This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
crosstool-0.28-rc26: 118 ugly toolchains and 74 pretty ones
- From: Dan Kegel <dank at kegel dot com>
- To: GCC Mailing List <gcc at gcc dot gnu dot org>
- Date: Tue, 29 Jun 2004 23:14:21 -0700
- Subject: crosstool-0.28-rc26: 118 ugly toolchains and 74 pretty ones
Crosstool is a portable script to build gcc/glibc crosstoolchains.
You can download crosstool snapshots from
http://kegel.com/crosstool/
Version 0.28-rc26 is ready for testing.
If you've sent me a fix that isn't included, please resend it.
Changelog:
0.28-rc26:
uClibc:
* added contrib/demo-uclibc.sh, contrib/crosstool-uclibc-0.28-rc5-ter.patch
for those who really want uclibc support
crosstool.sh:
* added 'sleep 2' before touching errlist-compat.c in case it's on
a filesystem with lousy time resolution (like FAT). Thanks
to jdubois@mc.com for the suggestion.
glibc-2.2.5:
* fixed typo in patches/glibc-2.2.5/dl-machine-mips.patch
To see which combinations of gcc and glibc versions build targeting which CPUs,
I ran a script (demo-loop.sh) which builds the 192 combinations
(gcc-{2.95-3,3.3.3,3.4.0} vs. glibc-{2.1.3,2.2.5,2.3.2} vs. 24 cpus).
On my Athlon XP 2GHz machine ($160 at Fry's for motherboard, RAM, and CPU),
this takes about five days.
It only checks whether the C/C++ toolchain itself builds,
not whether the gcc or glibc regression tests pass, as I don't have
access to most of these target CPUs.
See http://www.kegel.com/crosstool/crosstool-0.28-rc26/buildlogs/0.28/
for the HTML version of this table, and links to the build log tails.
For instance,
http://www.kegel.com/crosstool/crosstool-0.28-rc26/buildlogs/0.28/cris-gcc-3.4.0-glibc-2.2.5.txt
shows why the combination cris-gcc-3.4.0-glibc-2.2.5 failed
(hint: it's a bit ICEy out there in Cris land).
(Note: all tools are minimally patched, except for gcc-2.95.3 and
glibc-2.1.3, which have the Red Hat 6.2 patches.)
Tested combinations:
A: gcc-2.95.3 glibc-2.1.3 binutils-2.13.90.0.2 linux-2.4.21
B: gcc-2.95.3 glibc-2.2.5 binutils-2.11.2 linux-2.4.3
C: gcc-3.3.3 glibc-2.1.3 binutils-2.15 linux-2.4.24
D: gcc-3.3.3 glibc-2.2.5 binutils-2.15 linux-2.4.21
E: gcc-3.3.3 glibc-2.3.2 binutils-2.15 linux-2.4.24
F: gcc-3.4.0 glibc-2.1.3 binutils-2.15 linux-2.4.21
G: gcc-3.4.0 glibc-2.2.5 binutils-2.15 linux-2.4.21
H: gcc-3.4.0 glibc-2.3.2 binutils-2.15 linux-2.6.6
Results:
A B C D E F G H
alpha FAIL pass FAIL FAIL pass FAIL FAIL pass
arm9tdmi FAIL FAIL FAIL pass pass FAIL pass pass
arm pass pass FAIL pass pass FAIL pass pass
arm-iwmmxt FAIL FAIL FAIL FAIL FAIL FAIL pass pass
arm-softfloat FAIL FAIL FAIL pass pass FAIL FAIL FAIL
cris FAIL FAIL FAIL FAIL FAIL FAIL FAIL FAIL
hppa FAIL FAIL FAIL FAIL FAIL FAIL FAIL FAIL
i686 pass pass pass pass pass pass pass pass
ia64 FAIL FAIL FAIL FAIL pass FAIL FAIL pass
m68k pass FAIL FAIL FAIL pass FAIL FAIL pass
mipsel FAIL FAIL FAIL FAIL pass FAIL FAIL pass
powerpc-405 FAIL FAIL FAIL pass pass FAIL pass pass
powerpc-440 FAIL FAIL FAIL FAIL FAIL FAIL pass pass
powerpc-604 pass pass FAIL pass pass FAIL pass pass
powerpc-7450 FAIL FAIL FAIL FAIL FAIL FAIL FAIL pass
powerpc-750 pass pass FAIL pass pass FAIL pass pass
powerpc-860 FAIL pass FAIL pass pass FAIL pass pass
powerpc-970 FAIL FAIL FAIL FAIL FAIL FAIL FAIL pass
s390 FAIL FAIL FAIL FAIL pass FAIL FAIL pass
sh3 FAIL FAIL FAIL pass pass FAIL pass pass
sh4 FAIL FAIL FAIL pass pass FAIL pass pass
sparc64 FAIL FAIL FAIL FAIL FAIL FAIL FAIL pass
sparc FAIL pass FAIL pass pass FAIL FAIL pass
x86_64 FAIL FAIL FAIL FAIL pass FAIL FAIL pass
hppa support is not expected to work until glibc-2.3.3 or so, which
I'll test as soon as the glibc team decides to make a tarball :-)
Patches to fix any of the FAILs gratefully accepted. If I'm just
missing a patch from CVS, please just send me a link to the
patch in cvsweb. Thanks!
- Dan
--
My technical stuff: http://kegel.com
My politics: see http://www.misleader.org for examples of why I'm for regime change