On Mon, 15 Jun 2020 15:14:42 -0400
Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> On Sat, 13 Jun 2020 00:23:35 +0900
> Masami Hiramatsu <mhiramat@xxxxxxxxxx> wrote:
>
> > Fix bootconfig to return 0 if succeeded to show the bootconfig
> > in initrd. Without this fix, "bootconfig INITRD" command
> > returns !0 even if the command succeeded to show the bootconfig.
> >
> > Fixes: 950313ebf79c ("tools: bootconfig: Add bootconfig command")
> > Cc: stable@xxxxxxxxxxxxxxx
> > Signed-off-by: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
> > ---
> > tools/bootconfig/main.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/tools/bootconfig/main.c b/tools/bootconfig/main.c
> > index 21896a6675fd..ff2cc9520e10 100644
> > --- a/tools/bootconfig/main.c
> > +++ b/tools/bootconfig/main.c
> > @@ -209,8 +209,10 @@ int show_xbc(const char *path)
> > ret = load_xbc_from_initrd(fd, &buf);
> > if (ret < 0)
> > pr_err("Failed to load a boot config from initrd: %d\n", ret);
> > - else
> > + else {
> > xbc_show_compact_tree();
> > + ret = 0;
> > + }
>
> Usually for the above, I think goto is cleaner. As it is strange to
> have a successful case as the "else" condition. Not to mention, if you
> add brackets for one side of the else, it is usually recommended to add
> them for the other side.
>
> But in this case I think it would read better to have:
>
>
> if (ret < 0) {
> pr_err(...);
> goto out;
> }
> xbc_show_compact_tree();
> ret = 0;
> out:
Agreed. OK, I'll update it.
Thank you!
>
> >
> > close(fd);
> > free(buf);
>
> -- Steve
--
Masami Hiramatsu <mhiramat@xxxxxxxxxx>
