lttng-babeltrace: fix build failure on gcc14

The gcc14 machine has a moderately old host gcc, which was causing a
recurrent build failure of host-lttng-babeltrace:

 http://autobuild.buildroot.org/results/492/49216052c161874f41738e41e6e0c89a6dd04000/build-end.log

This commit adds a patch to fix this build failure, related to the
access of the ULLONG_MAX define on compilers that did not default to
the C99 variant of the C language.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 ...beltrace-01-no-posix-fallocate-in-uclibc.patch} |    0
 ...tng-babeltrace-02-fix-build-old-compilers.patch |   27 ++++++++++++++++++++
 package/lttng-babeltrace/lttng-babeltrace.mk       |    4 +++
 3 files changed, 31 insertions(+)
 rename package/lttng-babeltrace/{lttng-babeltrace-no-posix-fallocate-in-uclibc.patch => lttng-babeltrace-01-no-posix-fallocate-in-uclibc.patch} (100%)
 create mode 100644 package/lttng-babeltrace/lttng-babeltrace-02-fix-build-old-compilers.patch

>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:

 Thomas> The gcc14 machine has a moderately old host gcc, which was causing a
 Thomas> recurrent build failure of host-lttng-babeltrace:

 Thomas>  http://autobuild.buildroot.org/results/492/49216052c161874f41738e41e6e0c89a6dd04000/build-end.log

 Thomas> This commit adds a patch to fix this build failure, related to the
 Thomas> access of the ULLONG_MAX define on compilers that did not default to
 Thomas> the C99 variant of the C language.

 Thomas> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
 Thomas> ---
 Thomas>  ...beltrace-01-no-posix-fallocate-in-uclibc.patch} |    0
 Thomas>  ...tng-babeltrace-02-fix-build-old-compilers.patch |   27 ++++++++++++++++++++
 Thomas>  package/lttng-babeltrace/lttng-babeltrace.mk       |    4 +++
 Thomas>  3 files changed, 31 insertions(+)
 Thomas>  rename package/lttng-babeltrace/{lttng-babeltrace-no-posix-fallocate-in-uclibc.patch => lttng-babeltrace-01-no-posix-fallocate-in-uclibc.patch} (100%)
 Thomas>  create mode 100644 package/lttng-babeltrace/lttng-babeltrace-02-fix-build-old-compilers.patch

 Thomas> +++ b/package/lttng-babeltrace/lttng-babeltrace.mk
 Thomas> @@ -4,5 +4,9 @@ LTTNG_BABELTRACE_SOURCE  = babeltrace-$(LTTNG_BABELTRACE_VERSION).tar.bz2
 
 Thomas>  LTTNG_BABELTRACE_DEPENDENCIES = popt util-linux libglib2
 
 Thomas> +# We patch configure.ac to fix a problem that appears
 Thomas> +# only with some old host gcc
 Thomas> +HOST_LTTNG_BABELTRACE_AUTORECONF = YES

The patch also gets applied for target builds. Don't we need to
autoreconf there as well?

Dear Peter Korsgaard,

On Tue, 28 May 2013 15:29:42 +0200, Peter Korsgaard wrote:

>  Thomas> +# We patch configure.ac to fix a problem that appears
>  Thomas> +# only with some old host gcc
>  Thomas> +HOST_LTTNG_BABELTRACE_AUTORECONF = YES
> 
> The patch also gets applied for target builds. Don't we need to
> autoreconf there as well?

I was focused on the host gcc problem, but if the user decides to use
an old gcc as the cross-compiler, then there may be the same issue.

And anyway, patching configure.ac without doing the autoreconf means
that at build time, the thing will notice that configure.ac is more
recent than configure, and will do some kind of autoreconf.

So yes, adding LTTNG_BABELTRACE_AUTORECONF = YES is probably a good
thing to do.

BTW:

 * I've submitted the patch upstream. See
   http://lists.lttng.org/pipermail/lttng-dev/2013-May/020542.html and
   http://lists.lttng.org/pipermail/lttng-dev/2013-May/020543.html.

 * This host/target difference on autoreconf will disappear with the
   out-of-tree stuff, since we'll have only one source tree.

Best regards,

Thomas