Re: [FYI] {maint} news: Automake 1.13 will require Autoconf >= 2.65

On Mon, Apr 30, 2012 at 19:45, Stefano Lattarini wrote: > +  - Autoconf 2.65 or later will be required by the next major Automake > +    version (1.13).  Until now, Automake has been working with any > +    Autoconf version up to 2.62. > + Presumably you meant to write "Until now, Automake has re

[FYI] {maint} fixup: remove leftover "TERM=ansi" definition in 'color.test'

It should have actually be removed in commit 'v1.12-23-gb105d40'. Signed-off-by: Stefano Lattarini --- t/color.sh |3 --- 1 file changed, 3 deletions(-) diff --git a/t/color.sh b/t/color.sh index c9fe302..c9a1ab9 100755 --- a/t/color.sh +++ b/t/color.sh @@ -19,9 +19,6 @@ . ./defs || Exit

[PATCH 4/3] parallel-tests: save few forks when possible

* lib/am/check.am (am__check_pre): Save some forks, by analyzing more carefully the value of '$@' to decide how to extract its dirname part, and whether that should be created as a directory. Signed-off-by: Stefano Lattarini --- And this trivial optimization could be squashed in in this series

[FYI] Merge branch 'maint' into master

commit de9e3447ddbaa5488d0d695611eeaa880b09a035 Merge: 1564671 99dc7ac Author: Stefano Lattarini Date: Mon Apr 30 22:07:15 2012 +0200 Merge branch 'maint' * maint: news: Automake 1.13 will require Autoconf >= 2.65 news: put planned future backward-incompatibilities on top

[FYI] {maint} news: Automake 1.13 will require Autoconf >= 2.65

Autoconf 2.65 is almost three years old now, so requiring it is acceptable. Also, the soon-to-appear support for Objective C++ in Automake will be simplified if we can assume that Autoconf version (which is the one that introduced support for Objective C++). * NEWS (Planned future backward-incomp

[FYI] {maint} news: put planned future backward-incompatibilities on top

* NEWS (New in 1.12): List of planned future backward-incompatibilities moved from here ... (New in 1.12.1): ... to here. Signed-off-by: Stefano Lattarini --- NEWS | 40 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/NEWS b/NEWS index e

[FYI] {maint} readme: update INSTALL to latest version from Autoconf

* INSTALL: Updated to the version coming with Autoconf 2.69. This is warranted now that we use and require that same Autoconf version in our build system. Signed-off-by: Stefano Lattarini --- INSTALL |8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/INSTALL b/INSTAL

[FYI] {maint} sync: some auxiliary files synced from upstream

* lib/config.sub: Synced from upstream, by "make fetch". Signed-off-by: Stefano Lattarini --- lib/config.sub |8 +++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/lib/config.sub b/lib/config.sub index c894da4..59bb593 100755 --- a/lib/config.sub +++ b/lib/config.sub @@ -4

[FYI] {maint} hacking: copyright notices must be updated before releasing

* HACKING (Release procedure): Suggest to use "make update-copyright". Signed-off-by: Stefano Lattarini --- HACKING |4 1 file changed, 4 insertions(+) diff --git a/HACKING b/HACKING index e3153cc..7ab4e7f 100644 --- a/HACKING +++ b/HACKING @@ -223,6 +223,10 @@ running "make fetch".

[FYI] {maint} color-tests: coloring can be forced on non-ANSI terminals as well

Before this change, colorization of testsuite output was suppressed whenever the terminal was recognized to be a "dumb" one, incapable of handling ANSI coloring (i.e., when the environment variable TERM had a value of "dumb"). This happened even when the AM_COLOR_TESTS variable was set to a value

[PATCH 2/2] cygnus: remove support for Cygnus-style trees

Support for "Cygnus-style" tree as enabled by the 'cygnus' option has been deprecated in the documentation and with runtime warning since Automake 1.12.1. It's now time to remove it. Closes automake bug#11034. * lib/Automake/Options.pm (_process_option_list): Issue a proper error when the 'cygnu

[PATCH 1/2] cygnus: warn when it's used (flagged as 'obsolete' now)

Support for "Cygnus-style" tree as enabled by the 'cygnus' option is now reported as obsolete when the warnings in the 'obsolete' category are enabled. See automake bug#11034. * automake.in (check_cygnus): Report the use of 'cygnus' mode as obsolete. * NEWS: Update. * docs/automake.texi: Minor ad

