[smart_ptr] marking shared_ptr_move_test known failure
smart_ptr library test shared_ptr_move_test assumes the compiler support for rvalue references. The only platform participating in HEAD testing it compiles on is gcc-4.3.0_c++0x. I'd like to mark this test known failure for all the platforms in HEAD where it fails to compile because of the lack of the compiler support for rvalue references. Any objections? Thanks, Boris
Boris Gubenko wrote:
smart_ptr library test shared_ptr_move_test assumes the compiler support for rvalue references. The only platform participating in HEAD testing it compiles on is gcc-4.3.0_c++0x.
I'd like to mark this test known failure for all the platforms in HEAD where it fails to compile because of the lack of the compiler support for rvalue references. Any objections?
None. In fact you can even mark it a known failure for any compiler since this is probably going to be the right default for new toolsets for quite a while.
Peter Dimov wrote:
[...] In fact you can even mark it a known failure for any compiler since this is probably going to be the right default for new toolsets for quite a while.
You mean "<toolset name="*"/> so it unexpectedly passes on gcc-4.3.0_c++0x ? Thanks, Boris ----- Original Message ----- From: "Peter Dimov" <pdimov@mmltd.net> To: <boost@lists.boost.org> Cc: "Boris Gubenko" <Boris.Gubenko@hp.com> Sent: Thursday, June 28, 2007 1:11 PM Subject: Re: [boost] [smart_ptr] marking shared_ptr_move_test known failure
Boris Gubenko wrote:
smart_ptr library test shared_ptr_move_test assumes the compiler support for rvalue references. The only platform participating in HEAD testing it compiles on is gcc-4.3.0_c++0x.
I'd like to mark this test known failure for all the platforms in HEAD where it fails to compile because of the lack of the compiler support for rvalue references. Any objections?
None. In fact you can even mark it a known failure for any compiler since this is probably going to be the right default for new toolsets for quite a while.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Boris Gubenko wrote:
Peter Dimov wrote:
[...] In fact you can even mark it a known failure for any compiler since this is probably going to be the right default for new toolsets for quite a while.
You mean "<toolset name="*"/> so it unexpectedly passes on gcc-4.3.0_c++0x ?
Yes, I think that this will be best (unless it inhibits testing entirely). Unfortunately our current infrastructure gives us no easy way to say "only run this test when BOOST_HAS_RVALUE_REFS is defined, else consider it an expected failure". Marking it as expected failure everywhere seems the closest approximation.
Peter Dimov wrote:
Boris Gubenko wrote:
You mean "<toolset name="*"/> so it unexpectedly passes on gcc-4.3.0_c++0x ?
Yes, I think that this will be best (unless it inhibits testing entirely).
Done. Thanks, Boris ----- Original Message ----- From: "Peter Dimov" <pdimov@mmltd.net> To: <boost@lists.boost.org> Cc: "Boris Gubenko" <Boris.Gubenko@hp.com> Sent: Thursday, June 28, 2007 6:38 PM Subject: Re: [boost] [smart_ptr] marking shared_ptr_move_test known failure
Boris Gubenko wrote:
Peter Dimov wrote:
[...] In fact you can even mark it a known failure for any compiler since this is probably going to be the right default for new toolsets for quite a while.
You mean "<toolset name="*"/> so it unexpectedly passes on gcc-4.3.0_c++0x ?
Yes, I think that this will be best (unless it inhibits testing entirely). Unfortunately our current infrastructure gives us no easy way to say "only run this test when BOOST_HAS_RVALUE_REFS is defined, else consider it an expected failure". Marking it as expected failure everywhere seems the closest approximation.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
participants (2)
-
Boris Gubenko -
Peter Dimov