Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site pur-ee.UUCP
Path: utzoo!watmath!clyde!floyd!harpo!ihnp4!inuxc!pur-ee!ecn-ee!davy
From: davy@ecn-ee.UUCP
Newsgroups: net.lang.c
Subject: Re: Re: structure and array and string c - (nf)
Message-ID: <1745@pur-ee.UUCP>
Date: Tue, 27-Mar-84 17:20:55 EST
Article-I.D.: pur-ee.1745
Posted: Tue Mar 27 17:20:55 1984
Date-Received: Wed, 28-Mar-84 07:18:22 EST
Sender: notes@pur-ee.UUCP
Organization: Electrical Engineering Department , Purdue University
Lines: 25

#R:sdchema:-109700:ecn-ee:13100010:000:542
ecn-ee!davy    Mar 27 08:25:00 1984

In-line assembler functions are a nifty idea, and probably wouldn't
be that bad to implement (maybe they already have been?).  I've 
never seen the Berkeley sed scripts you describe which do this stuff,
but I have seen:


	foo(x, y, z)
	{
	#ifdef vax
		asm("assembly language version of the routine");
	#else
		C version of the routine
	#endif
	}

Seems to me this oughta be good enough for most people....it's kinda
ugly, but how often do you really want to go down into assembly language
for more than one or two statements?

--Dave Curry