On Wed, May 9, 2012 at 12:40 PM, Chad Rosier <[email protected]> wrote:
> > On May 9, 2012, at 11:30 AM, Chandler Carruth wrote: > > On Wed, May 9, 2012 at 11:29 AM, Chandler Carruth <[email protected]>wrote: > >> >> >> 2012/5/9 Chad Rosier <[email protected]> >> >>> Author: mcrosier >>> Date: Wed May 9 13:23:04 2012 >>> New Revision: 156493 >>> >>> URL: http://llvm.org/viewvc/llvm-project?rev=156493&view=rev >>> Log: >>> Update test case for new default of 3.0 >>> >>> Modified: >>> cfe/trunk/test/Driver/target-triple-deployment.c >>> >> >> Maybe we shouldn't check the version numbers here in specificity, and >> rely on the llvm::Triple tests? >> > > To be more clear -- should we not check *which* default version we get, > and just check that we did get a (non-zero) default version? I understand > we want to make sure that the llvm::Triple is in fact being used here. > > > Hi Chandler, > I've added the Triple unittests in llvm revision 156507. > Awesome. > Are you opposed to leaving this test case in place to ensure the driver > is working as expected? > Not at all, i like having this test. I was only wondering if we wanted to not hard code which version happens to be the default. But honestly, it's not a big deal either way. =] Sorry for explaining things poorly and wasting your time.
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
