[Linux-ia64] Re: ia64 cache flushing?

From: David Mosberger <davidm_at_napali.hpl.hp.com>
Date: 2002-12-17 13:15:28
>>>>> On Tue, 17 Dec 2002 13:11:46 +1100, Rusty Russell <rusty@rustcorp.com.au> said:

  Rusty> if (owner == me) should do this.

  Rusty> On top of your patch and my previous, this works!

Good to see that.

For someone who cares about modules and ia64 performance: you might
want to consider to use "brl" in the PLT stub instead of the
standardized code sequence.  That would avoid an indirect branch and
associated mis-prediction penalties.  (Of course, this would be for
McKinley and later CPUs only as Merced doesn't implement "brl" in
hardware).

	--david
Received on Mon Dec 16 18:17:12 2002

This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:11 EST