This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: How to configure a bi-arch PowerPC GCC?
Hi,
On Mon, Jan 30, 2017 at 01:13:13PM +0100, Sebastian Huber wrote:
> The compiler build shows these warnings:
>
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11023:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11034:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11034:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11050:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11050:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11060:1:
> warning: source missing a mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11060:1:
> warning: source missing a mode?
These are normal. Maybe it would work (and not warn) if we left off
the constraints as well, hrm.
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/rs6000.md:11998:1:
> warning: operand 1 missing mode?
This is an SPE pattern, won't be a problem for you. It should probably
be CCmode fwiw.
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/spe.md:3467:1:
> warning: operand 1 missing mode?
> /home/EB/sebastian_h/archive/gcc-git/gcc/config/rs6000/spe.md:3467:1:
> warning: operand 2 missing mode?
Similar.
> Statistics for recog:
> Number of decisions: 15193
> longest path: 604 (code: 593)
> longest backtrack: 43 (code: 1619)
> Statistics for split_insns:
> Number of decisions: 2991
> longest path: 53 (code: 576)
> longest backtrack: 18 (code: 574)
> Statistics for peephole2_insns:
> Number of decisions: 267
> longest path: 47 (code: 7)
> longest backtrack: 3 (code: 26)
>
> Is this something to worry about?
Nope, perfectly normal.
Segher