Re: [PATCHv4 01/12] atomic/tty: Fix up atomic abuse in ldsem

[Mark Rutland]

On Tue, Jul 24, 2018 at 11:20:36AM +0200, Peter Zijlstra wrote:
> On Tue, Jul 24, 2018 at 09:15:18AM +0200, Ingo Molnar wrote:
> > 
> > * Mark Rutland <mark.rutland@xxxxxxx> wrote:
> > 
> > > From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > > 
> > > Mark found ldsem_cmpxchg() needed an (atomic_long_t *) cast to keep
> > > working after making the atomic_long interface type safe.
> > > 
> > > Needing casts is bad form, which made me look at the code. There are no
> > > ld_semaphore::count users outside of these functions so there is no
> > > reason why it can not be an atomic_long_t in the first place, obviating
> > > the need for this cast.
> > > 
> > > That also ensures the loads use atomic_long_read(), which implies (at
> > > least) READ_ONCE() in order to guarantee single-copy-atomic loads.
> > > 
> > > When using atomic_long_try_cmpxchg() the ldsem_cmpxchg() wrapper gets
> > > very thin (the only difference is not changing *old on success, which
> > > most callers don't seem to care about).
> > > 
> > > So rework the whole thing to use atomic_long_t and its accessors
> > > directly.
> > > 
> > > While there, fixup all the horrible comment styles.
> > > 
> > > Cc: Peter Hurley <peter@xxxxxxxxxxxxxxxxxx>
> > > Acked-by: Will Deacon <will.deacon@xxxxxxx>
> > > Reported-by: Mark Rutland <mark.rutland@xxxxxxx>
> > > Reviewed-by: Andy Shevchenko <andy.shevchenko@xxxxxxxxx>
> > > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> > > Signed-off-by: Mark Rutland <mark.rutland@xxxxxxx>
> > > Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> > > ---
> > >  drivers/tty/tty_ldsem.c   | 82 ++++++++++++++++++++---------------------------
> > >  include/linux/tty_ldisc.h |  4 +--
> > >  2 files changed, 37 insertions(+), 49 deletions(-)
> > > 
> > > Note: Greg has queued this via the in the tty tree for v4.19, which can be seen at: 
> > > 
> > > https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git/commit/?h=tty-next&id=5fd691afdf929061c391d897fa627822c3b2fd5a
> > 
> > Can this patch be skipped, or do the others depend on it?
> 
> IIRC it depends on it, without this patch you get build issues due to
> atomic_long_cmpxchg() getting picky about it's arguments (type safety
> improved).

Yup. Without this patch, there will be a build regression at patch 9,
when we move to generated atomic_long_*() wrappers.

Thanks,
Mark.