Messages in this thread | | | From | "Rafael J. Wysocki" <> | Subject | Re: [PATCH v7 2/3] trivial: PM / Hibernate: clean up checkpatch in hibernate.c | Date | Wed, 05 Feb 2014 01:03:29 +0100 |
| |
On Tuesday, February 04, 2014 03:22:22 PM Sebastian Capella wrote: > Quoting Sebastian Capella (2014-02-04 14:37:33) > > Quoting Rafael J. Wysocki (2014-02-04 13:36:29) > > > > static int __init resumedelay_setup(char *str) > > > > { > > > > - resume_delay = simple_strtoul(str, NULL, 0); > > > > + int ret = kstrtoint(str, 0, &resume_delay); > > > > + /* mask must_check warn; on failure, leaves resume_delay unchanged */ > > > > + (void)ret; > > One unintended consequence of this change is that it'll now accept a > negative integer parameter.
Well, what about using kstrtouint(), then?
> I'll rework this to have the same behavior as before. > > BTW, one question, is the __must_check really needed on kstrtoint? > Wouldn't it be acceptable to rely on kstrtoint to not update resume_delay > if it's unable to parse an integer out of the string? Couldn't that be > a sufficient effect without requiring checking the return?
Well, kstrtoint() is used in some security-sensitive places AFAICS, so it really is better to check its return value in general. The __must_check reminds people about that.
Thanks!
-- I speak only for myself. Rafael J. Wysocki, Intel Open Source Technology Center.
| |