On Mon, 2016-06-06 at 11:17 +1000, David Gibson wrote: > On Fri, Jun 03, 2016 at 02:11:17PM +0200, Cédric Le Goater wrote: > > > > This is follow up to complete the serie "ppc: preparing pnv landing > > (round 2)" plus a little fix on instruction privileges. > > > > Tested on a POWER8 pserie guest and on mac99. > > > > Benjamin Herrenschmidt (2): > > ppc: Fix hreg_store_msr() so that non-HV mode cannot alter MSR:HV > > ppc: Better figure out if processor has HV mode > > > > Cédric Le Goater (1): > > ppc: fix hrfid, tlbia and slbia privilege > I've applied this series to ppc-for-2.7, swapping the order of patches > 1 & 2 as Marc requested. I've also applied the extra patch to fix tlb > flushing on 32-bit cpus. I aim to send a pull request once I've done > my usual tests. > I'm not sure that 32-bit patch is correct. We shouldn't have to flush on IR/DR transitions at all, that's the whole point of the split I/D code.
I think something else is wrong. Cheers, Ben.