> > > > Reproduced it again in normal boot with memmap=exactmap, but I agree it's a corner > > case, I remember I saw this in kdump kernel, but I still did not see it in today's > > testing so I think move this issue out of this patch series is fine to me. > > Please include the following oops when you submit this patch again. Sure, will do if I can reproduce it again in kdump kernel. Thanks Dave
