Re: [PATCH] KVM: PPC: Book3S PR: only call slbmte for valid SLB entries

2017-09-26 Thread Michael Ellerman
Paul Mackerras writes: > On Tue, Sep 26, 2017 at 03:24:05PM +1000, Michael Ellerman wrote: >> David Gibson writes: >> >> > On Fri, Sep 22, 2017 at 11:34:29AM +0200, Greg Kurz wrote: >> >> Userland passes an array of 64 SLB descriptors to KVM_SET_SREGS, >> >> some of which are valid (ie, SLB_ESI

Re: [PATCH] KVM: PPC: Book3S PR: only call slbmte for valid SLB entries

2017-09-26 Thread Paul Mackerras
On Tue, Sep 26, 2017 at 03:24:05PM +1000, Michael Ellerman wrote: > David Gibson writes: > > > On Fri, Sep 22, 2017 at 11:34:29AM +0200, Greg Kurz wrote: > >> Userland passes an array of 64 SLB descriptors to KVM_SET_SREGS, > >> some of which are valid (ie, SLB_ESID_V is set) and the rest are > >

Re: [PATCH] KVM: PPC: Book3S PR: only call slbmte for valid SLB entries

2017-09-25 Thread Michael Ellerman
David Gibson writes: > On Fri, Sep 22, 2017 at 11:34:29AM +0200, Greg Kurz wrote: >> Userland passes an array of 64 SLB descriptors to KVM_SET_SREGS, >> some of which are valid (ie, SLB_ESID_V is set) and the rest are >> likely all-zeroes (with QEMU at least). >> >> Each of them is then passed t

Re: [PATCH] KVM: PPC: Book3S PR: only call slbmte for valid SLB entries

2017-09-25 Thread David Gibson
On Fri, Sep 22, 2017 at 11:34:29AM +0200, Greg Kurz wrote: > Userland passes an array of 64 SLB descriptors to KVM_SET_SREGS, > some of which are valid (ie, SLB_ESID_V is set) and the rest are > likely all-zeroes (with QEMU at least). > > Each of them is then passed to kvmppc_mmu_book3s_64_slbmte(