Re: write_unlock: replace clear_bit with byte store

[David Mosberger <davidm@napali.hpl.hp.com>]

>>>>> On Fri, 29 Apr 2005 08:46:14 -0700 (PDT), Christoph Lameter <clameter@engr.sgi.com> said:

  Christoph> On Fri, 29 Apr 2005, David Mosberger wrote:
  >> >>>>> On Fri, 29 Apr 2005 08:30:27 -0700 (PDT), Christoph Lameter <clameter@engr.sgi.com> said:

  Christoph> So is my version of inline asm int the latest version
  Christoph> from yesterday.  There is an alternate version included
  Christoph> for !ASM_SUPPORTED in C which does not use an nta store.

  >> Yes, but _if_ it's a good idea to use .nta with GCC, there is no
  >> reason not to do the same with ICC.  Don't introduce unnecessary
  >> divergence.

  Christoph> The same situation of .nta only for GCC already exists
  Christoph> for regular spinlocks as a result of my nta unlock patch
  Christoph> that I posted a week or so ago.

And that's an argument to make the situation worse?  How about
cleaning up the previous patch instead?

	--david