Re: [gentoo-user] gcc-4.0.0?
On Sunday 24 April 2005 13:10, Bastian Balthazar Bux wrote: Jerry McBride wrote: Anyone done anything big with the new compiler yet? Like an emerge -e world? This bug : http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21167 make some things difficult, anyway you can compile 95% of emerge -e but do it only in a thrashing ready environment. Why bother with GCC at all? It is a horrible compiler which produces slow code (and has many optimizations above -O2 broken). It's main advantages are that it is portable and free. If you are looking for a compiler that actually produces good binaries you are looking the wrong way IMO. ICC is a better compiler by far. Shame ICC support is only in a couple of packages and not more widespread. -- Nicolai P. Guba Sony Computer Entertainment Europe, http://www.scee.net my opinions are mine -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
On 4/25/05, Nicolai Guba [EMAIL PROTECTED] wrote: Why bother with GCC at all? It is a horrible compiler which produces slow code (and has many optimizations above -O2 broken). It's main advantages are that it is portable and free. If you are looking for a compiler that actually produces good binaries you are looking the wrong way IMO. ICC is a better compiler by far. Shame ICC support is only in a couple of packages and not more widespread. I wouldn't want to start a flamefest, but I've had a lot of experience with the ICC compiler over the last year working on the Battlefield 2 game engine (I used ICC under Linux) and I thought I'd add my two cents. There are a number of issues with using ICC on large C++ codebases. The most important ones I encountered are, in brief: - It sometimes generates invalid DWARF debugging sections, which crashes both GDB (all versions) and Intel's own debugger, idb. Not very helpful. - It sometimes generates broken assembly code (with duplicate assembly labels) that will not assemble, especially when you use floating point intrinsics (which are on by default). The only remedy is to move certain parts of expressions around. - It silently violates floating point IEEE requirements on the default optimization level (by using SSE single precision). This breaks many programs to the point where they see NaN values everywhere. I'm just trying to show that it's not all roses. For instance, I wouldn't expect something as large as openoffice or mozilla to compile and run successfully with ICC anytime soon unless major dev work goes into porting and babysitting those builds for ICC. On the positive side, ICC has excellent ANSI/ISO C++ support (it's based on a very good frontend which is also shared by Comaeu C++ IIRC), and when you care about speed more than correctness, it's very hard to beat. // Andreas -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0 ?
Nicolai Guba wrote: Why bother with GCC at all? It is a horrible compiler which produces slow code (and has many optimizations above -O2 broken). It's main advantages are that it is portable and free. If you are looking for a compiler that actually produces good binaries you are looking the wrong way IMO. ICC is a better compiler by far. Shame ICC support is only in a couple of packages and not more widespread. sarcasm There's a compiler that is even better than ICC. Microsoft's C/C++ V7 (.NET) compiler produces even smaller and faster programs and libraries than ICC. It is really a shame that more Linux packages don't support using the .NET compilers. /sarcasm Do you even realize that Linux and Gentoo runs on many more processors than Intel Pentiums? Besides, calling _any_ open source program horrible is offensive to me. Hundreds of people/companies have donated tens-of-thousands of hours to developing GCC, and have asked you for *nothing* in return. If you don't want to use GCC, that's fine. If you want to suggest that ICC can produce faster code for Intel processors, that's fine. Publicly bashing their work is unacceptable. -Richard -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
Jerry McBride wrote: Anyone done anything big with the new compiler yet? Like an emerge -e world? This bug : http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21167 make some things difficult, anyway you can compile 95% of emerge -e but do it only in a thrashing ready environment. -- No problem is so formidable that you can't walk away from it. ~ Charles M. Schulz But sometimes run fast is better ~ Francesco R. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
