This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [libitm] Link with -litm and -pthread


On Tue, 14 Feb 2012, Eric Botcazou wrote:
> > It broke all targets that don't implement threads and as such
> > don't support -pthread.  And you need to gate *all* tm-related
> > tests on something like check_effective_target_pthread.
>
> The question is, how is libitm supposed to work on these systems?

I didn't think about that before since whatever was there
compiled and the tests passed, but right, that's a clue.

>  The
> configure.ac file contains:

Which is incidental, because configure.tgt already declares
non-support and I get:
checking for libitm support... no

Hm, so it seems an oversight in the framework that tm is
enabled for such systems?

> Just define GTM_SELF_SPECS to "" like Darwin and Windows.

That seems to imply that *every* such system should define
GTM_SELF_SPECS (and should have been part of your recent
changes).  But on a second look it seems best to follow the lead
of -fopenmp which implies instead that all tm tests should be
gated on a new check_effective_target_fgnu-tm.  Your ball.

brgds, H-P


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]