[Bug tree-optimization/77899] incorrect VR_RANGE for a signed char function argument

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77899

--- Comment #3 from Martin Sebor <msebor at gcc dot gnu.org> ---
The offset is this in plus_stmt_object_size() in tree-object-size.c:

 <ssa_name 0x7ffff0daa948
    type <integer_type 0x7ffff0da41f8 sizetype public unsigned DI
        size <integer_cst 0x7ffff0da0ca8 constant 64>
        unit size <integer_cst 0x7ffff0da0cc0 constant 8>
        align 64 symtab 0 alias set -1 canonical type 0x7ffff0da41f8 precision
64 min <integer_cst 0x7ffff0da0cd8 0> max <integer_cst 0x7ffff0db1480
18446744073709551615>>
    static visited
    def_stmt _1 = (sizetype) i_3(D);
    version 1
    ptr-info 0x7ffff0ec8f20>

I can work around it there by checking the type of i_3(D) (which is signed
char) and using its range to bound _1.  I wonder if this same workaround could
be put in get_range_info until POINTER_PLUS_EXPR is changed to take a signed
offset as you suggest.