Re: Are 64 Int or FP registers useful?

lfm@pgroup.com (Larry Meadows)
Sat, 19 Sep 1992 14:39:54 GMT

          From comp.compilers

Related articles
Re: Are 64 Int or FP registers useful? hrubin@pop.stat.purdue.edu (1992-09-14)
Re: Are 64 Int or FP registers useful? preston@helena.cs.rice.edu (1992-09-16)
Re: Are 64 Int or FP registers useful? hrubin@pop.stat.purdue.edu (1992-09-17)
Re: Are 64 Int or FP registers useful? daveg@synaptics.com (Dave Gillespie) (1992-09-17)
Re: Are 64 Int or FP registers useful? pardo@cs.washington.edu (1992-09-17)
Re: Are 64 Int or FP registers useful? lfm@pgroup.com (1992-09-19)
| List of all articles for this month |

Newsgroups: comp.compilers
From: lfm@pgroup.com (Larry Meadows)
Organization: The Portland Group, Portland, OR
Date: Sat, 19 Sep 1992 14:39:54 GMT
Keywords: registers, optimize
References: <1992Sep7.091904.2626@newsroom.bsc.no> 92-09-107

Dave Gillespie <daveg@synaptics.com> writes:
>On the other hand, I've never seen a compiler that could really take
>advantage of the 860's pipelining; our inner loops are all hand-coded.
>860 optimizers tend to be essentially vectorizing compilers, where the
>"vector instructions" are hand-coded assembly language routines.


Actually, our i860 compiler generates software pipelined code for the inner
loop, unless we happen to have a hand-code routine that matches the
inner loop. So, for example, a vector add will be matched, but some
complicated collection of adds and multiplies will be software
pipelined. On many codes it works quite well, although I'd be the first
to admit that hand code usually does better. A lot of the problem
is in communicating dependence information from the user to the compiler.
--
Larry Meadows The Portland Group
lfm@pgroup.com
--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.