Re: buildfailure in expr.c on darwin

[Devang Patel <dpatel at apple dot com>]

Increasing the stack size fixes one expr.c crash during bootstrap. I do 
not remember
this crash you are seeing is the same one or not, but try with higher 
stack size.

Devang

On Wednesday, November 21, 2001, at 01:30  PM, Andreas Tobler wrote:

> Hi,
>
> I get the following failure on darwin with the changes from last night.
> I can't figure out where the related change comes from.
> I produced a .i file and tried to find the failing reason. Below I
> copied also the part from the .i file which I think is related to. Any
> hints on how to find the reason?
>
> Thanks
> Andreas
>
> BTW, is gcc.gnu.org down?
>
> ----
> stage1/xgcc -Bstage1/
> -B/Volumes/reserved1/gcctmp/powerpc-apple-darwin5.1/bin/ -c  -DIN_GCC
> -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
> -Wtraditional -pedantic -Wno-long-long  -DHAVE_CONFIG_H    -I. -I.
> -I../../gcc/gcc -I../../gcc/gcc/. -I../../gcc/gcc/config
> -I../../gcc/gcc/../include ../../gcc/gcc/expr.c -o expr.o
> ../../gcc/gcc/expr.c: In function `expand_expr':
> ../../gcc/gcc/expr.c:8988: output_operand: invalid %%xn code
> ----
> .i part:
> ----
>   op0 = expand_expr ((__extension__ ({ const tree __t = exp; char __c = 
> tree_cod
> e_type[(int) (((enum tree_code) (__t)->common.code))]; if (__c != 'r' &&
> __c !=
> 's' && __c != '<' && __c != '1' && __c != '2' && __c != 'e') 
> tree_class_check_fa
> iled(__t, 'e', "../../gcc/gcc/expr.c", 8980, __FUNCTION__); __t; 
> })->exp.operand
> s[0]), subtarget, VOIDmode, 0);
>   op1 = expand_expr ((__extension__ ({ const tree __t = exp; char __c = 
> tree_cod
> e_type[(int) (((enum tree_code) (__t)->common.code))]; if (__c != 'r' &&
> __c !=
> 's' && __c != '<' && __c != '1' && __c != '2' && __c != 'e') 
> tree_class_check_fa
> iled(__t, 'e', "../../gcc/gcc/expr.c", 8981, __FUNCTION__); __t; 
> })->exp.operand
> s[1]), (rtx) 0, VOIDmode, 0);
>  binop2:
>   temp = expand_binop (mode, this_optab, op0, op1, target,
>                        unsignedp, OPTAB_LIB_WIDEN);
>   if (temp == 0)
>     fancy_abort ("../../gcc/gcc/expr.c", 8986, __FUNCTION__);
>   return temp;
> }
> ----
> gdb:
> ----
> (gdb) run
> Starting program: /Users/andreast/gccsrc/objdir/gcc/stage1/xgcc
> -Bstage1/ -B/Volumes/reserved1/gcctmp/powerpc-apple-darwin5.1/bin/ -c
> -DIN_GCC    -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes
> -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long
> -DHAVE_CONFIG_H    -I. -I. -I../../gcc/gcc -I../../gcc/gcc/.
> -I../../gcc/gcc/config -I../../gcc/gcc/../include ../../gcc/gcc/expr.c
> -o expr.o
> [Switching to thread 1 (process 14570 thread 0x1603)]
> ../../gcc/gcc/expr.c: In function `expand_expr':
> ../../gcc/gcc/expr.c:8988: output_operand: invalid %%xn code
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
>
> Breakpoint 2, 0x70005b88 in exit ()
> (gdb) bt
> #0  0x70005b88 in exit ()
> #1  0x00001a30 in _call_mod_init_funcs ()
> #2  0x70005b84 in exit ()
> (gdb)
>
>