https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95413
H.J. Lu <hjl.tools at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hjl.tools at gmail dot com
Status|UNCONFIRMED |WAITING
Last reconfirmed| |2020-05-29
Ever confirmed|0 |1
--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Rainer Orth from comment #0)
> Between 20200515 and 20200522, Linux/i686 bootstrap with --enable-targets=all
> broke: after configuring with
>
> $ .../configure CC='gcc -m32' CXX='g++ -m32' --enable-targets=all
> --enable-languages=c --build=i686-pc-linux-gnu --disable-lto-plugin
>
> configuring the stage1 64-bit libgomp fails:
>
> checking whether the C compiler works... no
> configure: error: in
> `/var/gcc/gcc-11.0.0-20200529/4.20.4-gcc-32/i686-pc-linux-gnu/64/libgomp':
> configure: error: C compiler cannot create executables
> See `config.log' for more details
> make[2]: *** [Makefile:20360: configure-stage1-target-libgomp] Error 1
>
> config.log has
>
> configure:3935:
> /var/gcc/regression/master/4.20.4-gcc-gas-gld-32/build/./gcc/xgcc
> -B/var/gcc/regression/master/4.20.4-gcc-gas-gld-32/build/./gcc/
> -B/vol/gcc/i686-pc-linux-gnu/bin/ -B/vol/gcc/i686-pc-linux-gnu/lib/ -isystem
> /vol/gcc/i686-pc-linux-gnu/include -isystem
> /vol/gcc/i686-pc-linux-gnu/sys-include -fno-checking -m64 -g -O2 -pthread
> -Wall -Werror -ftls-model=initial-exec -DUSING_INITIAL_EXEC_TLS -march=i486
> -mtune=i686 conftest.c >&5
> cc1: error: CPU you selected does not support x86-64 instruction set
>
> whereas before this was
>
> configure:3913: checking whether the C compiler works
> configure:3935:
> /var/gcc/regression/master/4.20.4-gcc-gas-gld-32/build/./gcc/xgcc
> -B/var/gcc/regression/master/4.20.4-gcc-gas-gld-32/build/./gcc/
> -B/vol/gcc/i686-pc-linux-gnu/bin/ -B/vol/gcc/i686-pc-linux-gnu/lib/ -isystem
> /vol/gcc/i686-pc-linux-gnu/include -isystem
> /vol/gcc/i686-pc-linux-gnu/sys-include -fno-checking -m64 -g -O2 -pthread
> conftest.c >&5
>
> I strongly suspect the CET patches, but haven't yet figured out what's going
> wrong.
How did CET changes add -march=i486 -mtune=i686? Can you bisect to the commit?