[PATCH 0/2] cygnus: deprecate and remove

The first patch deprecates support for Cygnus style-trees. It is meant to be applied to the 'maint' branch, so that it will appear in the next Automake maintenance version (1.12.1) The second patch removes support for Cygnus style-trees altogether. It is meant to be applied (on the top of the fi

Re: [PATCH 0/3] Optimize some parallel-tests recipe

On 04/30/2012 05:54 PM, Stefano Lattarini wrote: > I will push this to maint in a couple of days if there is no objection. > Reviews are most welcome. > > Regards, > Stefano > > -*-*-*- > > Stefano Lattarini (3): > coverage: add performance tests on some parallel-tests aspects > parallel-t

[PATCH 3/3] parallel-tests: optimize global log creation

With this change, the time required to execute the test case 'testsuite-summary-speed.sh' has dropped as follows: + Slow Debian i686 system, 1 core x 1.5 GHz, 768 MB of RAM, GNU make 3.81: - 1 run: 4 minutes => 30/40 seconds + Fast Solaris 10 i686 system, 4 cores x 3 GHz, 20 GB of R

[PATCH 1/3] coverage: add performance tests on some parallel-tests aspects

* t/testsuite-recheck-speed.sh: New test. * t/testsuite-summary-speed.sh: Likewise. * t/list-of-tests.mk: Update. The new tests are not meant to PASS/FAIL, but rather to help us to obtain quantitative measurements of the performance improvements offered by soon-to-appear optimization patches. Tim

[PATCH 2/3] parallel-tests: optimize 'recheck' target for speed

With this change, the time required to execute the test case 'testsuite-recheck-speed.sh' has dropped as follows: + Slow Debian i686 system, 1 core x 1.5 GHz, 768 MB of RAM, GNU make 3.81: - 1 run: 6 minutes => 40 seconds + Fast Solaris 10 i686 system, 4 cores x 3 GHz, 20 GB of RAM,

[PATCH 0/3] Optimize some parallel-tests recipe

I will push this to maint in a couple of days if there is no objection. Reviews are most welcome. Regards, Stefano -*-*-*- Stefano Lattarini (3): coverage: add performance tests on some parallel-tests aspects parallel-tests: optimize 'recheck' target for speed parallel-tests: optimize gl

Re: [PATCH 5/6] objc++: test more automatic dependency tracking

On 04/30/2012 03:31 PM, Peter Breitenlohner wrote: > On Mon, 30 Apr 2012, Stefano Lattarini wrote: > >> Hi Peter, thanks for the patch. However, I've opted for a more >> "semantic" approach, implemented by the attached test case. > > Hi Stefano, > > fine with me (except for a typo s/autocon,/au

Re: [PATCH 3/5] objc++: test support for '.mm' suffix in _SOURCES enrries

On 04/27/2012 10:19 AM, Stefano Lattarini wrote: > From: Peter Breitenlohner > > * t/ext.sh: Enhance by adding a file with the '.mm' suffix in > the 'foo_SOURCES' definition. > * t/nodep2.sh: Likewise. > > Signed-off-by: Peter Breitenlohner > Signed-off-by: Stefano Lattarini > Oops, this patch

Re: [PATCH 5/6] objc++: test more automatic dependency tracking

On Mon, 30 Apr 2012, Stefano Lattarini wrote: Hi Peter, thanks for the patch. However, I've opted for a more "semantic" approach, implemented by the attached test case. Hi Stefano, fine with me (except for a typo s/autocon,/autoconf,/. But please note that these tests will fail if gcc/g++ d

Re: [PATCH 3/6] objc++: test automatic dependency tracking

Hi Peter, thanks for the patch. However, I've opted for a more "semantic" approach, implemented by the test case attached to the patch [5/6] later in this series. Regards, Stefano

Re: [PATCH 4/6] objc: test automatic dependency tracking

Hi Peter, thanks for the patch. However, I've opted for a more "semantic" approach, implemented by the test case attached to the patch [6/6] later in this series. Regards, Stefano

Re: [PATCH 5/6] objc++: test more automatic dependency tracking

Hi Peter, thanks for the patch. However, I've opted for a more "semantic" approach, implemented by the attached test case. Regards, Stefano objcxx-deps.sh Description: Bourne shell script

Re: [PATCH 6/6] objc: test more automatic dependency tracking

Hi Peter, thanks for the patch. However, I've opted for a more "semantic" approach, implemented by the attached test case. Regards, Stefano objc-deps.sh Description: Bourne shell script

