2008-05-07 Jakub Jelinek <jakub@redhat.com>
+ PR middle-end/36137
+ * fold-const.c (fold_binary): Use STRIP_SIGN_NOPS instead of
+ STRIP_NOPS on arguments even for MIN_EXPR and MAX_EXPR.
+
PR middle-end/36106
* omp-low.c (expand_omp_atomic_pipeline): Load value using the
integral type rather than floating point, then VIEW_CONVERT_EXPR
to the floating point type.
- * testsuite/libgomp.c/atomic-5.c: New test.
- * testsuite/libgomp.c/atomic-6.c: New test.
- * testsuite/libgomp.c/autopar-1.c: New test.
-
2008-05-07 Uros Bizjak <ubizjak@gmail.com>
* config/i386/i386.c (ix86_expand_copysign): Force non-zero constant
/* Fold a constant sub-tree into a single node for C-compiler
Copyright (C) 1987, 1988, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
- 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
+ 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
Free Software Foundation, Inc.
This file is part of GCC.
safe for every expression, except for a comparison expression
because its signedness is derived from its operands. So, in
the latter case, only strip conversions that don't change the
- signedness.
+ signedness. MIN_EXPR/MAX_EXPR also need signedness of arguments
+ preserved.
Note that this is done as an internal manipulation within the
constant folder, in order to find the simplest representation
cases, the appropriate type conversions should be put back in
the tree that will get out of the constant folder. */
- if (kind == tcc_comparison)
+ if (kind == tcc_comparison || code == MIN_EXPR || code == MAX_EXPR)
{
STRIP_SIGN_NOPS (arg0);
STRIP_SIGN_NOPS (arg1);
--- /dev/null
+/* PR middle-end/36137 */
+extern void abort (void);
+
+#define MIN(a, b) ((a) < (b) ? (a) : (b))
+#define MAX(a, b) ((a) > (b) ? (a) : (b))
+
+int
+main ()
+{
+ unsigned int u;
+ int i = -1;
+
+ u = MAX ((unsigned int) MAX (i, 0), 1);
+ if (u != 1)
+ abort ();
+
+ u = MIN ((unsigned int) MAX (i, 0), (unsigned int) i);
+ if (u != 0)
+ abort ();
+ return 0;
+}