After the newest changes, the CI fails on macOS 15 and 26. Namely, there is a
compilation error.

On macOS 26:

depbase=`echo builtin.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
        gcc    -I../../lib -I../lib -Wall   -g -O2 -MT builtin.o -MD -MP -MF 
$depbase.Tpo -c -o builtin.o ../../src/builtin.c &&\
        mv -f $depbase.Tpo $depbase.Po
../../src/builtin.c:1165:7: error: operand argument to checked integer 
operation must be an integer type other than plain 'char', 'bool', bit-precise, 
or an enumeration ('bool' invalid)
 1165 |   if (ckd_add (&alloc, MAX (digits, min), negative))
      |       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Applications/Xcode_26.5.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/21/include/stdckdint.h:37:54:
 note: expanded from macro 'ckd_add'
   37 | #define ckd_add(R, A, B) __builtin_add_overflow((A), (B), (R))
      |                                                      ^~~
1 error generated.
make[2]: *** [builtin.o] Error 1

Similarly on macOS 15.

Apparently the problem is that 'negative' is of type 'bool'.

The attached proposed patch fixes it.

>From c601cf6423e91401adef57fbe8a70df7eff89312 Mon Sep 17 00:00:00 2001
From: Bruno Haible <[email protected]>
Date: Thu, 25 Jun 2026 20:41:43 +0200
Subject: [PATCH] m4: fix compilation error on macOS (regression 2026-06-24)

* src/builtin.c (m4_eval): Convert third argument of ckd_add to int.
---
 src/builtin.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/builtin.c b/src/builtin.c
index c3d5413f..b9aa3efa 100644
--- a/src/builtin.c
+++ b/src/builtin.c
@@ -1162,7 +1162,7 @@ m4_eval (struct obstack *obs, int argc, token_data **argv)
     }
 
   size_t alloc;
-  if (ckd_add (&alloc, MAX (digits, min), negative))
+  if (ckd_add (&alloc, MAX (digits, min), +negative))
     xalloc_die ();
 
   obstack_blank (obs, alloc);
-- 
2.53.0

  • CI failure on ... Bruno Haible via Bug reports for the GNU m4 macro processor
    • Fwd: CI f... Bruno Haible via Bug reports for the GNU m4 macro processor

Reply via email to