Re: [PATCH v7 1/4] spinlock: A new lockref structure for locklessupdate of refcount

[Linus Torvalds]

On Tue, Sep 3, 2013 at 2:34 PM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> I'll try to hack that up too, but it's looking like it really is just
> the "lock xadd", not the memory dependency chain..

Yeah, no difference: Better code generation with my quick hack for a
percpu spinlock:

       â    ffffffff81078e70 <lg_local_lock>:
  0.59 â      push   %rbp
  0.25 â      mov    %rsp,%rbp
  0.07 â      mov    $0x100,%eax
 97.55 â      lock   xadd   %ax,%gs:(%rdi)
  0.01 â      movzbl %ah,%edx
       â      cmp    %al,%dl
  0.68 â    â je     29
       â      nop
       â20:   pause
       â      mov    %gs:(%rdi),%al
       â      cmp    %dl,%al
       â    â jne    20
       â29:   pop    %rbp
  0.84 â    â retq

but the actual cost is pretty much the same:

   6.81%   lg_local_lock

so it doesn't seem to be some odd weakness of the microarchitecture.

            Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/