This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH/committed] tst-fanotify: check for linux/fanotify.h existence
- From: "Joseph S. Myers" <joseph at codesourcery dot com>
- To: Mike Frysinger <vapier at gentoo dot org>
- Cc: <libc-alpha at sourceware dot org>
- Date: Tue, 31 Dec 2013 16:00:00 +0000
- Subject: Re: [PATCH/committed] tst-fanotify: check for linux/fanotify.h existence
- Authentication-results: sourceware.org; auth=none
- References: <201309052328 dot 17452 dot vapier at gentoo dot org> <1388477828-16559-1-git-send-email-vapier at gentoo dot org>
On Tue, 31 Dec 2013, Mike Frysinger wrote:
> +# Until we start requiring 2.6.37+ headers, we need to check for the
> +# availability of linux/fanotify.h for testing purposes.
> +AC_CHECK_HEADER(linux/fanotify.h,
> + [DEFINES="$DEFINES -DHAVE_LINUX_FANOTIFY_H=1"], ,
> + [/* No default includes. */])
Why are you using DEFINES instead of the normal approach of AC_DEFINE?
We know it's not ideal for the toplevel config.h.in to contain
system-specific defines - see bug 14068 - but I don't think using DEFINES
is a good workaround; it's better to keep all such cases using the same
mechanism so it's easier to find them to make them use a newer cleaner
mechanism.
--
Joseph S. Myers
joseph@codesourcery.com