blob: f35df97bfcfeece850f1d733237ffb4b51216d85 [file] [log] [blame]
Brad Bishopd7bf8c12018-02-25 22:55:05 -05001* e.g. with gentoo gcc-7.1 they define _FORTIFY_SOURCE by default with:
2 https://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo/src/patchsets/gcc/7.1.0/gentoo/10_all_default-fortify-source.patch?view=markup
3 which results in following error while building grub-efi-native:
4 ./config-util.h:1504:48: error: this use of "defined" may not be portable [-Werror=expansion-to-defined]
5 || (defined _FORTIFY_SOURCE && 0 < _FORTIFY_SOURCE \
6 ^~~~~~~~~~~~~~~
7 this part comes from gnulib and it's used only for Apple and BSD,
8 so we can ignore it, but we cannot add -Wno-error=expansion-to-defined
9 because this warning was introduced only in gcc-7 and older gcc
10 will fail with:
11 cc1: error: -Werror=expansion-to-defined: no option -Wexpansion-to-defined
12 use #pragma to work around this
13
14Upstream-Status: Pending (should be fixed in gnulib which is then rarely updated in grub)
15
16Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
17
18diff -uNr grub-2.02.old/m4/extern-inline.m4 grub-2.02/m4/extern-inline.m4
19--- grub-2.02.old/m4/extern-inline.m4 2016-02-28 15:22:21.000000000 +0100
20+++ grub-2.02/m4/extern-inline.m4 2017-08-22 19:26:45.213637276 +0200
21@@ -39,6 +39,10 @@
22 OS X 10.9 has a macro __header_inline indicating the bug is fixed for C and
23 for clang but remains for g++; see <http://trac.macports.org/ticket/41033>.
24 Assume DragonFly and FreeBSD will be similar. */
25+#pragma GCC diagnostic push
26+#if __GNUC__ >= 7
27+#pragma GCC diagnostic ignored "-Wexpansion-to-defined"
28+#endif
29 #if (((defined __APPLE__ && defined __MACH__) \
30 || defined __DragonFly__ || defined __FreeBSD__) \
31 && (defined __header_inline \
32@@ -50,6 +52,7 @@
33 && defined __GNUC__ && ! defined __cplusplus))))
34 # define _GL_EXTERN_INLINE_STDHEADER_BUG
35 #endif
36+#pragma GCC diagnostic pop
37 #if ((__GNUC__ \
38 ? defined __GNUC_STDC_INLINE__ && __GNUC_STDC_INLINE__ \
39 : (199901L <= __STDC_VERSION__ \