[PATCH xserver] test: create a link to the generated hw/xfree86/sdksyms.c at build time

[Gaetan Nadon]

On 14-02-24 05:27 PM, Keith Packard wrote:
> Gaetan Nadon <memsize at videotron.ca> writes:
>
>> Note that we do not "need" the subdir-objects feature, but automake 2.0
>> will turn it on any (and it cannot be turned off). We happen to have
>> code that will trip this feature. So we need to fix it before 2.0 gets
>> pervasive.
> Sure would be nice if automake 2.0 had a compatibility mode that would
> let us continue with our current scheme. Has anyone looked at the 2.0
> code to see if there's a way we can avoid changing how we build things
> just to make autofu happy?
>
Does not look like there is any new code in git.
http://git.savannah.gnu.org/cgit/automake.git

Being a "2.0" version, some changes will not be backwards compatible.
According to their "PLANS", the way it works now is a "historical
accident" and they are determined to correcting it.

>From the PLANS/subdir-objs.txt file from Automake 1.14 package:

    The fact that Automake-generated Makefiles place compiled object files in
    the current directory by default, also when the corresponding source file
    is in a subdirectory, is basically an historical accident, due to the fact
    that the 'subdir-objects' option had only been introduced in April 1999,
    starting with commit 'user-dep-gen-branchpoint-56-g88b5959', and never
    made the default (likely to avoid backwards-compatibility issues).

    Since I believe the behaviour enabled by the 'subdir-objects' is the most
    useful one, and in fact the *only* natural one, I'd like to make it the
    only one available, simplifying the Automake implementation and APIs a
    little in the process.

I originally submitted a patch series which fixes all of the makefiles
using git links. It's not the best solution, but it shows it is a
reasonable endeavour.

Reviewers asked for a better solution which I couldn't provide alone.
Jon Turney responded with patches (already reviewed) which fixes most of
the DDX code reuse case. That leaves miinitext for which Daniel Stone
provided a hint. This is the hardest one.

Mark Kettenis suggested how os-support source can be organized and looks
fairly simple. I provided a couple patches to fix the test makefile and
one of them has already landed in master.

In my opinion, this is energy well spent. Time is on our side (rarely
happens) as the issue has been detected early thanks to warnings in
1.14. The reusable source code will be better organized, which is not a
negligible benefit.




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.x.org/archives/xorg-devel/attachments/20140224/84614317/attachment.html>