diff mbox

Tweak tm/pr47573.C testcase (PR c++/64884)

Message ID 20150212212539.GF1746@tucnak.redhat.com
State New
Headers show

Commit Message

Jakub Jelinek Feb. 12, 2015, 9:25 p.m. UTC
Hi!

This patch tweaks the pr47573.C testcase, so that it is run only on
comdat_group target.  It is true that comdat_group is stricter than
comdat, but we don't have an effective-target for the latter and I'm not
sure what exactly to test for.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

2015-02-12  Jakub Jelinek  <jakub@redhat.com>

	PR c++/64884
	* g++.dg/tm/pr47573.C: Only run on comdat_group effective targets.


	Jakub

Comments

Richard Henderson Feb. 12, 2015, 9:31 p.m. UTC | #1
On 02/12/2015 01:25 PM, Jakub Jelinek wrote:
> Hi!
> 
> This patch tweaks the pr47573.C testcase, so that it is run only on
> comdat_group target.  It is true that comdat_group is stricter than
> comdat, but we don't have an effective-target for the latter and I'm not
> sure what exactly to test for.
> 
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
> 
> 2015-02-12  Jakub Jelinek  <jakub@redhat.com>
> 
> 	PR c++/64884
> 	* g++.dg/tm/pr47573.C: Only run on comdat_group effective targets.

Yes, thanks.  I don't see what else we can test for either,
and it's entirely likely the set of comdat supporters that
don't also support comdat_group is probably empty anyway.


r~
diff mbox

Patch

--- gcc/testsuite/g++.dg/tm/pr47573.C.jj	2011-11-08 23:35:11.000000000 +0100
+++ gcc/testsuite/g++.dg/tm/pr47573.C	2015-02-12 14:26:23.873151135 +0100
@@ -1,4 +1,7 @@ 
-// { dg-do compile }
+// Without comdat support, we don't see the body of the
+// extern template class constructor, so limit this to
+// known comdat targets.
+// { dg-do compile { target comdat_group } }
 // { dg-options "-fgnu-tm" }
 
 template<typename _Tp> class allocator