Path: utzoo!utgpu!water!watmath!clyde!rutgers!cmcl2!nrl-cmf!ames!lll-lcc!pyramid!prls!mips!mash
From: mash@mips.COM (John Mashey)
Newsgroups: comp.arch
Subject: Re: FREQUENCY statements: fill a much-needed gap
Message-ID: <1385@winchester.mips.COM>
Date: 21 Jan 88 21:54:31 GMT
References: <839@ima.ISC.COM> <28200085@ccvaxa> <1267@winchester.UUCP> <2861@clash.rutgers.edu> <1722@faline.bellcore.com>
Reply-To: mash@winchester.UUCP (John Mashey)
Organization: MIPS Computer Systems, Sunnyvale, CA
Lines: 38

In article <1722@faline.bellcore.com> hammond@faline.UUCP (Rich A. Hammond) writes:

>In this case the programmer could be relatively certain of the
>frequency (or alternatively doesn't care if the error recovery is
>slower) and the compiler could put code, such as save all the registers
>before the error routine calls,  that would enable better code for the
>case that we wish to go quickly.
This is a specific instance of the general idea of moving register
saves and restores around when possible. Rich's case is:
	glurp()
	{
		if (something) {
			low-frequency-function-call
		} else {
			a few statements
		}
	}
You could move at least a return-address register-save/restore around
the function call (which our compiler folks some-times call "shrink-wrapping"
it.)  However, the idea really has nothing to do with function calls:
it could just as well be: 

	glurp()
	{
		if (something) {
			simple statment that needs few registers
		} else {
			a whole bunch of code that needs many regs
		}
	}

In this case, you move any save/restore code into the "else" part.
Has anybody actually implemented this? (we haven't yet)
-- 
-john mashey	DISCLAIMER: <generic disclaimer, I speak for me only, etc>
UUCP: 	{ames,decwrl,prls,pyramid}!mips!mash  OR  mash@mips.com
DDD:  	408-991-0253 or 408-720-1700, x253
USPS: 	MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086