On Fri, Mar 27, 2015 at 5:52 PM, H.J. Lu <hjl.to...@gmail.com> wrote: > On Wed, Mar 18, 2015 at 12:11 PM, H.J. Lu <hjl.to...@gmail.com> wrote: >> On Wed, Mar 18, 2015 at 11:58 AM, Mike Stump <mikest...@comcast.net> wrote: >>> On Mar 18, 2015, at 2:55 AM, Uros Bizjak <ubiz...@gmail.com> wrote: >>>> >>>>> Here is the updated patch with testcases. Tested on Linux/x86. OK >>>>> for trunk? >>> >>>> This patch needs global reviewer approval (I have added Jakub to CC) >>>> and Darwin maintainer approval. >>> >>> So, my concern would be this, does the bug also impact darwin, and does the >>> bug fix also fix darwin? >> >> This bug doesn't impact darwin. >> >>> If no and the change doesn’t change code-gen for darwin (which I think it >>> does not), then the darwin bits are ok. >>> >>> I did a quick check of the test case on darwin, no protected variables (we >>> ignore the protected request). When compiled, the program works (returns >>> 0). >> >> It is expected. This bug only affects targets which support protected >> visibility and use copy relocation. >> > > Hi Jakub, > > I'd like to fix this bug for GCC 5. Is that OK for trunk: > > https://gcc.gnu.org/ml/gcc-patches/2015-03/msg00325.html > > It only impacts Linux/x86.
Jakub, does the patch look OK to you? I am not that familiar with this part of the code ... Uros.
