Newsgroups: comp.lang.c
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: Log Library - How is it done in the library code?
Message-ID: <1991Mar21.170830.24983@zoo.toronto.edu>
Date: Thu, 21 Mar 1991 17:08:30 GMT
References: <1991Mar20.173249.3819@zoo.toronto.edu> <1991Mar20.204034.28931@bingvaxu.cc.binghamton.edu> <1991Mar20.223644.16769@zoo.toronto.edu> <1991Mar21.000655.28999@bingvaxu.cc.binghamton.edu>
Organization: U of Toronto Zoology

In article <1991Mar21.000655.28999@bingvaxu.cc.binghamton.edu> kym@bingvaxu.cc.binghamton.edu (R. Kym Horsell) writes:
>>... There are *two* separate facilities that
>>are relevant, -f68881 and the inlining facility.  They're not the same thing.
>
>Yes I may be confused. The following comes from the man page ...
>
>-f68881   Generate  in-line  code   for   Motorola
>		 MC68881  floating-point  processor ...

However, this doesn't generate in-line code for library functions, only for
things like addition and multiplication.  There is a separate inlining
facility for library code, which is what's wanted if you're trying to
evaluate the performance of log().  See the discussion of the .il suffix.
-- 
"[Some people] positively *wish* to     | Henry Spencer @ U of Toronto Zoology
believe ill of the modern world."-R.Peto|  henry@zoo.toronto.edu  utzoo!henry