gcc -O flags (was GLIBC Coredump)

Shawn Starr spstarr at visualnet.rootpass.net
Fri Jul 14 16:32:27 PDT 2000


That's what I'm planning on doing, leaving the default without -g since i
dont want debugging :)

On Fri, 14 Jul 2000, Thomas T. Veldhouse wrote:

> Do you honestly think you are going to notice the performance difference?  I
> would stick with what is known to work well.
> 
> Tom Veldhouse
> veldy at veldy.net
> 
> Mattias Axelsson wrote:
> 
> > Is the advice from FreeBSD to _never_ use -O2 or -O3 on intel machines?
> > >From what I can understand, -O3 among other things makes loop unrolling
> > and function inlining, which I thought were safe performance boosters.
> >
> > glibc seems to fail compiling with -O3 (I'm not ready with -O2 yet) but
> > how is it with other packages? Are there any more packages that will fail
> > to compile, or worse, compile but fail to work, when using -O2 or -O3 then
> > I would be glad to hear about it. Or should I just give up and use just -O
> > all the time?
> >
> > /Mattias
> >
> > "Thomas T. Veldhouse" wrote:
> > >
> > > I use:  -O -pipe -march=pentiumpro
> > >
> > > That is it. I use it for both CFLAGS and CXXFLAGS.
> > >
> > > I am going to rebuild all my C++ packages with
> > >
> > > CXXFLAGS= -O -pipe -march=pentiumpro -fmemoize-lookups -fsave-memoized
> > >
> > > -O2 and above is known to cause some problems in rare cases on i386 - and
> > > many cases on alpha machines.  I did run into a problem with -O2 once (for
> > > gcc-2.95.2), but I don't recall what package it was (it was before I started
> > > building my own system).
> > >
> > > Here is a quote from the /etc/make.conf in FreeBSD 4.0-STABLE.
> > >
> > > # CFLAGS controls the compiler settings used when compiling C code.
> > > # Note that optimisation settings above -O (-O2, ...) are not recommended
> > > # or supported for compiling the world or the kernel - please revert any
> > > # nonstandard optimisation settings to "-O" before submitting bug reports
> > > # to the developers.
> > > # Note also that at this time the -O2 setting is known to produce BROKEN
> > > # CODE on the Alpha platform.
> > >
> > ________________________________________________________________________
> > Get Your Private, Free Fortune from Barbapapa at http://www.barbanet.com
> > --
> > Mail archive: http://www.pcrdallas.com/mail-archives/lfs-discuss
> > IRC access: server: irc.linuxfromscratch.org port: 6667 channel: #LFS
> > Unsubscribe: email lfs-discuss-request at linuxfromscratch.org and put
> > "unsubscribe" (without the quotation marks) in the body of the message
> > (no subject is required)
> 
> --
> Mail archive: http://www.pcrdallas.com/mail-archives/lfs-discuss
> IRC access: server: irc.linuxfromscratch.org port: 6667 channel: #LFS
> Unsubscribe: email lfs-discuss-request at linuxfromscratch.org and put
> "unsubscribe" (without the quotation marks) in the body of the message
> (no subject is required)
> 

--
Mail archive: http://www.pcrdallas.com/mail-archives/lfs-discuss
IRC access: server: irc.linuxfromscratch.org port: 6667 channel: #LFS
Unsubscribe: email lfs-discuss-request at linuxfromscratch.org and put
"unsubscribe" (without the quotation marks) in the body of the message
(no subject is required)



More information about the lfs-dev mailing list