2020-05-19 16:11:30

by Gustavo A. R. Silva

Subject: [PATCH] RDMA/rtrs: client: Fix function return on success

The function should return 0 on success, instead of err.

Addresses-Coverity-ID: 1493753 ("Identical code for different branches")
Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality")
Signed-off-by: Gustavo A. R. Silva <[email protected]>
---
drivers/infiniband/ulp/rtrs/rtrs-clt.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
index 468fdd0d8713c..465515e46bb1a 100644
--- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
+++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
@@ -1594,7 +1594,8 @@ static int create_con_cq_qp(struct rtrs_clt_con *con)

if (err)
return err;
- return err;
+
+ return 0;
}

static void destroy_con_cq_qp(struct rtrs_clt_con *con)
--
2.26.2

2020-05-19 16:18:34

by Bart Van Assche

Subject: Re: [PATCH] RDMA/rtrs: client: Fix function return on success

On 2020-05-19 09:13, Gustavo A. R. Silva wrote:
> The function should return 0 on success, instead of err.
>
> Addresses-Coverity-ID: 1493753 ("Identical code for different branches")
> Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality")
> Signed-off-by: Gustavo A. R. Silva <[email protected]>
> ---
> drivers/infiniband/ulp/rtrs/rtrs-clt.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> index 468fdd0d8713c..465515e46bb1a 100644
> --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> @@ -1594,7 +1594,8 @@ static int create_con_cq_qp(struct rtrs_clt_con *con)
>
> if (err)
> return err;
> - return err;
> +
> + return 0;
> }

Why to keep the if-statement? Has the following been considered?

diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
index 8dfa56dc32bc..a7f5d55f8542 100644
--- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
+++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
@@ -1587,8 +1587,6 @@ static int create_con_cq_qp(struct rtrs_clt_con *con)
* since destroy_con_cq_qp() must be called.
*/

- if (err)
- return err;
return err;
}

Thanks,

Bart.

2020-05-19 16:25:50

by Jinpu Wang

Subject: Re: [PATCH] RDMA/rtrs: client: Fix function return on success

On Tue, May 19, 2020 at 6:16 PM Bart Van Assche <[email protected]> wrote:
>
> On 2020-05-19 09:13, Gustavo A. R. Silva wrote:
> > The function should return 0 on success, instead of err.
> >
> > Addresses-Coverity-ID: 1493753 ("Identical code for different branches")
> > Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality")
> > Signed-off-by: Gustavo A. R. Silva <[email protected]>
> > ---
> > drivers/infiniband/ulp/rtrs/rtrs-clt.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> > index 468fdd0d8713c..465515e46bb1a 100644
> > --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> > +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> > @@ -1594,7 +1594,8 @@ static int create_con_cq_qp(struct rtrs_clt_con *con)
> >
> > if (err)
> > return err;
> > - return err;
> > +
> > + return 0;
> > }
>
> Why to keep the if-statement? Has the following been considered?
>
> diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> index 8dfa56dc32bc..a7f5d55f8542 100644
> --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c
> @@ -1587,8 +1587,6 @@ static int create_con_cq_qp(struct rtrs_clt_con *con)
> * since destroy_con_cq_qp() must be called.
> */
>
> - if (err)
> - return err;
> return err;
> }
>
> Thanks,
>
> Bart.

This one seems better, thanks Bart and Gustavo, Gustavo, would you
like to send a v2 patch as Bart suggested?

Thanks,