This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Bogus trees from Ada front-end (more VRP vs Ada) stuff)

> This sounds like fold is merging the two comparisons above incorrectly into 
> the one upper comparison.


> It can do the merge, but it needs to convert to base types first.

The type at stake is a ENUMERAL_TYPE and those have no base type

 <enumeral_type 0x5572333c ada__strings__trim_end readonly sizes-gimplified 
unsigned QI
    size <integer_cst 0x556d71e0 type <integer_type 0x556e505c bit_size_type> 
constant invariant visited 8>
    unit size <integer_cst 0x556d71f8 type <integer_type 0x556e5000 long int> 
constant invariant visited 1>
    user align 8 symtab 0 alias set -1 precision 8 min <integer_cst 0x5571adb0 
0> max <integer_cst 0x5571ae40 2>
    RM size <integer_cst 0x5571aea0 type <integer_type 0x556e505c 
bit_size_type> constant invariant 2>>

so the provision for base types in build_range_check doesn't trigger:

  if (value != 0 && ! TREE_OVERFLOW (value))
      /* There is no requirement that LOW be within the range of ETYPE
	 if the latter is a subtype.  It must, however, be within the base
	 type of ETYPE.  So be sure we do the subtraction in that type.  */
      if (INTEGRAL_TYPE_P (etype) && TREE_TYPE (etype))
	  etype = TREE_TYPE (etype);
	  exp = fold_convert (etype, exp);
	  low = fold_convert (etype, low);
	  value = fold_convert (etype, value);

      return build_range_check (type,
				fold_build2 (MINUS_EXPR, etype, exp, low),
				1, build_int_cst (etype, 0), value);

I gather that we should restrict the transformation to INTEGER_TYPEs.

Eric Botcazou

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]