No :( I tried reverting only r315256 but still got the errors.
On Tue, Oct 10, 2017 at 6:10 PM Richard Smith wrote:
> On 10 Oct 2017 05:41, "Eric Liu via cfe-commits" <
> cfe-commits@lists.llvm.org> wrote:
>
> Hi Richard,
>
> This is breaking the boostrap stage in the internal integration. I'm
>
On 10 Oct 2017 05:41, "Eric Liu via cfe-commits"
wrote:
Hi Richard,
This is breaking the boostrap stage in the internal integration. I'm seeing
"unsupported: typedef changes linkage of anonymous type, but linkage was
already computed" error for many `struct`s defined with typedef. I'm not
sure i
Thanks for reverting.
We saw build failures in Chromium too. E.g. from
https://build.chromium.org/p/chromium.clang/builders/ToTLinux/builds/134:
In file included from
../../third_party/breakpad/breakpad/src/common/linux/symbol_upload.cc:34:
../../third_party/breakpad/breakpad/src/common/linux/sym
FYI, I tweaked the tree in rL315283. Shall we restore it?
On Tue, Oct 10, 2017 at 9:41 PM Eric Liu via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> Hi Richard,
>
> This is breaking the boostrap stage in the internal integration. I'm
> seeing "unsupported: typedef changes linkage of anonymou
Hi Richard,
This is breaking the boostrap stage in the internal integration. I'm seeing
"unsupported: typedef changes linkage of anonymous type, but linkage was
already computed" error for many `struct`s defined with typedef. I'm not
sure if it is user code or clang that needs fixing; however, as