Re: [PATCH 1/5] objc++: initial support for Objective C++

On 04/30/2012 01:11 PM, Peter Breitenlohner wrote: > On Mon, 30 Apr 2012, Stefano Lattarini wrote: > >> Unfortunately, this patch breaks automake with older autoconfs (e.g., 2.62), >> even when the Objective C++ is not used. For example: >> ... >> This is due to the unconditional use of 'AC_PRO

Re: [PATCH] Skip the test when Autoconf is too old for ObjC++

Hi Peter. Since the test 't/objcxx-basic.sh' was indeed failing with autoconf 2.62, I've folded in the following diff, inspired by your patch: diff --git a/t/objcxx-basic.sh b/t/objcxx-basic.sh index 50e1399..04a42a4 100755 --- a/t/objcxx-basic.sh +++ b/t/objcxx-basic.sh @@ -26,10 +26,2

Re: [PATCH 1/5] objc++: initial support for Objective C++

On Mon, 30 Apr 2012, Stefano Lattarini wrote: Unfortunately, this patch breaks automake with older autoconfs (e.g., 2.62), even when the Objective C++ is not used. For example: ... This is due to the unconditional use of 'AC_PROG_OBJCXX' in AM_INIT_AUTOMAKE. I've squashed in the diff below,

Re: [PATCH] Skip the test when Autoconf is too old for ObjC++

On 04/28/2012 03:35 PM, Peter Breitenlohner wrote: > Signed-off-by: Peter Breitenlohner > --- > t/objcxx-demo.sh | 12 ++-- > 1 files changed, 10 insertions(+), 2 deletions(-) > I've folded in a slightly modified version of this patch. Thanks, Stefano

Re: [PATCH 1/5] objc++: initial support for Objective C++

Hi Peter, sorry for the delay. On 04/27/2012 10:19 AM, Stefano Lattarini wrote: > From: Peter Breitenlohner > > Original thread (dating back to almost three years ago): > > > * automake.in: Register new language 'objcxx'

[PATCH 5/6] objc++: test more automatic dependency tracking

* t/objcxx-deps2.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objcxx-deps2.sh | 58 + 1 files changed, 58 insertions(+), 0 deletions(-) create mode 100755 t/objcxx-deps2.sh diff --git a/t/objcxx-deps2.sh b/t/objcxx-deps2.sh new fi

[PATCH 6/6] objc: test more automatic dependency tracking

* t/objc-deps2.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objc-deps2.sh | 49 + 1 files changed, 49 insertions(+), 0 deletions(-) create mode 100755 t/objc-deps2.sh diff --git a/t/objc-deps2.sh b/t/objc-deps2.sh new file mode 100755

[PATCH 1/6] objc++: test global, per-target, and user flags

* t/objcxx-flags.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objcxx-flags.sh | 82 + 1 files changed, 82 insertions(+), 0 deletions(-) create mode 100755 t/objcxx-flags.sh diff --git a/t/objcxx-flags.sh b/t/objcxx-flags.sh new fi

[PATCH 4/6] objc: test automatic dependency tracking

* t/objc-deps.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objc-deps.sh | 37 + 1 files changed, 37 insertions(+), 0 deletions(-) create mode 100755 t/objc-deps.sh diff --git a/t/objc-deps.sh b/t/objc-deps.sh new file mode 100755 index 000..8

[PATCH 3/6] objc++: test automatic dependency tracking

* t/objcxx-deps.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objcxx-deps.sh | 46 ++ 1 files changed, 46 insertions(+), 0 deletions(-) create mode 100755 t/objcxx-deps.sh diff --git a/t/objcxx-deps.sh b/t/objcxx-deps.sh new file mode 1007

[PATCH 2/6] objc: test global, per-target, and user flags

* t/objc-flags.sh: New test. Signed-off-by: Peter Breitenlohner --- t/objc-flags.sh | 73 +++ 1 files changed, 73 insertions(+), 0 deletions(-) create mode 100755 t/objc-flags.sh diff --git a/t/objc-flags.sh b/t/objc-flags.sh new file mode

objc++: more tests (also for objc)

Hi Stefano, here the additional tests for Objective C++ you have asked for. While at it I also have created analogous tests for Objective C. [PATCH 1/6] objc++: test global, per-target, and user flags: inspired lflags.sh. [PATCH 2/6] objc: test global, per-target, and user flags: likewise. [PAT