Age | Commit message (Collapse) | Author |
|
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
Check for a specific file instead of a specific set of versions from
uname, to cope with manpage section alignment coming to 11.4 instead
of 12.0.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
|
|
Analogous to previous commit.
With this commit we no longer get the following erroneous messages
during `make distcheck'.
"
failed to copy INSTALL from util-macros: installing possibly empty INSTALL.
...
failed to copy INSTALL from util-macros: installing possibly empty INSTALL.
"
Cc: Gaetan Nadon <memsize@videotron.ca>
Cc: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
|
|
Currently CHANGELOG_CMD issues a
touch ChangeLog; echo "Scary message" >&2
even if the file is already present. For example, when running `make
distcheck' the following is observed:
"
/bin/sh: ../../.changelog.tmp: Permission denied
git directory not found: installing possibly empty changelog.
...
/bin/sh: ../../.changelog.tmp: Permission denied
git directory not found: installing possibly empty changelog.
"
That in itself happens since srcdir is RO, thus shell redirection [to
srcddir/.changelog.tmp] fails. At the same time the latter message is
wrong and misleading since the file is already there, with the correct
contents.
Silence the permissing warning (/dev/null is our friend), and add
a simple `test -e .../ChangeLog || (...' prior to the touch/echo combo.
As we're here also update the message to be more generic, since there
are multiple other reasons why things can fail.
Cc: Gaetan Nadon <memsize@videotron.ca>
Cc: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
|
|
Signed-off-by: Matt Turner <mattst88@gmail.com>
|
|
Signed-off-by: Mihail Konev <k.mvc@ya.ru>
|
|
Place quotes around the $srcdir, $ORIGDIR and $0 variables to prevent
fall-outs, when they contain space.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
|
|
Syncs the invocation of configure with the one from the server.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
|
|
Starting with xmlto version 0.0.27 the return code of
xmlto --skip-validation txt conftest.xml
is non-zero if conftest.xml is an empty file.
As a consequence the macro XORG_WITH_XMLTO returns
"xmlto cannot generate text format, this format skipped"
and therefore libraries like libxi, libxdmcp and others won't convert
docbook XML to text format.
This changed behavior was introduced with the following change in xmlto:
xmlto.in: use correctly exit code from xsltproc
See also: https://fedorahosted.org/xmlto/changeset/77
This patch fixes this by additionally testing xmlto with a non-empty XML
file.
More details can be found at:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=613674
Thanks to Peter Pearse, Helmut Grohne and Gaetan Nadon.
v2: To maintain compatibility with older xorg tarballs don't replace
the original test with the empty XML file but instead add a fallback
to additionally test with a non-empty XML file if the original test fails.
Use the alternate solution with <x></x> to skip compatibility issues
with different docbook versions.
Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
Signed-off-by: Julien Cristau <jcristau@debian.org>
|
|
All others (including other SunOS/Solaris releases) use the
traditional Bell Labs / BSD / Linux section numbering.
(The [[01]] is an m4-quoted shell character set of "[01]" to make sure
autoconf doesn't lose the [] when generating the configure script.)
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
|
|
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
In some cases we may want to have -traditional for proper whitespace preserving
without -undef, as we actually want the system definitions to be in place
so we can #ifdef on them. IE in xinit various .cpp files contain things like
#ifdef __APPLE__ and #ifdef __linux__
So this patch adds a TRADITIONALCPPFLAGS variable which contains just
-traditional where applicable without the other RAWCPPFLAGS for unsetting
the system definitions.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Now that the minimum level for automake is 1.11, replace the current
INSTALL with the one from automake 1.11.
Recall that the file in this package is installed in $pkgdatadir
and is copied by each xorg module through the make 'dist' target.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Originally from XCB, this macro checks for the presence of m4 or gm4
which supports -I dir.
The AC_PATH_PROGS_FEATURE_CHECK autoconf macro requires autoconf 2.62.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
The main motivation is to catch-up with the development reality and
allow use of features in Autoconf 2.62 as well as Automake 1.11.
As usual this means no features found only in versions above those
specified are allowed.
This is implementing the policy change which is described in:
http://www.x.org/wiki/Building_the_X_Window_System/?updated#index2h3
Discussion on xorg minimum autotools requirements:
http://lists.x.org/archives/xorg-devel/2013-October/038325.html
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
We've fixed several bugs it uncovered already, only one warning left that
it adds in a default build.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Seen in the XCB configuration and found missing in xorg.
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
It is part of the graphviz package and it may be used by doxygen if
the HAVE_DOT is set to "yes" in the doxygen configuration file.
The calling module handles if the tool is mandatory or not.
Doxygen will work fine without it, but no fancy graphics.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
On a full X build, this warning accounts for the 5,888 warnings on a total
of 12,042. The line is moved to the "noisy" section of warnings.
There is a discussion about the pros and cons here:
http://lists.x.org/archives/xorg-devel/2013-November/039069.html
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Improve configure speed by checking the autoconf cache before compiling and
running the test.
As this means cross-compilation environments can trivially seed this cache as
appropriate, remove the cross-compiling value (which for embedded systems using
uclibc is wrong).
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
This reverts commit 689ea0ec5d8b7594ba2fa9e27b2458cea8a58724.
