Dear all, in find_substring_ref we erroneously handled given substring start and end indices as unsigned integers. However, gives indices could be negative, which is legal as long as end < start, leading to a string of length zero. The current behavior could lead to a wrong length as well as an invalid read from (compiler) memory.
The fix allows to reintroduce code in testcase substr_6.f90 that was erroneously considered as illegal. Regtested on x86_64-pc-linux-gnu. OK for mainline? As this is invalid code, I'd like to backport this fix. Thanks, Harald Fortran - fix handling of substring start and end indices gcc/fortran/ChangeLog: PR fortran/85130 * expr.c (find_substring_ref): Handle given substring start and end indices as signed integers, not unsigned. gcc/testsuite/ChangeLog: PR fortran/85130 * gfortran.dg/substr_6.f90: Revert commit r8-7574, adding again test that was erroneously considered as illegal.
diff --git a/gcc/fortran/expr.c b/gcc/fortran/expr.c index dfecc3012e1..604e63e6164 100644 --- a/gcc/fortran/expr.c +++ b/gcc/fortran/expr.c @@ -1724,8 +1724,8 @@ find_substring_ref (gfc_expr *p, gfc_expr **newp) *newp = gfc_copy_expr (p); free ((*newp)->value.character.string); - end = (gfc_charlen_t) mpz_get_ui (p->ref->u.ss.end->value.integer); - start = (gfc_charlen_t) mpz_get_ui (p->ref->u.ss.start->value.integer); + end = (gfc_charlen_t) mpz_get_si (p->ref->u.ss.end->value.integer); + start = (gfc_charlen_t) mpz_get_si (p->ref->u.ss.start->value.integer); if (end >= start) length = end - start + 1; else diff --git a/gcc/testsuite/gfortran.dg/substr_6.f90 b/gcc/testsuite/gfortran.dg/substr_6.f90 index 0d5e3d75e88..83e788a55a6 100644 --- a/gcc/testsuite/gfortran.dg/substr_6.f90 +++ b/gcc/testsuite/gfortran.dg/substr_6.f90 @@ -6,6 +6,8 @@ CHARACTER(5), parameter :: c0(1) = (/ "123" // ACHAR(0) // "5" /) CHARACTER*5 c(1) CHARACTER(1), parameter :: c1(5) = (/ "1", "2", "3", ACHAR(0), "5" /) +c = c0(1)(-5:-8) +if (c(1) /= " ") STOP 1 c = (/ c0(1)(1:5) /) do i=1,5 if (c(1)(i:i) /= c1(i)) STOP 2