Dear All,

I've committed the attached patch for a regression on some 32-bit
targets as obvious after a hint by Mikael and verification of the
fix by Rainer in the PR as r16-6322-gabf3f0b24c4d07 .

Thanks,
Harald

From abf3f0b24c4d07649266371c0f8f9625dff58cd3 Mon Sep 17 00:00:00 2001
From: Harald Anlauf <[email protected]>
Date: Sun, 21 Dec 2025 23:03:28 +0100
Subject: [PATCH] fortran: fix testsuite regression for gfortran.dg/value_9.f90
 [PR123201]

Commit r16-3499 introduced a regression on targets where truncation of a
string argument passed to a CHARACTER(len=1),VALUE dummy argument missed
the special treatment needed for passing single characters.

	PR fortran/123201

gcc/fortran/ChangeLog:

	* trans-expr.cc (conv_dummy_value): Convert string of length 1 to a
	single character for passing as actual argument.
---
 gcc/fortran/trans-expr.cc | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/gcc/fortran/trans-expr.cc b/gcc/fortran/trans-expr.cc
index b549a62880e..20f74927153 100644
--- a/gcc/fortran/trans-expr.cc
+++ b/gcc/fortran/trans-expr.cc
@@ -6654,6 +6654,14 @@ conv_dummy_value (gfc_se * parmse, gfc_expr * e, gfc_symbol * fsym,
 						  e->value.character.string);
       parmse->string_length = build_int_cst (gfc_charlen_type_node, flen);
 
+      if (flen == 1)
+	{
+	  tree slen1 = build_int_cst (gfc_charlen_type_node, 1);
+	  gfc_conv_string_parameter (parmse);
+	  parmse->expr = gfc_string_to_single_character (slen1, parmse->expr,
+							 e->ts.kind);
+	}
+
       /* Indicate value,optional scalar dummy argument as present.  */
       if (fsym->attr.optional)
 	vec_safe_push (optionalargs, boolean_true_node);
-- 
2.51.0

Reply via email to