Re: Please requeue atlas3

Re: Please requeue atlas3

[Matthew Wilcox]

On Wed, Jun 09, 2004 at 10:05:15AM -0600, LaMont Jones wrote:
> On Wed, Jun 09, 2004 at 11:31:38AM -0400, Camm Maguire wrote:
> > TO close the book on ia64 atlas3, are there any isa extensions
> > available on certain subclasses of the cpu which might warrant a
> > separate build, as is the case for sse2 on i386?
> 
> Absolutely no clue here.  Anyone else?

I think the only ISA addition so far is BranchLong which gets emulated
by the kernel.  I hear murmurings of additional instructions in future
processors, but let's cross that bridge when there's silicon to be had.

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain

Re: Please requeue atlas3

[David Mosberger]

>>>>> On Wed, 9 Jun 2004 17:51:03 +0100, Matthew Wilcox <willy@debian.org> said:

  Matthew> I think the only ISA addition so far is BranchLong which
  Matthew> gets emulated by the kernel.  I hear murmurings of
  Matthew> additional instructions in future processors, but let's
  Matthew> cross that bridge when there's silicon to be had.

There is also the 16-byte-atomic-ops extension.  The Ski simulator has
been supporting that for some time, so in that sense it exists
already:

 $ cat /proc/cpuinfo | grep features
 features   : branchlong, 16-byte atomic ops

;-)

	--david