On Wed, Feb 20, 2019 at 02:47:05PM +0000, Will Deacon wrote: > On Tue, Feb 19, 2019 at 02:41:47PM +0100, Peter Zijlstra wrote: > > On Tue, Feb 19, 2019 at 12:47:38PM +0000, Will Deacon wrote: > > > Fine for now, but I agree that we should drop the hook altogether. AFAICT, > > > this only exists to help an ia64 optimisation which looks suspicious to > > > me since it uses: > > > > > > mm == current->active_mm && atomic_read(&mm->mm_users) == 1 > > > > > > to identify a "single-threaded fork()" and therefore perform only local TLB > > > invalidation. Even if this was the right thing to do, it's not clear to me > > > that tlb_migrate_finish() is called on the right CPU anyway. > > > > > > So I'd be keen to remove this hook before it spreads, but in the meantime: > > > > Agreed :-) > > > > The obvious slash and kill patch ... untested > > I'm also unable to test this, unfortunately. Can we get it into next after > the merge window and see if anybody reports issues? While I do have a pair of Itanium systems in my basement, neither are sn2 machines, which was the only sub-architecture that implemented tlb_migrate_finish(). I see NASA decomissioned Columbia in 2013, and I imagine most sn2 machines have been similarly scrapped.
