| Message ID | 20170217214140.GA8844@fergus.ozlabs.ibm.com |
|---|---|
| State | Accepted |
| Headers | show |
On Sat, Feb 18, 2017 at 08:41:40AM +1100, Paul Mackerras wrote: > The new HPT resizing code added in commit b5baa6877315 ("KVM: PPC: > Book3S HV: KVM-HV HPT resizing implementation", 2016-12-20) doesn't > have code to handle the new HPTE format which POWER9 uses. Thus it > would be best not to advertise it to userspace on POWER9 systems > until it works properly. > > Also, since resize_hpt_rehash_hpte() contains BUG_ON() calls that > could be hit on POWER9, let's prevent it from being called on POWER9 > for now. > > Signed-off-by: Paul Mackerras <paulus@ozlabs.org> Acked-by: David Gibson <david@gibson.dropbear.id.au> I probably would have put the check on entry to the prepare path, rather than deep in the rehashing, but we need the fix one way or another. > --- > arch/powerpc/kvm/book3s_64_mmu_hv.c | 6 ++++++ > arch/powerpc/kvm/powerpc.c | 3 ++- > 2 files changed, 8 insertions(+), 1 deletion(-) > > diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c > index b68b342..f3158fb 100644 > --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c > +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c > @@ -1370,6 +1370,12 @@ static int resize_hpt_rehash(struct kvm_resize_hpt *resize) > unsigned long i; > int rc; > > + /* > + * resize_hpt_rehash_hpte() doesn't handle the new-format HPTEs > + * that POWER9 uses, and could well hit a BUG_ON on POWER9. > + */ > + if (cpu_has_feature(CPU_FTR_ARCH_300)) > + return -EIO; > for (i = 0; i < kvmppc_hpt_npte(&kvm->arch.hpt); i++) { > rc = resize_hpt_rehash_hpte(resize, i); > if (rc != 0) > diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c > index 2b3e4e6..fcb253b 100644 > --- a/arch/powerpc/kvm/powerpc.c > +++ b/arch/powerpc/kvm/powerpc.c > @@ -613,7 +613,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) > r = 1; > break; > case KVM_CAP_SPAPR_RESIZE_HPT: > - r = !!hv_enabled; > + /* Disable this on POWER9 until code handles new HPTE format */ > + r = !!hv_enabled && !cpu_has_feature(CPU_FTR_ARCH_300); > break; > #endif > case KVM_CAP_PPC_HTM:
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index b68b342..f3158fb 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -1370,6 +1370,12 @@ static int resize_hpt_rehash(struct kvm_resize_hpt *resize) unsigned long i; int rc; + /* + * resize_hpt_rehash_hpte() doesn't handle the new-format HPTEs + * that POWER9 uses, and could well hit a BUG_ON on POWER9. + */ + if (cpu_has_feature(CPU_FTR_ARCH_300)) + return -EIO; for (i = 0; i < kvmppc_hpt_npte(&kvm->arch.hpt); i++) { rc = resize_hpt_rehash_hpte(resize, i); if (rc != 0) diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 2b3e4e6..fcb253b 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -613,7 +613,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) r = 1; break; case KVM_CAP_SPAPR_RESIZE_HPT: - r = !!hv_enabled; + /* Disable this on POWER9 until code handles new HPTE format */ + r = !!hv_enabled && !cpu_has_feature(CPU_FTR_ARCH_300); break; #endif case KVM_CAP_PPC_HTM:
The new HPT resizing code added in commit b5baa6877315 ("KVM: PPC: Book3S HV: KVM-HV HPT resizing implementation", 2016-12-20) doesn't have code to handle the new HPTE format which POWER9 uses. Thus it would be best not to advertise it to userspace on POWER9 systems until it works properly. Also, since resize_hpt_rehash_hpte() contains BUG_ON() calls that could be hit on POWER9, let's prevent it from being called on POWER9 for now. Signed-off-by: Paul Mackerras <paulus@ozlabs.org> --- arch/powerpc/kvm/book3s_64_mmu_hv.c | 6 ++++++ arch/powerpc/kvm/powerpc.c | 3 ++- 2 files changed, 8 insertions(+), 1 deletion(-)