Re: Free UNIX for non-commerical use.

From: mike (mmilewski_at_comcast.net)
Date: 07/29/03 

Date: Mon, 28 Jul 2003 20:58:26 -0400

> > > Don't agree: the optimization of the GCC has become much better:
> > > in 9 out of 10 times the GCC produces faster binaries than vendor
> compilers
> > > Perhaps not, if you will tune every single object, and put a lot of
> #pragma
> > > in your source to keep the vendor compilers from producing invalid code.
> > > On AIX IBM's xlc build faster common-mode bins, but if you mind peak
> > > performance, you should try the GCC 3.x and the cpu-specific
> optimizations.
> > > Never got working binaries from any GNU code with xlc and "-O5".
> > > GCC's results were up to 10% faster on PowerPC, with greater difference
> for
> > > float compared to integer.
> > > Same for Solaris x86, SCO and NCR's MP-RAS.
> > > A great time-saver of the GCC is the fix of bugs in the O/S include
> headers.
> >
> > I have never yet seen gcc 3.2 or previous generate faster code than
> > Sun's C compiler. I've seen improvements up to 10* by switching to
> > Sun's cc, depending on the code. Thats with specific cpu/architecture
> > selection and high optimization selected for both gcc and cc,
> > compiling plain C code.
> >
> Don't have written that this is valid for Solaris on Sparc, but I've never
> seen a factor of ten on the E450 I've compared GCC 2.95.2 and Sun's Workshop
> 4 a few years ago.
> It is maybe??? possible to write code leading to such extreme results, but
> you could never expect such factor in real life. Not even with the code for
> SPECint and SPECfp which no commercial vendor will ignore.
> If you can build the Apache/SSL/PHP4 running 10times as fast as the GCC
> build I will do the packaging for you.
> But Sun will make the next version slower, since no one buys new hw anymore
> ;-)

Regarding the GCC issue, I would like to state that I have seen a performance
increase with SGI's MIPSpro compilers for IRIX - most noteworthy example would
be a xmame build I did once and it performed much better with MIPSpro. It has
also been my experience that sometimes it just depends on the program itself.

Relevant Pages

  • Re: Which PIC18 C Compiler?
    ... Walter gave a good answer regarding recursion in 8051 compilers, totally unrelated to gcc and/or sdcc (since there is no gcc port for the 8051). ... But in my experience on ColdFire's and PPC's, it has generated similar to or better than the few commercial compilers I have seen. ...
    (comp.arch.embedded)
  • Re: math.nroot [was Re: A brief question.]
    ... >>> Ah, but as I've said before, virtually all C compilers on 754 boxes ... >>> happy to use gcc, they all could have used the older gcc spellings ... > use glibc, they all could have used the older glibc spellings. ... for setting traps) interfaces, and I can't be arsed to go through old ...
    (comp.lang.python)
  • Re: math.nroot [was Re: A brief question.]
    ... >> happy to use gcc, they all could have used the older gcc spellings ... parings of C compilers and C runtime libraries are ... use glibc, they all could have used the older glibc spellings. ... trap doesn't actually fire until the _next_ fp operation ...
    (comp.lang.python)
  • Re: What is Linux built with? (or, what does this question mean?)
    ... Isn't the kernel/distro nearly always built w/ gcc? ... Now some of us actually have other compilers as well and these could be ... specific to provide cache and other optimizing that can't be generic. ... gnu compile in a way that it gives some registers over to the kernel as far ...
    (alt.os.linux.suse)
  • Re: g77 mex problem under win
    ... Don't have my matlab available, but as far as I can remember, ... Why do you insist on using gcc? ... You can buy good C compilers for a couple of hundred ... The same goes for me with the fortran compilers I have seen. ...
    (comp.soft-sys.matlab)