On Thu, 22 Jul 2021, Alan Modra wrote: > On Wed, Jul 21, 2021 at 08:59:04AM -0400, David Edelsohn wrote: > > On Wed, Jul 21, 2021 at 4:29 AM Alan Modra <amo...@gmail.com> wrote: > > > > > > On Wed, Jul 14, 2021 at 08:24:16PM -0400, David Edelsohn wrote: > > > > > > > * config/rs6000/morestack.S (R2_SAVE): Define. > > > > > > > (__morestack): Save and restore r2. Set up r2 for called > > > > > > > functions. > > > > > > > > This patch is okay. > > > > > > Thanks David, the patch is needed on gcc-11 and gcc-10 too. > > > OK for the branches too? > > > > Backports are fine, but I believe that Richi is planning to cut GCC 11 > > RC today, so you really should check with him about a backport at the > > last minute. > > Hi Richard, > Is this patch OK at this late stage for the gcc-11 branch? > https://gcc.gnu.org/pipermail/gcc-patches/2021-June/573978.html > > The impacts of the bug are segfaults and other undesirable behaviour > with Go (or more generally -fsplit-stack) on power10 when libgcc is > not power10 pcrel. A non-pcrel libgcc is very likely how distros > will ship gcc.
If you think it's safe (well, there are not many __morestack users, so based on that it's pretty safe) then go ahead. Richard.
