On Wed, 10 Jun 2009 10:11:49 -0600, Jim Cromie wrote: > On Thu, Jun 4, 2009 at 1:30 AM, Jean Delvare <khali at linux-fr.org> wrote: > > > Make sure __devexit and devexit_p() match in all hwmon drivers. > > Suggested by a similar fix from Mike Frysinger. > > > > Signed-off-by: Jean Delvare <khali at linux-fr.org> > > Cc: Hans de Goede <hdegoede at redhat.com> > > Cc: Jonathan Cameron <jic23 at cam.ac.uk> > > --- > > drivers/hwmon/f71882fg.c | 2 +- > > drivers/hwmon/sht15.c | 10 +++++----- > > 2 files changed, 6 insertions(+), 6 deletions(-) > > > > --- linux-2.6.30-rc8.orig/drivers/hwmon/f71882fg.c 2009-06-03 > > 10:55:43.000000000 +0200 > > +++ linux-2.6.30-rc8/drivers/hwmon/f71882fg.c 2009-06-04 > > 09:02:47.000000000 +0200 > > @@ -253,7 +253,7 @@ static struct platform_driver f71882fg_d > > .name = DRVNAME, > > }, > > .probe = f71882fg_probe, > > - .remove = __devexit_p(f71882fg_remove), > > + .remove = f71882fg_remove, > > }; > > > > > I think that one is reversed, certainly vs following. This is on purpose. f71882fg_remove() is not marked with __devexit, so there is no point in protecting references thereto with __devexit_p(). Note that, contrary to the other way, this one does not cause build errors. It's just needlessly complex, and confusing for the reader. > > - .remove = sht15_remove, > > + .remove = __devexit_p(sht15_remove), > > > > -- Jean Delvare
