296,8 +296,8 @@ create_udp_flow(struct usnic_ib_qp_grp *qp_grp,
 
 	/* Create qp_flow */
 	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
-	if (IS_ERR_OR_NULL(qp_flow)) {
-		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
+	if (!qp_flow) {
+		err = -ENOMEM;
 		goto out_dealloc_flow;
 	}
 	qp_flow->flow = flow;
-- 
1.9.1

Re: [PATCH] usnix: correctly handle kzalloc return value

[Dave Goodell]

On Sat, Oct 17, 2015 at 07:06:56PM +0000, Insu Yun wrote:
> Since kzalloc returns memory address, not error code,
> it should be checked whether it is null or not. 
> 
> Signed-off-by: Insu Yun <wuninsu@gmail.com>
> ---
>  drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)

The subject says "usnix" where it should say "usnic".  Otherwise:

Reviewed-by: Dave Goodell <dgoodell@cisco.com>

If you have more of these fixes for drivers/infiniband/hw/usnic/, you
can send them all through just Doug Ledford and linux-rdma@, no need to
copy all of linux-kernel@.  You can also CC me instead of Upinder, whose
email bounces at this point, though I will see your patches sent to
linux-rdma@ even if I'm not explicitly CC-ed.  The patch to update
MAINTAINERS for usnic has been submitted but hasn't yet percolated
somewhere you would have seen it:

https://patchwork.kernel.org/patch/7410421/

-Dave

Re: [PATCH] usnix: correctly handle kzalloc return value

[Doug Ledford]

[-- Attachment #1: Type: text/plain, Size: 1396 bytes --]

On 10/17/2015 03:06 PM, Insu Yun wrote:
> Since kzalloc returns memory address, not error code,
> it should be checked whether it is null or not. 
> 
> Signed-off-by: Insu Yun <wuninsu@gmail.com>

Thanks, applied.

> ---
>  drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
> index db3588d..a678a66 100644
> --- a/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
> +++ b/drivers/infiniband/hw/usnic/usnic_ib_qp_grp.c
> @@ -221,8 +221,8 @@ create_roce_custom_flow(struct usnic_ib_qp_grp *qp_grp,
>  
>  	/* Create Flow Handle */
>  	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
> -	if (IS_ERR_OR_NULL(qp_flow)) {
> -		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
> +	if (!qp_flow) {
> +		err = -ENOMEM;
>  		goto out_dealloc_flow;
>  	}
>  	qp_flow->flow = flow;
> @@ -296,8 +296,8 @@ create_udp_flow(struct usnic_ib_qp_grp *qp_grp,
>  
>  	/* Create qp_flow */
>  	qp_flow = kzalloc(sizeof(*qp_flow), GFP_ATOMIC);
> -	if (IS_ERR_OR_NULL(qp_flow)) {
> -		err = qp_flow ? PTR_ERR(qp_flow) : -ENOMEM;
> +	if (!qp_flow) {
> +		err = -ENOMEM;
>  		goto out_dealloc_flow;
>  	}
>  	qp_flow->flow = flow;
> 


-- 
Doug Ledford <dledford@redhat.com>
              GPG KeyID: 0E572FDD



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 884 bytes --]