This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: 3.0 bootstrap broken on irix6.5
- To: gcc-bugs at gcc dot gnu dot org
- Subject: Re: 3.0 bootstrap broken on irix6.5
- From: Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com>
- Date: Tue, 2 Oct 2001 03:33:24 -0500 (CDT)
- Cc: David dot Billinghurst at riotinto dot com
- Organization: Networks and Infrastructure Lab (IL02/2240), Motorola Labs
In article <8D00C32549556B4E977F81DBC24E985D40FF81@crtsmail1.technol_exch.corp.riotinto.org> you write:
> 3.0 bootstrap is broken when using cc on irix6.5. We are passing gcc flags
> to cc when compiling gengenrtl.o in stage2, giving
>
> make[1]: Entering directory `/exd4/billingd/obj/gcc-3.0/gcc'
> make CC="stage1/xgcc -Bstage1/
> -B/exd4/billingd/tmp/gcc-3.0/mips-sgi-irix6.5/bin/" \
> STAGE_PREFIX=stage1/ \
> HOST_CC="cc" CFLAGS="-g -O2" LDFLAGS=""
Someone shoot me (figuratively speaking only). I'm fairly sure that
this is related to the patch I installed earlier today where I
propagate HOST_CC from stage 1 to later stages. Obviously (to me now
in light of your findings), one must propagate all HOST_ variables
that might be recomputed in a submake if we pass any of them... I
wonder how one is suppose to determine that complete set without any
trial and error. Or, should they all be propagated without study?
To fix the problem David uncovered, I need input from the
gcc/Makefile.in gurus. Here is the full thread on the broken (or
half-complete) patch:
http://gcc.gnu.org/ml/gcc-patches/2001-09/msg01179.html
http://gcc.gnu.org/ml/gcc-patches/2001-10/msg00032.html
Regards,
Loren
--
Loren J. Rittle
Senior Staff Software Engineer, Distributed Object Technology Lab
Networks and Infrastructure Research Lab (IL02/2240), Motorola Labs
rittle@rsch.comm.mot.com, KeyID: 2048/ADCE34A5, FDC0292446937F2A240BC07D42763672