[dragonegg] r177268 - File names in debug info get an extra layer of indirection since dblaikie's

David Blaikie dblaikie at gmail.com
Mon Mar 18 09:59:17 PDT 2013


On Mon, Mar 18, 2013 at 3:17 AM, Duncan Sands <baldrick at free.fr> wrote:
> Author: baldrick
> Date: Mon Mar 18 05:17:47 2013
> New Revision: 177268
>
> URL: http://llvm.org/viewvc/llvm-project?rev=177268&view=rev
> Log:
> File names in debug info get an extra layer of indirection since dblaikie's
> commit 177239.  Adjust this test correspondingly.
>
> Modified:
>     dragonegg/trunk/test/validator/c++/2010-07-23-DeclLoc.cpp
>
> Modified: dragonegg/trunk/test/validator/c++/2010-07-23-DeclLoc.cpp
> URL: http://llvm.org/viewvc/llvm-project/dragonegg/trunk/test/validator/c%2B%2B/2010-07-23-DeclLoc.cpp?rev=177268&r1=177267&r2=177268&view=diff
> ==============================================================================
> --- dragonegg/trunk/test/validator/c++/2010-07-23-DeclLoc.cpp (original)
> +++ dragonegg/trunk/test/validator/c++/2010-07-23-DeclLoc.cpp Mon Mar 18 05:17:47 2013
> @@ -1,9 +1,10 @@
>  // RUN: %dragonegg -S -g %s -o - | FileCheck %s
>  // Require the template function declaration refer to the correct filename.
>  // First, locate the function decl in metadata, and pluck out the file handle:
> -// CHECK: {{extract_dwarf_data_from_header.*extract_dwarf_data_from_header.*extract_dwarf_data_from_header.*[^ ]+", metadata !}}[[filehandle:[0-9]+]],
> -// Second: Require that filehandle refer to the correct filename:
> -// CHECK: {{^!}}[[filehandle]] = metadata {{![{].*}} metadata !"decl_should_be_here.hpp",
> +// CHECK: {{extract_dwarf_data_from_header.*extract_dwarf_data_from_header.*extract_dwarf_data_from_header.*[^ ]+", metadata !}}[[filetag:[0-9]+]],
> +// Second: Require that filetag refers to metadata holding the correct filename:
> +// CHECK: {{^!}}[[filetag]] = metadata !{i32 786473, metadata ![[filedata:[0-9]+]]}
> +// CHECK: {{^!}}[[filedata]] = metadata !{metadata !"decl_should_be_here.hpp",

Thanks for the fix & sorry for the breakage.

If you like you can check for the IR annotation comments (in
requires:asserts, at least - I don't think we print the comments from
non-asserts builds) that would be a bit more robust against changes in
the schema. I did this to at least one test case in r177238.

- David

>  typedef long unsigned int __darwin_size_t;
>  typedef __darwin_size_t size_t;
>  typedef unsigned char uint8_t;
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits



More information about the llvm-commits mailing list