On Mon, Dec 25, 2017 at 8:16 PM, Martin Husemann <mar...@duskware.de> wrote: > On Mon, Dec 25, 2017 at 12:05:28PM +0900, Ryota Ozaki wrote: >> Anyway I annoy that we often cannot have suspect commits because of cvs >> (you know the kernel version doesn't work at all for the purpose). I wish >> we had revision IDs of svn or commit IDs of git to know them. > > The sparc64 hang happened with: > > $NetBSD: ip_output.c,v 1.288 2017/12/22 11:22:37 ozaki-r Exp $
Thanks. Hmm, I have no idea why and how the change hangs the system. If it's a deadlock can you run ps in DDB and gather threads stuck on turnstile? If not, ...I have no idea :-/ ozaki-r