On Mar 12, 2013, at 1:49 PM, Eric Christopher <[email protected]> wrote:
> > > +// RUN: %clang_cc1 -g -emit-obj -o %t %s > +// RUN: llvm-dwarfdump %t | FileCheck %s > +// In the attached test case a post-r166236 clang coalesces two > +// instances of an inlined function in a way that makes it appear as > +// if the function was only inlined once. > + > > Two things: > > a) this test should only check for IR emitted, there should be a backend test > that determines whether or not the correct debug info was generated from IR. > Please split this into two testcases. > b) update the comment that explicitly calls out whether or not we generate > column info rather than a revision in svn. revision numbers aren't > particularly useful since future revisions can change behavior. Good point. I’m fixing this right now! thanks, adrian _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