On Sunday 24 April 2005 01:50 am, Willie Wong wrote: last I checked, the gcc4 ebuild isn't even hard-masked yet... it exists, but that's about it. Don't know if the gentoo-user list is the best place to ask this question at this moment. Thanks to you and Bastian, I'm coming up to speed on this stuff... OTOH, searching around on the gentoo forums, it seems that some people are adventurous enough. Some problems they reported: glibc 2.3.5 does not adhere to gcc4's strictness wget won't compile openssh won't compile e2fsprogs won't compile ...and more Yeah, noticed that. I guess this is on hold for systems that have to run... There's one gigantic gcc 4.0 thread in the forums, it began life as gcc 3.4.0 thread, however, go a bit past half way before you hit the 4.0 stuff. Thanks, I'll dig it up. to quote irf2003 whom many on the thread agreed as the one to listen to: Don't mess ur box up! You are on stable, with gcc4 you need to go beyond ~x86. Play with it in a chroot. The only advantage of gcc4 at this time is that it compiles fast. If you want a real boost to your system, go for gcc-3.4.x, you will not regret it. It's literally like having a hardware upgrade, when one is migrating from gcc 3.3. Give it a try. Is gcc-3.4.3-20050110-r2 the favored release of 3.4.3? HTH, W It did, thanks guys. -- ** Registered Linux User Number 185956 FSF Associate Member number 2340 since 05/20/2004 Join me in chat at #linux-users on irc.freenode.net Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00! 11:32am up 15 days, 18:39, 3 users, load average: 0.00, 0.00, 0.00 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
On Sun, Apr 24, 2005 at 11:30:45AM -0400, Jerry McBride wrote: Is gcc-3.4.3-20050110-r2 the favored release of 3.4.3? Don't know about favored. It's the one I use =) So far it hasn't failed me... W -- * Address: 45 Spelman Hall, Princeton University 08544 * * Phone: x68958 AIM: AngularJerk* *E-mail: [EMAIL PROTECTED]From: sep.dynalias.net * This is not an optical illusion. It just looks like one. Sortir en Pantoufles: up 13 days, 2:51 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
On Sunday 24 April 2005 12:52 pm, Willie Wong wrote: On Sun, Apr 24, 2005 at 11:30:45AM -0400, Jerry McBride wrote: Is gcc-3.4.3-20050110-r2 the favored release of 3.4.3? Don't know about favored. It's the one I use =) So far it hasn't failed me... It hasn't failed you? You mean everything has compiled without problems? -- ** Registered Linux User Number 185956 FSF Associate Member number 2340 since 05/20/2004 Join me in chat at #linux-users on irc.freenode.net Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00! 2:16pm up 15 days, 21:23, 3 users, load average: 0.00, 0.00, 0.00 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
Well... modulo tweaking USE flags to disable some things and changing my CFLAGS, everything I compiled runs like expected, and everything I wanted compiled. Can't vouch for all the packages in portage though (= Best W On Sun, Apr 24, 2005 at 02:10:41PM -0400, Jerry McBride wrote: On Sunday 24 April 2005 12:52 pm, Willie Wong wrote: On Sun, Apr 24, 2005 at 11:30:45AM -0400, Jerry McBride wrote: Is gcc-3.4.3-20050110-r2 the favored release of 3.4.3? Don't know about favored. It's the one I use =) So far it hasn't failed me... It hasn't failed you? You mean everything has compiled without problems? -- * Address: 45 Spelman Hall, Princeton University 08544 * * Phone: x68958 AIM: AngularJerk* *E-mail: [EMAIL PROTECTED]From: sep.dynalias.net * Dude, this is making the same approximation twice in a row. It's like a whack-a-mole game. ~DeathMech, Some Student. P-town PHY 205 Sortir en Pantoufles: up 13 days, 5:09 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc-4.0.0?
last I checked, the gcc4 ebuild isn't even hard-masked yet... it exists, but that's about it. Don't know if the gentoo-user list is the best place to ask this question at this moment. OTOH, searching around on the gentoo forums, it seems that some people are adventurous enough. Some problems they reported: glibc 2.3.5 does not adhere to gcc4's strictness wget won't compile openssh won't compile e2fsprogs won't compile ...and more There's one gigantic gcc 4.0 thread in the forums, it began life as gcc 3.4.0 thread, however, go a bit past half way before you hit the 4.0 stuff. to quote irf2003 whom many on the thread agreed as the one to listen to: Don't mess ur box up! You are on stable, with gcc4 you need to go beyond ~x86. Play with it in a chroot. The only advantage of gcc4 at this time is that it compiles fast. If you want a real boost to your system, go for gcc-3.4.x, you will not regret it. It's literally like having a hardware upgrade, when one is migrating from gcc 3.3. Give it a try. HTH, W On Sun, Apr 24, 2005 at 12:57:30AM -0400, Jerry McBride wrote: Anyone done anything big with the new compiler yet? Like an emerge -e world? -- ** Registered Linux User Number 185956 FSF Associate Member number 2340 since 05/20/2004 Join me in chat at #linux-users on irc.freenode.net Buy an Xbox for $149.00, run linux on it and Microsoft loses $150.00! 1:03am up 15 days, 8:10, 3 users, load average: 0.00, 0.00, 0.00 -- gentoo-user@gentoo.org mailing list -- * Address: 45 Spelman Hall, Princeton University 08544 * * Phone: x68958 AIM: AngularJerk* *E-mail: [EMAIL PROTECTED]From: sep.dynalias.net * The longest word in the English language is the one that follows And now a word from out sponsor. ~Jack Benny Sortir en Pantoufles: up 12 days, 15:40 -- gentoo-user@gentoo.org mailing list