Community consensus is that, while this style is no longer a
portability issue, it's a style issue.
|
|
This was only supported from GCC 3.0 and OpenBSD was still using GCC 2.95
until recently.
Signed-off-by: Arnaud Fontaine <arnau@debian.org>
|
|
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
sed -i 's/[[:space:]]*$//' *
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
Whilst populating the xorg_cv_ cache entries, and the subsequent
xorg_testset_, we always include the FLAG_PREFIX in the name, to
distinguish between C and C++ iterations. However, in the body of the
test function, we dropped the FLAG_PREFIX expansion and so always used
the empty string and hence did not correctly detect incorrect flags.
The bug was introduced with
commit 51f69514e65637f575df901ef942515e70314f6a
Author: Chase Douglas <chase.douglas@canonical.com>
Date: Mon Mar 12 11:49:28 2012 -0700
Separate unknown warning options by language
Reported-and-tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Sedat Dilek <sedat.dilek@gmail.com>
Cc: Chase Douglas <chase.douglas@canonical.com>
Cc: Gaetan Nadon<memsize@videotron.ca>
Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
http://people.gnome.org/~walters/docs/build-api.txt
Signed-off-by: Adam Jackson <ajax@redhat.com>
|
|
Signed-off-by: Adam Jackson <ajax@redhat.com>
|
|
Signed-off-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
If XORG_COMPILER_FLAGS is called more than once with separate languages,
the unknown warning options used internally for unknown warning checking
will be set the first time and then the cached value will be used for
subsequent languages. This is a problem if the compilers differ between
the languages.
This change ensures that the unknown warning options are namespaced so
multiple XORG_COMPILER_FLAGS calls with different languages are checked
separately.
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Gaetan Nadon<memsize@videotron.ca>
|
|
When the language is C++, the flag checking message references $CC
instead of $CXX. The cache id is also xorg_cv_cc_* instead of
xorg_cv_cxx_*. This change fixes both issues.
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Gaetan Nadon<memsize@videotron.ca>
|
|
Further to commit 52ef6f6, since AS_ECHO isn't supported in autoconf 2.61 either,
rewrite use of cache variable in XORG_TESTSET_CFLAG so 'echo' isn't needed at all.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Gaetan Nadon<memsize@videotron.ca>
Reviewed-by: Chase Douglas <chase.douglas@canonical.com>
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
|
|
This macro enables a builder to enable/disable integration testing
It makes no assumption about the test cases' implementation
Test cases may or may not use Automake "Support for test suites"
Unit test support usually requires less dependencies and may be built
and run under less stringent environments than integration tests. Thus,
two separate options for unit and integration testing is desirable.
Copied from XORG_ENABLE_UNIT_TESTS.
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Gaetan Nadon<memsize@videotron.ca>
|
|
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
You can use this functionality by setting the language to C++ before
calling XORG_DEFAULT_OPTIONS. This can accomplished with AC_LANG(C++).
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
Reviewed-by: Gaetan Nadon<memsize@videotron.ca>
|
|
The AS_ECHO_N macro was only introduced in autoconf 2.62.
Backticks remove trailing newlines, so we don't need to use echo -n
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Since checking all those warning flags takes an amount of time I find irritating,
here is an attempt at caching the results of checking compiler warning flag support
A couple of aspects which need definitely need review or testing:
* I've tried to get things right when not the first flag in the list of alternates is
supported, but this really needs testing with a compiler other than gcc
* The cache variable naming policy is a bit opaque me, so the names used may need corecting
* I've tried to be careful, but who knows what portability sins I have committed :-)
This change reduces the time to run the ./configure script produced from a configure.ac
containing just
AC_INIT([test], 1.0)
XORG_COMPILER_FLAGS
XORG_CWARNFLAGS
XORG_STRICT_OPTION
AC_OUTPUT()
from ~60s to ~20s on my cygwin machine, and from ~15s to ~9s on a linux VM running on the
same hardware
v2: AS_TR_SH transforms characters which are invalid in shell variable names, so we don't
have to use sed to transform spaces ourself
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Provide end users with a workaround if they find code doesn't build with
the -Werror=... flags provided by BASE_CFLAGS. This can be useful if
newer util-macros are used with older source code that had problems, or
in the rare case when warnings are expanded to hit more cases (although
this case should not happen too much in practice).
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
This might be useful for performance tuning builds against specific
compilers, but it is a bit too noisy to be used in general.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
|
|
Catches build problems caused by compilers that ignore unknown flags
at compile time and pass them to the linker at link time, when the
linker considers unknown flags to be fatal errors.
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
This warning was was added to clang in r144365
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Based on the checks already in libXt/configure.ac and proposed for
xserver/configure.ac - makes a common implementation before we make
more copies.
Callers should be able to just add
TESTS_ENVIRONMENT = $(XORG_MALLOC_DEBUG_ENV)
If built with an older xorg-macros, then the tests will simply be
less stringent, matching what happens on platforms without any
simple malloc debugging facilities available.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Otherwise the test fails on Solaris releases that actually have support
for ld -wrap because the linker instead returns an error that __wrap_exit
is an undefined symbol.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Allows specifying if ld -wrap support is 'required' or 'optional' for
unit tests, so that builds like the xserver which only need ld -wrap
for some tests can allow users to enable the others when ld -wrap is
not available.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|
|
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
|