Messages in this thread | | | Date | Thu, 13 Mar 2008 08:44:48 +0800 | From | "Dave Young" <> | Subject | Re: [PATCH -mm] do not check condition twice in WARN_ON_SECS |
| |
On Wed, Mar 12, 2008 at 9:47 PM, Johannes Weiner <hannes@saeurebad.de> wrote: > Hi Dave, > > > > Dave Young <hidave.darkstar@gmail.com> writes: > > > Don't check condition twice, change WARN_ON(condition) to WARN_ON(1) > > Thanks Marcin Slusarz <marcin.slusarz@gmail.com> for pointing out > > > > Signed-off-by: Dave Young <hidave.darkstar@gmail.com> > > > > --- > > include/asm-generic/bug.h | 3 ++- > > 1 file changed, 2 insertions(+), 1 deletion(-) > > > > diff -upr linux/include/asm-generic/bug.h linux.new/include/asm-generic/bug.h > > --- linux/include/asm-generic/bug.h 2008-03-12 08:45:08.000000000 +0800 > > +++ linux.new/include/asm-generic/bug.h 2008-03-12 09:04:07.000000000 +0800 > > @@ -80,7 +80,8 @@ extern void warn_on_slowpath(const char > > int __ret_warn_on = !!(condition); \ > > if (unlikely(__ret_warn_on)) \ > > if (__ratelimit(secs * HZ, 1)) \ > > - WARN_ON(condition); \ > > + WARN_ON(1); \ > > + unlikely(__ret_warn_on); \ > > }) > > What's wrong with: > > #define WARN_ON_SECS(condition, secs) \ > WARN_ON(condition && __ratelimit(secs * HZ, 1))
Looks concise. Should I update the third time?
| |