Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!hplabs!hp-pcd!hplsla!jima
From: jima@hplsla.HP.COM (Jim Adcock)
Newsgroups: comp.lang.c++
Subject: Re: Having constructors is expensive.
Message-ID: <6590137@hplsla.HP.COM>
Date: 1 Jun 89 22:24:06 GMT
References: <11615@ulysses.homer.nj.att.com>
Organization: HP Lake Stevens, WA
Lines: 41

> jeffb@grace.cs.washington.edu (Jeff Bowden) writes:
> >
> >	The problem (which I found by examining the output of CC -F) is that
> >the empty constructor (which is defined in the class declaration, so it is
> >should to be inline) somehow gets incarnated and called as a normal function!
> >
> >	This wouldn't be so bad except that I declare many arrays of vectors
> >of this class and I generally use only 10% of the elements.  (I have an empty
> >constructor because of the vectors; fascist cfront doesn't like it when you
> >declare vectors of things which haven't any no-argument constructor).
> 
> The reason C++ demands a no argument constructor is so that it can
> maintain what I call the "allocation guarantee".  Namely every area
> of storage that it will call an X must be initialized by a
> constructor.  In the case of something like
> 
> 	X array[1000]  
> 
> this means that the X() constructor must be called 1000 times to
> "initialize" the areas that are holding an X.  This is done by a
> function that is passed the constructor as an argument.  This is why
> The X() constructor is being incarnated, and presumably it is also
> account for the profiling showing many calls to X().

One might hope for a compiler that inlines code for array initialization
if a class's X() constructor is declared inline.  I see nothing that
requires a compiler to use a vector new function expecting a pointer
to an X() constructor.  For example, why couldn't a compiler pass a null
function pointer to the vector new function, test in the vector new function
whether or not to call that function, and then generate an inline for-loop
with the inline X() constuctor to do the actual init?  Good optimizing
backend C "code generators" would be smart enough to void the loop if the
X() constructor is null.

Its ugly to have to come up with hack work-arounds for compiler deficiencies.  
Compilers need to do a much better job of respecting a user's request for 
inlining -- resolving inline verses non-inline compiler deficiencies 
presently takes me more time than all other aspects of C++ code writing!

One might question how usable present compilers are for arrays of objects.
I'm always having to find hack work-arounds.  Is this what C++ is about ???