llunak added a subscriber: aganea. llunak added a comment. In D69778#2035318 <https://reviews.llvm.org/D69778#2035318>, @dblaikie wrote:
> Do you have a sense of the larger testing that PR44953 was reduced from? Have > you tried compiling a non-trivial codebase (I assume you might've tested it > with Open Office, for instance) - wonder why PR44953 didn't show up there? > (perhaps just the construct needed to tickle the issue isn't common/isn't > common in Open Office code, etc) I've been building LibreOffice and some of its 3rd-party dependencies such as Skia for months with this patch, without problems. PR44953 seems to be a corner case, we do not use D3DX12 headers in LO, and (I presume) __declspec(selectany) is a niche feature. And while I do not understand why PR44953 was happening, I do understand why my original patch was triggering it, and it will not happen with this patch (or, if it will happen, then it'll happen with modules as well). And I've tested with the testcase in PR44953. (It's LibreOffice BTW, OpenOffice is anything but dead.) > and/or maybe see if the person who filed PR44953 might be able to test this > version of the patch on the original codebase that bug was reduced from? @aganea Have you tried how this version of the patch affects PR44953? If not, could you please try? Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69778/new/ https://reviews.llvm.org/D69778 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
