From: "Arsen Arsenović" <arsen@aarsen.me>
To: gcc-patches@gcc.gnu.org
Cc: libstdc++@gcc.gnu.org, "Arsen Arsenović" <arsen@aarsen.me>
Subject: [PATCH 02/10] libstdc++: Filter out unconditional <stdio.h> default include
Date: Fri, 30 Sep 2022 18:45:48 +0200 [thread overview]
Message-ID: <20220930164556.1198044-3-arsen@aarsen.me> (raw)
In-Reply-To: <20220930164556.1198044-1-arsen@aarsen.me>
_AC_INCLUDES_DEFAULT_REQUIREMENTS including <stdio.h> when checking for
stdint.h has prevented proper detection of whether stdint.h is present,
since it'd poison the cache variables with test results failing due to
<stdio.h> failing to include. As a solution, for autoconf versions under
2.70, we filter out that bit of code from ac_includes_default.
This issue was fixed in autoconf-2.70.
This also applies to various other headers, but was noticed when looking
into why HAVE_STDINT_H was misdefined.
libstdc++-v3/ChangeLog:
* configure.ac: Remove any lines that unconditionally include
<stdio.h> from ac_includes_default, when running Autoconf <2.70.
* configure: Regenerate.
Signed-off-by: Arsen Arsenović <arsen@aarsen.me>
---
libstdc++-v3/configure.ac | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/libstdc++-v3/configure.ac b/libstdc++-v3/configure.ac
index b351622bf46..c05fcdda7e9 100644
--- a/libstdc++-v3/configure.ac
+++ b/libstdc++-v3/configure.ac
@@ -58,6 +58,15 @@ else
GLIBCXX_IS_NATIVE=true
fi
+# In the case that we're building without headers, we won't have <stdio.h>
+# available. In these cases, we have to instruct autotools to never include
+# <stdio.h> as a part of default headers.
+m4_version_prereq([2.70], [], [
+if test "x$with_headers" = "xno"; then
+ ac_includes_default=`echo "$ac_includes_default" | sed '/^#include <stdio.h>$/d'`
+fi
+])
+
# Sets up automake. Must come after AC_CANONICAL_SYSTEM. Each of the
# following is magically included in AUTOMAKE_OPTIONS in each Makefile.am.
# 1.x: minimum required version
--
2.37.3
next prev parent reply other threads:[~2022-09-30 16:46 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-30 16:45 [PATCH 00/10] c-family,libstdc++: P1642 and related changes Arsen Arsenović
2022-09-30 16:45 ` [PATCH 01/10] libstdc++: Make _GLIBCXX_HOSTED respect -ffreestanding [PR103626] Arsen Arsenović
2022-09-30 16:45 ` Arsen Arsenović [this message]
2022-09-30 16:45 ` [PATCH 03/10] libstdc++: Adjust precompiled headers for freestanding Arsen Arsenović
2022-09-30 16:45 ` [PATCH 04/10] libstdc++: Mark headers that must be hosted as such [PR103626] Arsen Arsenović
2022-09-30 16:45 ` [PATCH 05/10] c-family: Implement new `int main' semantics in freestanding Arsen Arsenović
2022-09-30 16:45 ` [PATCH 06/10] libstdc++: Rework how freestanding install works [PR106953] Arsen Arsenović
2022-09-30 16:45 ` [PATCH 07/10] libstdc++: Make some tests work on freestanding [PR103626] Arsen Arsenović
2022-09-30 16:45 ` [PATCH 08/10] libstdc++: Add effective-target 'hosted' for testsuite [PR103626] Arsen Arsenović
2022-09-30 16:45 ` [PATCH 09/10] libstdc++: Re-enable std::hash<std::bitset> in freestanding [PR103626] Arsen Arsenović
2022-09-30 16:45 ` [PATCH 10/10] libstdc++: Disable hosted-only tests [PR103626] Arsen Arsenović
2022-09-30 18:15 ` [PATCH 00/10] c-family,libstdc++: P1642 and related changes Jonathan Wakely
2022-10-03 14:49 ` Jonathan Wakely
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220930164556.1198044-3-arsen@aarsen.me \
--to=arsen@aarsen.me \
--cc=gcc-patches@gcc.gnu.org \
--cc=libstdc++@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).