On 6 Apr 2018, at 21:00, Oliver Pinter <oliver.pin...@hardenedbsd.org> wrote: > > On 4/6/18, John Baldwin <j...@freebsd.org> wrote: >> On Monday, April 02, 2018 12:27:47 PM Ed Maste wrote: >>> On 31 March 2018 at 14:41, Mark Linimon <lini...@lonesome.com> wrote: >>> This is the most important point of this discussion: we do need to >>> ensure there's good communication and coordination between teams where >>> dependencies like this exist. I'll take the blame here: Dimitry asked >>> me about merging the Clang update to stable/11 and I agreed that it >>> was reasonable to merge sooner rather than later to have as much lead >>> time as possible before the 11.2 process starts. I also assumed that >>> outstanding Clang 6 issues in ports were farther along in being >>> addressed. >>> >>> The key lesson from this discussion is that for significant commits >>> and merges like this one we should make sure to always have sufficient >>> advance notice. >> >> Is this driven by -mretpoline? > > Don't think so since it requires LLD as linker.
You can use lld on 11, but it isn't the default. > The LLVM 5 which was already > part of the 11-stable has the retpoline option. See r331219. Indeed, r331219 contains backports (from upstream) for llvm, clang *and* lld. If you want to make use of -mretpoline, you should also pass -fuse-ld=lld on the command line. > From other side, I like to see a newer compiler sets in the recent > releases, because > there are lot of performance improvements and other fixes. That is certainly the case, the flip side is that there are always some new bugs, regressions and newly added warnings. Having the default C++ dialect bumped is really an anomaly, this won't happen every upstream release. :) -Dimitry
Description: Message signed with OpenPGP