On Tue, Jun 05, 2018 at 02:00:22PM +0200, Peter Zijlstra wrote: > On Tue, May 29, 2018 at 07:07:40PM +0100, Mark Rutland wrote: > > In ldsem_cmpxchg a pointer to unsigned long is passed to > > atomic_long_cmpxchg(), which expects a pointer to atomic_long_t. > > > - long tmp = atomic_long_cmpxchg(&sem->count, *old, new); > > + long tmp = atomic_long_cmpxchg((atomic_long_t *)&sem->count, *old, new); > > Needing to cast anything to atomic types is dodgy at best, so I had me a > look at this code. > > Would not the following patch be much better? > > --- > drivers/tty/tty_ldsem.c | 82 > ++++++++++++++++++++--------------------------- > include/linux/tty_ldisc.h | 4 +-- > 2 files changed, 37 insertions(+), 49 deletions(-)
At first glance, yes, this does look a lot better, it's less lines at the least :) I think Peter Hurley was just trying to hid the atomic mess behind function calls, which is why he didn't open-code it like you did here. But this makes it a bit more obvious as to what "magic" is happening, so I like it. Care to turn it into a real patch so I can queue it up after 4.18-rc1 is out? Or do you want me to do that? thanks, greg k-h
