[PATCH] Unchecked return value error in fdt_rw.c

[PATCH] Unchecked return value error in fdt_rw.c

[Ryan Long]

Hi,

I filed an issue on github, but I was told it'd be good to send this to
the mailing list as well.

RTEMS (RTEMS.org) is a free single process, multi-threaded real-time operating 
system with a long history. Your dtc toolchain has been incorporated into 
RTEMS. The RTEMS Project is a member of the Coverity Scan program and it 
flagged an issue with /dtc/libfdt/fdt_rw.c. In trying to be good citizens of 
the wider open source community, the project wants these issues to be reported 
to the upstream owner along with a fix or suggestions. This is one of those 
reports.

When Coverity Scan was ran on some of your code, a "Unchecked return value" 
error was found at line 352 in fdt_rw.c. For similar errors that we received 
for RTEMS, we created a macro that will assert the value returned and "use" the 
return value like so.

int status = fdt_next_tag(fdt, parentoffsetm &nextoffset);
assert(status == 0);
(void)status;

For this patch, I just put the "(void)" statement in front of the
function call, and it achieves the same results. Could this be put in?
We don't want to deviate from our upstream codebases, but we would also
like to get rid of these errors.

Thanks,
Ryan

Ryan Long (1):
  fdt_rw.c: Fix Unchecked return value error

 cpukit/dtc/libfdt/fdt_rw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

-- 
1.8.3.1

[PATCH] fdt_rw.c: Fix Unchecked return value error

[Ryan Long]

Fixed "Unchecked return value" error, by adding "(void)" in front of the
function call.
---
 cpukit/dtc/libfdt/fdt_rw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/cpukit/dtc/libfdt/fdt_rw.c b/cpukit/dtc/libfdt/fdt_rw.c
index 1385425..cdae5dd 100644
--- a/cpukit/dtc/libfdt/fdt_rw.c
+++ b/cpukit/dtc/libfdt/fdt_rw.c
@@ -348,7 +348,7 @@ int fdt_add_subnode_namelen(void *fdt, int parentoffset,
 		return offset;
 
 	/* Try to place the new node after the parent's properties */
-	fdt_next_tag(fdt, parentoffset, &nextoffset); /* skip the BEGIN_NODE */
+	(void) fdt_next_tag(fdt, parentoffset, &nextoffset); /* skip the BEGIN_NODE */
 	do {
 		offset = nextoffset;
 		tag = fdt_next_tag(fdt, offset, &nextoffset);
-- 
1.8.3.1

Re: [PATCH] fdt_rw.c: Fix Unchecked return value error

[Rob Herring]

On Mon, Mar 1, 2021 at 4:27 PM Ryan Long <thisisryanlong-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
> Fixed "Unchecked return value" error, by adding "(void)" in front of the
> function call.

You don't need cover letters for single patches. Move the explanation
here because it's a lot better at explaining the why. The what is not
needed here so much because we can read what the patch does.

You also need a S-o-b.

> ---
>  cpukit/dtc/libfdt/fdt_rw.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/cpukit/dtc/libfdt/fdt_rw.c b/cpukit/dtc/libfdt/fdt_rw.c
> index 1385425..cdae5dd 100644
> --- a/cpukit/dtc/libfdt/fdt_rw.c
> +++ b/cpukit/dtc/libfdt/fdt_rw.c
> @@ -348,7 +348,7 @@ int fdt_add_subnode_namelen(void *fdt, int parentoffset,
>                 return offset;
>
>         /* Try to place the new node after the parent's properties */
> -       fdt_next_tag(fdt, parentoffset, &nextoffset); /* skip the BEGIN_NODE */
> +       (void) fdt_next_tag(fdt, parentoffset, &nextoffset); /* skip the BEGIN_NODE */

Can't we just actually check the error here?

>         do {
>                 offset = nextoffset;
>                 tag = fdt_next_tag(fdt, offset, &nextoffset);
> --
> 1.8.3.1
>

Re: [PATCH] Unchecked return value error in fdt_rw.c

[Rob Herring]

On Mon, Mar 1, 2021 at 4:27 PM Ryan Long <thisisryanlong-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
> Hi,
>
> I filed an issue on github, but I was told it'd be good to send this to
> the mailing list as well.
>
> RTEMS (RTEMS.org) is a free single process, multi-threaded real-time operating
> system with a long history. Your dtc toolchain has been incorporated into
> RTEMS. The RTEMS Project is a member of the Coverity Scan program and it
> flagged an issue with /dtc/libfdt/fdt_rw.c. In trying to be good citizens of
> the wider open source community, the project wants these issues to be reported
> to the upstream owner along with a fix or suggestions. This is one of those
> reports.
>
> When Coverity Scan was ran on some of your code, a "Unchecked return value"
> error was found at line 352 in fdt_rw.c. For similar errors that we received
> for RTEMS, we created a macro that will assert the value returned and "use" the
> return value like so.

Humm, I think Coverity scans run on the Linux kernel too, but I've
never seen any reports on the kernel's copy. Just wondering why...

Rob

Re: [PATCH] Unchecked return value error in fdt_rw.c

[David Gibson]

[-- Attachment #1: Type: text/plain, Size: 1460 bytes --]

On Mon, Mar 01, 2021 at 08:29:49PM -0600, Rob Herring wrote:
> On Mon, Mar 1, 2021 at 4:27 PM Ryan Long <thisisryanlong-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> >
> > Hi,
> >
> > I filed an issue on github, but I was told it'd be good to send this to
> > the mailing list as well.
> >
> > RTEMS (RTEMS.org) is a free single process, multi-threaded real-time operating
> > system with a long history. Your dtc toolchain has been incorporated into
> > RTEMS. The RTEMS Project is a member of the Coverity Scan program and it
> > flagged an issue with /dtc/libfdt/fdt_rw.c. In trying to be good citizens of
> > the wider open source community, the project wants these issues to be reported
> > to the upstream owner along with a fix or suggestions. This is one of those
> > reports.
> >
> > When Coverity Scan was ran on some of your code, a "Unchecked return value"
> > error was found at line 352 in fdt_rw.c. For similar errors that we received
> > for RTEMS, we created a macro that will assert the value returned and "use" the
> > return value like so.
> 
> Humm, I think Coverity scans run on the Linux kernel too, but I've
> never seen any reports on the kernel's copy. Just wondering why...

It runs on dtc upstream as well.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]