Hi, On May 6, 2010, at 5:02 PM, Jan Rovins wrote:
> Rene Rebe wrote: >> That is really unfortunate. Hope they can find something. >> >> In any case, we can selectively use older versions of GCC for some >> architectures. > That was going to be my next question. Is there an example anywhere on > how to do this? Is it in the current sparc64 arch? yep, AVR32 and Blackfin currently do this: t2-trunk# l architecture/*/package/gcc -d René >> E.g. I also got trouble on sparc/sparc64. In general gcc 4.4 did work there, >> just that the boot loader (silo) binaries are compiled in a way that >> circumvents this boot loader's very strange binary patching to work. gcc 4.5 >> does also not compile for sparc64 at all. I have not investigated the later, >> one of it's configure scripts errored out not being able to execute some >> compiled binary, ... >> > Ah yes, things like this always seem to go smoothly on x86, but bomb out > horribly on the other architectures. The price to pay for being on the > bleeding edge. > > Jan >> René >> >> On May 5, 2010, at 11:09 PM, Jan Rovins wrote: >> >> >>> I tried the latest GCC 4.5 snapshot, and am still getting the same errors. >>> I have opened a bug with GCC on the e2fsprogs/N32-ABI failure >>> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43995 >>> They are currently looking into it. >>> >>> The other one may be the same bug too, so I will not submit that yet. >>> >>> Jan >>> >>> Jan Rovins wrote: >>> >>>> Hi All, >>>> >>>> I have recently updated my trunk to the latest, and have been trying to >>>> build my slightly modified generic-minimal target. >>>> >>>> In the past I was able to successfully build a N64 ABI (single lib), and >>>> a N32 ABI (single lib) but was not able to build multilib yet. >>>> In switching to GCC 4.5.0, I am now getting internal compiler errors in >>>> two places: >>>> >>>> In the N64 ABI build, the failure is in building stage 1 GCC: >>>> ----------------------------------------------------------------------------------------- >>>> mips64-t2-linux-gnu-gcc -c -g -O2 -DIN_GCC -W -Wall -Wwrite-strings >>>> -Wcast-qual -Wstrict-prototypes -Wmissing-prototypes >>>> -Wmissing-format-attribute -pedantic -Wno-long-long -Wno-variadic-macros >>>> -Wno-overlength-strings -Wold-style-definition -Wc++-compat >>>> -DHAVE_CONFIG_H >>>> -I. -I. -I../../gcc -I../../gcc/. -I../../gcc/../include >>>> -I../../gcc/../libcpp/include -I../../gcc/../libdecnumber >>>> -I../../gcc/../libdecnumber/dpd -I../libdecnumber >>>> ../../gcc/regrename.c -o >>>> regrename.o >>>> Please submit a full bug report, >>>> with preprocessed source if appropriate. >>>> See <http://gcc.gnu.org/bugs.html> for instructions. >>>> mips64-t2-linux-gnu-gcc -c -g -O2 -DIN_GCC -W -Wall -Wwrite-strings >>>> -Wcast-qual -Wstrict-prototypes -Wmissing-prototypes >>>> -Wmissing-format-attribute -pedantic -Wno-long-long -Wno-variadic-macros >>>> -Wno-overlength-strings -Wold-style-definition -Wc++-compat >>>> -DHAVE_CONFIG_H >>>> -I. -I. -I../../gcc -I../../gcc/. -I../../gcc/../include >>>> -I../../gcc/../libcpp/include -I../../gcc/../libdecnumber >>>> -I../../gcc/../libdecnumber/dpd -I../libdecnumber ../../gcc/regstat.c >>>> -o >>>> regstat.o >>>> make[2]: *** [lto-streamer-in.o] Error 1 >>>> make[2]: *** Waiting for unfinished jobs.... >>>> rm gcc.pod >>>> make[2]: Leaving directory >>>> `/opt/T2/t2-trunk/src.gcc.GW-Octeon-Pure64.20100504.121745.9249.ubuntu1/gcc-4.5.0/objdir/gcc' >>>> make[1]: *** [all-gcc] Error 2 >>>> make[1]: Leaving directory >>>> `/opt/T2/t2-trunk/src.gcc.GW-Octeon-Pure64.20100504.121745.9249.ubuntu1/gcc-4.5.0/objdir' >>>> make: *** [all] Error 2 >>>> Due to previous errors, no 1-gcc.log file! >>>> (Try enabling xtrace in the config to track an error inside the build >>>> system.) >>>> --- BUILD ERROR --- >>>> >>>> -------------------------------------------------------------------------------- >>>> In the N32 ABI, it gets through the stage 1 GCC build, but fails during >>>> stage 1 e2fsprogs. >>>> --------------------------------------------------------------------------------- >>>> CC crc32.c >>>> recovery.c: In function 'do_one_pass': >>>> recovery.c:713:1: internal compiler error: Segmentation fault >>>> Please submit a full bug report, >>>> with preprocessed source if appropriate. >>>> See <http://gcc.gnu.org/bugs.html> for instructions. >>>> make[2]: *** [recovery.o] Error 1 >>>> make[2]: Leaving directory >>>> `/opt/T2/t2-trunk/src.e2fsprogs.GW-Octeon-N32.20100504.194156.29137.ubuntu1/e2fsprogs-1.41.11/e2fsck' >>>> make[1]: *** [all-progs-recursive] Error 1 >>>> make[1]: Leaving directory >>>> `/opt/T2/t2-trunk/src.e2fsprogs.GW-Octeon-N32.20100504.194156.29137.ubuntu1/e2fsprogs-1.41.11' >>>> make: *** [all] Error 2 >>>> >>>> -------------------------------------------------------------------------------------- >>>> >>>> I'm not sure if doing anything inside of T2 will fix this. The first thing >>>> I will try is to use a newer GCC 4.5.X development snapshot, and see if >>>> the errors go away. If that does not work, I should probibaly submit this >>>> as a bug to the GCC project folk. >>>> >>>> >>>> Jan >>>> >>>> >>>> >>>> ----------------------------------------------------------- If you wish to >>>> unsubscribe from this mailing, send mail to >>>> [email protected] with a subject of: unsubscribe t2 >>>> >>> >>> ----------------------------------------------------------- If you wish to >>> unsubscribe from this mailing, send mail to >>> [email protected] with a subject of: unsubscribe t2 >>> >> >> > -- René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin DE Legal: Amtsgericht Charlottenburg HRB 105123B, Tax-ID#: DE251602478 Managing Directors: Susanne Klaus, René Rebe http://exactcode.com | http://t2-project.org | http://rene.rebe.de
----------------------------------------------------------- If you wish to unsubscribe from this mailing, send mail to [email protected] with a subject of: unsubscribe t2
