Re: bootstrap failure on darwin

[Geoff Keating <geoffk at geoffk dot org>]

On 17/08/2006, at 12:33 PM, Geoff Keating wrote:

> Hi All,
>
> Between revision 116195 and 116224, GCC started to fail to  
> bootstrap on powerpc-darwin, in stage2.
>
> The build log says
>
> /Users/regress/tbox/svn-gcc/gcc/dwarf2out.c: In function  
> 'output_call_frame_info':
> /Users/regress/tbox/svn-gcc/gcc/dwarf2out.c:2160: internal compiler  
> error: in build_polynomial_chrec, at tree-chrec.h:108

I reproduced the same failure on i386-darwin9, so it doesn't seem  
especially tied to architecture or OS version.

...

> ---------------------------------------------------------------------- 
> --
> r116213 | rakdver | 2006-08-17 01:22:05 -0700 (Thu, 17 Aug 2006) |  
> 11 lines
>
>         PR tree-optimization/27865
>         * tree-vrp.c (adjust_range_with_scev): Do not use TYPE_ 
> {MIN,MAX}_VALUE
>         for pointer types.
>         * tree-scalar-evolution.c (fold_used_pointer_cast,  
> pointer_offset_p,
>         fold_used_pointer, pointer_used_p): New functions.
>         (analyze_scalar_evolution_1): Use fold_used_pointer.
>         * tree-chrec.c (convert_affine_scev): Convert no-op casts  
> correctly.
>         * tree-ssa-loop-ivopts.c (generic_type_for): Return  
> integral type
>         for pointers.

I can confirm that backing out this change lets the build proceed.

Attachment: smime.p7s
Description: S/MIME cryptographic signature