Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1574447

Re: [PATCH 4/5] atomic: Introduce atomic_try_cmpxchg()

Show all headers | View raw

On Mon, Feb 06, 2017 at 12:24:28PM +0800, Boqun Feng wrote:
> On Fri, Feb 03, 2017 at 02:26:02PM +0100, Peter Zijlstra wrote:

> > 
> > 	for (;;) {
> > 		new = val $op $imm;
> > 		if (try_cmpxchg(ptr, &val, new))
> > 			break;
> > 	}
> > 
> > while also generating better code (GCC6 and onwards).
> > 
> 
> But switching to try_cmpxchg() will make @val a memory location, which
> could not be put in a register. And this will generate unnecessary
> memory accesses on archs having enough registers(PPC, e.g.).

GCC was perfectly capable of making @val a register in the code I was
looking at.

> > +#ifndef atomic_try_cmpxchg
> > +
> > +#define __atomic_try_cmpxchg(type, _p, _po, _n)				\
> > +({									\
> > +	typeof(_po) __po = (_po);					\
> > +	typeof(*(_po)) __o = *__po;					\
> > +	bool success = (atomic_cmpxchg##type((_p), __o, (_n)) == __o);	\
> > +	*__po = __o;							\
> 
> Besides, is this part correct? atomic_cmpxchg_*() wouldn't change the
> value of __o, so *__po wouldn't be changed.. IOW, in case of failure,
> *ptr wouldn't be updated to a new value.
> 
> Maybe this should be:
> 
> 	bool success;
> 	*__po = atomic_cmpxchg##type((_p), __o, (_n));
> 	sucess = (*__po == _o);
> 
> , right?

Yes, botched that. Don't think I even compiled it to be honest :/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 0/5] refcount_t and various related bits Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100
  [PATCH 4/5] atomic: Introduce atomic_try_cmpxchg() Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100
    Re: [PATCH 4/5] atomic: Introduce atomic_try_cmpxchg() Boqun Feng <boqun.feng@gmail.com> - 2017-02-06 05:30 +0100
      Re: [PATCH 4/5] atomic: Introduce atomic_try_cmpxchg() Boqun Feng <boqun.feng@gmail.com> - 2017-02-06 07:40 +0100
      Re: [PATCH 4/5] atomic: Introduce atomic_try_cmpxchg() Peter Zijlstra <peterz@infradead.org> - 2017-02-06 09:20 +0100
  [PATCH 2/5] kref: Implement using refcount_t Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100
    Re: [PATCH 2/5] kref: Implement using refcount_t Greg KH <gregkh@linuxfoundation.org> - 2017-02-06 14:10 +0100
  [PATCH 5/5] refcount: Use atomic_try_cmpxchg() Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100
  [PATCH 1/5] refcount_t: A special purpose refcount type Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100
    Re: [PATCH 1/5] refcount_t: A special purpose refcount type Kees Cook <keescook@chromium.org> - 2017-02-03 19:10 +0100
    Re: [PATCH 1/5] refcount_t: A special purpose refcount type Kees Cook <keescook@chromium.org> - 2017-02-04 00:40 +0100
  [PATCH 3/5] x86: Implement __WARN using UD2 Peter Zijlstra <peterz@infradead.org> - 2017-02-03 14:40 +0100

csiph-web