On Mon, Jul 2, 2012 at 1:51 PM, Jordan Rose <[email protected]> wrote:
> > On Jul 2, 2012, at 13:48 , Chandler Carruth <[email protected]> wrote: > > > On Mon, Jul 2, 2012 at 1:45 PM, Jordan Rose <[email protected]> wrote: > >> Yes, that did it. (Sorry, for no feedback; out at lunch.) >> >> On Mon, Jul 2, 2012 at 11:00 AM, Jordan Rose <[email protected]>wrote: >>>> >>>>> This is not a standalone build, but it is a separate build from LLVM. >>>> >>>> >>>> I'm not aware of such a variation? the 'CLANG_BUILT_STANDALONE' is set >>>> explicitly in the cmake files when the Clang tree is built separately from >>>> the LLVM tree. >>>> >>> >> >> Sorry, I thought a standalone build meant "no LLVM sources available", >> since there is a separate case in tests/CMakeLists.txt for that. It looks >> like it's supposed to be possible to build Clang merely with LLVM headers >> and built products, but that's not what I'm doing. >> > > Amusingly, I think that what you're describing *is* the standalone mode... > :: shrug :: > > > Do we need a different terminology for when CLANG_PATH_TO_LLVM_SOURCE is > empty but CLANG_PATH_TO_LLVM_BUILD is present? (Is that even supported? Eh.) > > > Unfortunately, even though "check-clang" builds now in standalone mode, it >> still prints out PASS messages for everything that passes, >> > > Wait wait, I think I've seen that. I can try a few blind commits (or mail > you patches) that might help. > > > Patches accepted. :-) I'll try out what you've got locally first. > It was obvious: r159582. I can't promise that's *enough* but it seems definitely a good initial step. Let me know if that helps.
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
