Re: C++ in kernel (was Re: exception in a device driver)
kwrohrer@ce.mediaone.net
Tue, 12 Jan 1999 00:38:00 -0600 (EST)
And lo, Khimenko Victor saith unto me:
> >>
> >>foo1.C
> >>
> >> blah->foofunc();
> >>
> >>
> >>foo2.C
> >>
> >>blahobject::foofunc()
> >>{
> >> printf("Hi");
> >>}
> >>
> >>a C++ compiler ends up passing "this" needlessly because it can't tell that
> >>foofunc of object blahobject need not know who it is.
>
> > A good optimizing compiler should detect the lack of use of the this
> > variable and optimize it away. Besides the reason "this" is passed to the
> > function is to access object variables, so a C version of a C++ object would
> > have its associated functions passed a structure's address as well.
>
> Since C++ has no support for modular programming (shame for modern language)
> aven genious compiler could not do anything with this sample. Since compiler
> is NOT aware about lack of this usage in foofunc while compiling foo1.C and
> thus could not remove it's usage there and thus could not remove it's usage
> in foo2.C as well even if foofunc is not virtual ...
A correct compiler could not skip passing this, because it can not be known
until link time that this will not be needed. I don't know of a compiler
that leaves enough info in the .o file for the linker to optimize at that
stage...and ld.so would never take the time to do so.
> Programmer should do this
> by hand by declaring foofunc as static :-)
Amen...and it's not like you have occasion to do so all that often.
Keith
--
"Well, look at that. The sun's | Linux: http://www.linuxhq.com |"Zooty,
coming up." -- John Sheridan, | KDE: http://www.kde.org | zoot
"Sleeping in Light", Babylon 5 | Keith: kwrohrer@enteract.com | zoot!"
www.midwinter.com/lurk/lurker.html | http://www.enteract.com/~kwrohrer | --Rebo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/