[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20100604151613.2594bf7c.akpm@linux-foundation.org>
Date: Fri, 4 Jun 2010 15:16:13 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Thomas Weber <weber@...science.de>
Cc: zippel@...ux-m68k.org, linux-kbuild@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] scripts: confdata: Fix fwrite warning
On Wed, 2 Jun 2010 16:44:19 +0200
Thomas Weber <weber@...science.de> wrote:
> I get for:
> scripts/kconfig/confdata.c:508
> scripts/kconfig/confdata.c:759
> scripts/kconfig/confdata.c:760
>
> warning: ignoring return value of ___fwrite___, declared with attribute warn_unused_result
>
> So check the return value and handle the error.
>
> Signed-off-by: Thomas Weber <weber@...science.de>
> ---
> scripts/kconfig/confdata.c | 15 ++++++++++++---
> 1 files changed, 12 insertions(+), 3 deletions(-)
>
> diff --git a/scripts/kconfig/confdata.c b/scripts/kconfig/confdata.c
> index c4dec80..4741fea 100644
> --- a/scripts/kconfig/confdata.c
> +++ b/scripts/kconfig/confdata.c
> @@ -505,7 +505,10 @@ int conf_write(const char *name)
> while (1) {
> l = strcspn(str, "\"\\");
> if (l) {
> - fwrite(str, l, 1, out);
> + if (fwrite(str, l, 1, out) != 1) {
> + fprintf(stderr, "write str failed\n");
> + return 1;
> + }
Missing an fclose(out), although that probably doesn't matter a lot in
practice
Also, we prefer to avoid putting `return' statements deep inside large
C functions as they can easily lead to resource leaks. Such as
forgetting to fclose(out) ;)
> str += l;
> }
> if (!*str)
> @@ -756,8 +759,14 @@ int conf_write_autoconf(void)
> while (1) {
> l = strcspn(str, "\"\\");
> if (l) {
> - fwrite(str, l, 1, out);
> - fwrite(str, l, 1, out_h);
> + if (fwrite(str, l, 1, out) != 1) {
> + fprintf(stderr, "write str failed\n");
> + return 1;
> + }
> + if (fwrite(str, l, 1, out_h) != 1) {
> + fprintf(stderr, "write str failed\n");
> + return 1;
> + }
Dittoes.
Also, this patch would be neater if it added a little helper function
around fwrite(), rather than triplicating the same code sequence.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists