Messages in this thread | | | Date | Thu, 13 Nov 2008 13:35:20 -0500 (EST) | From | Steven Rostedt <> | Subject | Re: [PATCH 1/2] tracing/function-return-tracer: Make the function return tracer lockless |
| |
On Thu, 13 Nov 2008, Andi Kleen wrote:
> On Thu, Nov 13, 2008 at 12:32:23PM -0500, Steven Rostedt wrote: > > > > On Thu, 13 Nov 2008, Andi Kleen wrote: > > > > > > So the answer to this is: > > > > > > > > i = index++; > > > > barrier(); > > > > write to index i (not index); > > > > > > That was my first thought when I wrote the original email, > > > but the disadvantage is that barrier() is a big hammer > > > that flushes everything and can make the code much worse. > > > That is why I suggested local_add_return() instead. > > > > barrier() is a compiler barrier, does nothing with the caches, and is > > quite cheap. We only need a compiler barrier because we are only > > I did not refer to CPU caches, but the compiler's register allocation > [ok if you want the registers are the "level 0 cache"]. A memory barrier > all messes it up. That is why it is better to only clobber specific > memory regions, which is what local_* does. >
#define barrier() __asm__ __volatile__("": : :"memory")
static inline long local_add_return(long i, local_t *l) { long __i; #ifdef CONFIG_M386 unsigned long flags; if (unlikely(boot_cpu_data.x86 <= 3)) goto no_xadd; #endif /* Modern 486+ processor */ __i = i; asm volatile(_ASM_XADD "%0, %1;" : "+r" (i), "+m" (l->a.counter) : : "memory"); return i + __i;
#ifdef CONFIG_M386 no_xadd: /* Legacy 386 processor */ local_irq_save(flags); __i = local_read(l); local_set(l, i + __i); local_irq_restore(flags); return i + __i; #endif }
Now tell me again how local_* is more efficient than barrier?
Not to mention, if this is ever used on other archs with load-linked and store-conditional, it gets even worse.
-- Steve
| |