apple's plans for llvm back

(L) [2007/10/29] [lycium] [apple's plans for llvm] Wayback!

while reading across ars technica's review of os x leopard, i came across this interesting news (from [LINK http://arstechnica.com/reviews/os/mac-os-x-10-5.ars/11]):


Don't be misled by its humble use in Leopard; Apple has grand plans for LLVM. How grand? How about swapping out the guts of the gcc compiler Mac OS X uses now and replacing them with the LLVM equivalents? That project is well underway. Not ambitious enough? How about ditching gcc entirely, replacing it with a completely new LLVM-based (but gcc-compatible) compiler system? That project is called Clang, and it's already yielded some impressive performance results.


it seems they've hired the lead developer and are going hardcore with it. here's the link to clang: [LINK http://clang.llvm.org/]

(L) [2007/11/05] [tbp] [apple's plans for llvm] Wayback!

Lattner has been working for Apple for some time now; about a year ago, when a gcc Link Time Optimization backend was yet again revived, LLVM was proposed but put aside after much discussion (one argument being it wasn't as mature as advertised). As an outsider not caring much for political fuss and seeing how antiquated is gcc's backend i would have preferred to see more synergy.

Not that LLVM has lived up to expectations yet... i've tried about each and every release on real code and got nothing but ICEs or crappy perf when lucky.


Note: It used to be a real bitch to build with a gcc frontend, it's slightly simpler now. Give it a try.
_________________
May you live in interesting times.

[LINK https://gna.org/projects/radius/ radius] | [LINK http://ompf.org/ ompf] | [LINK http://ompf.org/wiki/ WompfKi]

(L) [2007/11/05] [lycium] [apple's plans for llvm] Wayback!

good to see you around again, oh benevolent dictator ;)


i will indeed have to check out llvm after my uni exams are over (soonsoon). my primary interest is using it as a shader compiler, it's no tight coherent-shading monster (maybe it can be made into one) but it should do the job better than my previous "infix-to-postfix stack-evaluation of simple maths functions" plans :)


in any case, i found this moving and shaking in the free compiler world quite interesting. i know competition in the commercial environment is good, but hmm perhaps co-operation is better in complex opensource projects like compilers where the really elite skills are at a premium.

(L) [2007/11/05] [tbp] [apple's plans for llvm] Wayback!

LLVM would be the right tool for the job indeed, if only because gcc is prohibited  - historically to avoid vendors to hack around GPL - from providing stable standardized external hooks into its subsystems; it's no surprise that LLVM, mainly sponsored by Apple, takes the other stance.

But, specifically, you can't talk about competition in this case because gcc has no LTO backend at all and won't for some time [SMILEY Smile]
_________________
May you live in interesting times.

[LINK https://gna.org/projects/radius/ radius] | [LINK http://ompf.org/ ompf] | [LINK http://ompf.org/wiki/ WompfKi]