Not logged in
Log in now
Create an account
Subscribe to LWN
LWN.net Weekly Edition for May 23, 2013
An "enum" for Python 3
An unexpected perf feature
LWN.net Weekly Edition for May 16, 2013
A look at the PyPy 2.0 release
GNU sed 4.2.2 released; maintainer resigns
Posted Dec 23, 2012 15:23 UTC (Sun) by stevenb (guest, #11536)
Posted Dec 23, 2012 22:35 UTC (Sun) by lu_zero (guest, #72556)
Posted Dec 24, 2012 18:05 UTC (Mon) by heijo (guest, #88363)
This idea that you need to support bad compiler is bullshit, just use your x86-64 g++ to compile an x86-64 to $bootstrap g++, then use the latter to compile a native $bootstrap g++.
Posted Dec 26, 2012 10:49 UTC (Wed) by CopperWing (guest, #82856)
Posted Dec 26, 2012 10:40 UTC (Wed) by CopperWing (guest, #82856)
So you leave GCC because of its switch to C++, to embrace llvm and clang which are completely written in C++ as well? Does not make sense at all.
Posted Dec 26, 2012 12:11 UTC (Wed) by lu_zero (guest, #72556)
Posted Dec 26, 2012 13:40 UTC (Wed) by CopperWing (guest, #82856)
I build quite often the arm-linux-gnueabi GCC cross-compiler and a Linux root file system for my work, and after the switch to C++ I didn't find any issue in doing that. It's neither more nor less straightforward than before, just the same.
Conversely if you plan to rebuild the Linux kernel with llvm/clang, be prepared to get some headaches. Last time I checked, some components still needed to be recompiled with GCC and the resulting kernel was not at all stable.
So, for real work on Linux, GCC is still the only choice. Until someone volunteers to improve the situation (FreeBSD with clang didn't came exactly for free).
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds