linux/arch/powerpc/kvm/book3s_64_mmu.c

669 lines
16 KiB
C
Raw Normal View History

/*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, version 2, as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*
* Copyright SUSE Linux Products GmbH 2009
*
* Authors: Alexander Graf <agraf@suse.de>
*/
#include <linux/types.h>
#include <linux/string.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <linux/highmem.h>
#include <asm/tlbflush.h>
#include <asm/kvm_ppc.h>
#include <asm/kvm_book3s.h>
#include <asm/mmu-hash64.h>
/* #define DEBUG_MMU */
#ifdef DEBUG_MMU
#define dprintk(X...) printk(KERN_INFO X)
#else
#define dprintk(X...) do { } while(0)
#endif
static void kvmppc_mmu_book3s_64_reset_msr(struct kvm_vcpu *vcpu)
{
kvmppc_set_msr(vcpu, MSR_SF);
}
static struct kvmppc_slb *kvmppc_mmu_book3s_64_find_slbe(
struct kvm_vcpu *vcpu,
gva_t eaddr)
{
int i;
u64 esid = GET_ESID(eaddr);
u64 esid_1t = GET_ESID_1T(eaddr);
for (i = 0; i < vcpu->arch.slb_nr; i++) {
u64 cmp_esid = esid;
if (!vcpu->arch.slb[i].valid)
continue;
if (vcpu->arch.slb[i].tb)
cmp_esid = esid_1t;
if (vcpu->arch.slb[i].esid == cmp_esid)
return &vcpu->arch.slb[i];
}
dprintk("KVM: No SLB entry found for 0x%lx [%llx | %llx]\n",
eaddr, esid, esid_1t);
for (i = 0; i < vcpu->arch.slb_nr; i++) {
if (vcpu->arch.slb[i].vsid)
dprintk(" %d: %c%c%c %llx %llx\n", i,
vcpu->arch.slb[i].valid ? 'v' : ' ',
vcpu->arch.slb[i].large ? 'l' : ' ',
vcpu->arch.slb[i].tb ? 't' : ' ',
vcpu->arch.slb[i].esid,
vcpu->arch.slb[i].vsid);
}
return NULL;
}
static int kvmppc_slb_sid_shift(struct kvmppc_slb *slbe)
{
return slbe->tb ? SID_SHIFT_1T : SID_SHIFT;
}
static u64 kvmppc_slb_offset_mask(struct kvmppc_slb *slbe)
{
return (1ul << kvmppc_slb_sid_shift(slbe)) - 1;
}
static u64 kvmppc_slb_calc_vpn(struct kvmppc_slb *slb, gva_t eaddr)
{
eaddr &= kvmppc_slb_offset_mask(slb);
return (eaddr >> VPN_SHIFT) |
((slb->vsid) << (kvmppc_slb_sid_shift(slb) - VPN_SHIFT));
}
static u64 kvmppc_mmu_book3s_64_ea_to_vp(struct kvm_vcpu *vcpu, gva_t eaddr,
bool data)
{
struct kvmppc_slb *slb;
slb = kvmppc_mmu_book3s_64_find_slbe(vcpu, eaddr);
if (!slb)
return 0;
return kvmppc_slb_calc_vpn(slb, eaddr);
}
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
static int mmu_pagesize(int mmu_pg)
{
switch (mmu_pg) {
case MMU_PAGE_64K:
return 16;
case MMU_PAGE_16M:
return 24;
}
return 12;
}
static int kvmppc_mmu_book3s_64_get_pagesize(struct kvmppc_slb *slbe)
{
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
return mmu_pagesize(slbe->base_page_size);
}
static u32 kvmppc_mmu_book3s_64_get_page(struct kvmppc_slb *slbe, gva_t eaddr)
{
int p = kvmppc_mmu_book3s_64_get_pagesize(slbe);
return ((eaddr & kvmppc_slb_offset_mask(slbe)) >> p);
}
static hva_t kvmppc_mmu_book3s_64_get_pteg(struct kvm_vcpu *vcpu,
struct kvmppc_slb *slbe, gva_t eaddr,
bool second)
{
struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu);
u64 hash, pteg, htabsize;
u32 ssize;
hva_t r;
u64 vpn;
htabsize = ((1 << ((vcpu_book3s->sdr1 & 0x1f) + 11)) - 1);
vpn = kvmppc_slb_calc_vpn(slbe, eaddr);
ssize = slbe->tb ? MMU_SEGSIZE_1T : MMU_SEGSIZE_256M;
hash = hpt_hash(vpn, kvmppc_mmu_book3s_64_get_pagesize(slbe), ssize);
if (second)
hash = ~hash;
hash &= ((1ULL << 39ULL) - 1ULL);
hash &= htabsize;
hash <<= 7ULL;
pteg = vcpu_book3s->sdr1 & 0xfffffffffffc0000ULL;
pteg |= hash;
dprintk("MMU: page=0x%x sdr1=0x%llx pteg=0x%llx vsid=0x%llx\n",
page, vcpu_book3s->sdr1, pteg, slbe->vsid);
/* When running a PAPR guest, SDR1 contains a HVA address instead
of a GPA */
if (vcpu->arch.papr_enabled)
r = pteg;
else
r = gfn_to_hva(vcpu->kvm, pteg >> PAGE_SHIFT);
if (kvm_is_error_hva(r))
return r;
return r | (pteg & ~PAGE_MASK);
}
static u64 kvmppc_mmu_book3s_64_get_avpn(struct kvmppc_slb *slbe, gva_t eaddr)
{
int p = kvmppc_mmu_book3s_64_get_pagesize(slbe);
u64 avpn;
avpn = kvmppc_mmu_book3s_64_get_page(slbe, eaddr);
avpn |= slbe->vsid << (kvmppc_slb_sid_shift(slbe) - p);
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
if (p < 16)
avpn >>= ((80 - p) - 56) - 8; /* 16 - p */
else
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
avpn <<= p - 16;
return avpn;
}
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
/*
* Return page size encoded in the second word of a HPTE, or
* -1 for an invalid encoding for the base page size indicated by
* the SLB entry. This doesn't handle mixed pagesize segments yet.
*/
static int decode_pagesize(struct kvmppc_slb *slbe, u64 r)
{
switch (slbe->base_page_size) {
case MMU_PAGE_64K:
if ((r & 0xf000) == 0x1000)
return MMU_PAGE_64K;
break;
case MMU_PAGE_16M:
if ((r & 0xff000) == 0)
return MMU_PAGE_16M;
break;
}
return -1;
}
static int kvmppc_mmu_book3s_64_xlate(struct kvm_vcpu *vcpu, gva_t eaddr,
KVM: PPC: Book3S PR: Better handling of host-side read-only pages Currently we request write access to all pages that get mapped into the guest, even if the guest is only loading from the page. This reduces the effectiveness of KSM because it means that we unshare every page we access. Also, we always set the changed (C) bit in the guest HPTE if it allows writing, even for a guest load. This fixes both these problems. We pass an 'iswrite' flag to the mmu.xlate() functions and to kvmppc_mmu_map_page() to indicate whether the access is a load or a store. The mmu.xlate() functions now only set C for stores. kvmppc_gfn_to_pfn() now calls gfn_to_pfn_prot() instead of gfn_to_pfn() so that it can indicate whether we need write access to the page, and get back a 'writable' flag to indicate whether the page is writable or not. If that 'writable' flag is clear, we then make the host HPTE read-only even if the guest HPTE allowed writing. This means that we can get a protection fault when the guest writes to a page that it has mapped read-write but which is read-only on the host side (perhaps due to KSM having merged the page). Thus we now call kvmppc_handle_pagefault() for protection faults as well as HPTE not found faults. In kvmppc_handle_pagefault(), if the access was allowed by the guest HPTE and we thus need to install a new host HPTE, we then need to remove the old host HPTE if there is one. This is done with a new function, kvmppc_mmu_unmap_page(), which uses kvmppc_mmu_pte_vflush() to find and remove the old host HPTE. Since the memslot-related functions require the KVM SRCU read lock to be held, this adds srcu_read_lock/unlock pairs around the calls to kvmppc_handle_pagefault(). Finally, this changes kvmppc_mmu_book3s_32_xlate_pte() to not ignore guest HPTEs that don't permit access, and to return -EPERM for accesses that are not permitted by the page protections. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:51 +00:00
struct kvmppc_pte *gpte, bool data,
bool iswrite)
{
struct kvmppc_slb *slbe;
hva_t ptegp;
u64 pteg[16];
u64 avpn = 0;
u64 v, r;
u64 v_val, v_mask;
u64 eaddr_mask;
int i;
u8 pp, key = 0;
bool found = false;
bool second = false;
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
int pgsize;
ulong mp_ea = vcpu->arch.magic_page_ea;
/* Magic page override */
if (unlikely(mp_ea) &&
unlikely((eaddr & ~0xfffULL) == (mp_ea & ~0xfffULL)) &&
!(vcpu->arch.shared->msr & MSR_PR)) {
gpte->eaddr = eaddr;
gpte->vpage = kvmppc_mmu_book3s_64_ea_to_vp(vcpu, eaddr, data);
gpte->raddr = vcpu->arch.magic_page_pa | (gpte->raddr & 0xfff);
gpte->raddr &= KVM_PAM;
gpte->may_execute = true;
gpte->may_read = true;
gpte->may_write = true;
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
gpte->page_size = MMU_PAGE_4K;
return 0;
}
slbe = kvmppc_mmu_book3s_64_find_slbe(vcpu, eaddr);
if (!slbe)
goto no_seg_found;
avpn = kvmppc_mmu_book3s_64_get_avpn(slbe, eaddr);
v_val = avpn & HPTE_V_AVPN;
if (slbe->tb)
v_val |= SLB_VSID_B_1T;
if (slbe->large)
v_val |= HPTE_V_LARGE;
v_val |= HPTE_V_VALID;
v_mask = SLB_VSID_B | HPTE_V_AVPN | HPTE_V_LARGE | HPTE_V_VALID |
HPTE_V_SECONDARY;
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
pgsize = slbe->large ? MMU_PAGE_16M : MMU_PAGE_4K;
mutex_lock(&vcpu->kvm->arch.hpt_mutex);
do_second:
ptegp = kvmppc_mmu_book3s_64_get_pteg(vcpu, slbe, eaddr, second);
if (kvm_is_error_hva(ptegp))
goto no_page_found;
if(copy_from_user(pteg, (void __user *)ptegp, sizeof(pteg))) {
printk(KERN_ERR "KVM can't copy data from 0x%lx!\n", ptegp);
goto no_page_found;
}
if ((vcpu->arch.shared->msr & MSR_PR) && slbe->Kp)
key = 4;
else if (!(vcpu->arch.shared->msr & MSR_PR) && slbe->Ks)
key = 4;
for (i=0; i<16; i+=2) {
/* Check all relevant fields of 1st dword */
if ((pteg[i] & v_mask) == v_val) {
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
/* If large page bit is set, check pgsize encoding */
if (slbe->large &&
(vcpu->arch.hflags & BOOK3S_HFLAG_MULTI_PGSIZE)) {
pgsize = decode_pagesize(slbe, pteg[i+1]);
if (pgsize < 0)
continue;
}
found = true;
break;
}
}
if (!found) {
if (second)
goto no_page_found;
v_val |= HPTE_V_SECONDARY;
second = true;
goto do_second;
}
v = pteg[i];
r = pteg[i+1];
pp = (r & HPTE_R_PP) | key;
if (r & HPTE_R_PP0)
pp |= 8;
gpte->eaddr = eaddr;
gpte->vpage = kvmppc_mmu_book3s_64_ea_to_vp(vcpu, eaddr, data);
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
eaddr_mask = (1ull << mmu_pagesize(pgsize)) - 1;
gpte->raddr = (r & HPTE_R_RPN & ~eaddr_mask) | (eaddr & eaddr_mask);
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
gpte->page_size = pgsize;
gpte->may_execute = ((r & HPTE_R_N) ? false : true);
gpte->may_read = false;
gpte->may_write = false;
switch (pp) {
case 0:
case 1:
case 2:
case 6:
gpte->may_write = true;
/* fall through */
case 3:
case 5:
case 7:
case 10:
gpte->may_read = true;
break;
}
dprintk("KVM MMU: Translated 0x%lx [0x%llx] -> 0x%llx "
"-> 0x%lx\n",
eaddr, avpn, gpte->vpage, gpte->raddr);
/* Update PTE R and C bits, so the guest's swapper knows we used the
* page */
if (gpte->may_read && !(r & HPTE_R_R)) {
/*
* Set the accessed flag.
* We have to write this back with a single byte write
* because another vcpu may be accessing this on
* non-PAPR platforms such as mac99, and this is
* what real hardware does.
*/
char __user *addr = (char __user *) &pteg[i+1];
r |= HPTE_R_R;
put_user(r >> 8, addr + 6);
}
KVM: PPC: Book3S PR: Better handling of host-side read-only pages Currently we request write access to all pages that get mapped into the guest, even if the guest is only loading from the page. This reduces the effectiveness of KSM because it means that we unshare every page we access. Also, we always set the changed (C) bit in the guest HPTE if it allows writing, even for a guest load. This fixes both these problems. We pass an 'iswrite' flag to the mmu.xlate() functions and to kvmppc_mmu_map_page() to indicate whether the access is a load or a store. The mmu.xlate() functions now only set C for stores. kvmppc_gfn_to_pfn() now calls gfn_to_pfn_prot() instead of gfn_to_pfn() so that it can indicate whether we need write access to the page, and get back a 'writable' flag to indicate whether the page is writable or not. If that 'writable' flag is clear, we then make the host HPTE read-only even if the guest HPTE allowed writing. This means that we can get a protection fault when the guest writes to a page that it has mapped read-write but which is read-only on the host side (perhaps due to KSM having merged the page). Thus we now call kvmppc_handle_pagefault() for protection faults as well as HPTE not found faults. In kvmppc_handle_pagefault(), if the access was allowed by the guest HPTE and we thus need to install a new host HPTE, we then need to remove the old host HPTE if there is one. This is done with a new function, kvmppc_mmu_unmap_page(), which uses kvmppc_mmu_pte_vflush() to find and remove the old host HPTE. Since the memslot-related functions require the KVM SRCU read lock to be held, this adds srcu_read_lock/unlock pairs around the calls to kvmppc_handle_pagefault(). Finally, this changes kvmppc_mmu_book3s_32_xlate_pte() to not ignore guest HPTEs that don't permit access, and to return -EPERM for accesses that are not permitted by the page protections. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:51 +00:00
if (iswrite && gpte->may_write && !(r & HPTE_R_C)) {
/* Set the dirty flag */
/* Use a single byte write */
char __user *addr = (char __user *) &pteg[i+1];
r |= HPTE_R_C;
put_user(r, addr + 7);
}
mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
KVM: PPC: Book3S PR: Better handling of host-side read-only pages Currently we request write access to all pages that get mapped into the guest, even if the guest is only loading from the page. This reduces the effectiveness of KSM because it means that we unshare every page we access. Also, we always set the changed (C) bit in the guest HPTE if it allows writing, even for a guest load. This fixes both these problems. We pass an 'iswrite' flag to the mmu.xlate() functions and to kvmppc_mmu_map_page() to indicate whether the access is a load or a store. The mmu.xlate() functions now only set C for stores. kvmppc_gfn_to_pfn() now calls gfn_to_pfn_prot() instead of gfn_to_pfn() so that it can indicate whether we need write access to the page, and get back a 'writable' flag to indicate whether the page is writable or not. If that 'writable' flag is clear, we then make the host HPTE read-only even if the guest HPTE allowed writing. This means that we can get a protection fault when the guest writes to a page that it has mapped read-write but which is read-only on the host side (perhaps due to KSM having merged the page). Thus we now call kvmppc_handle_pagefault() for protection faults as well as HPTE not found faults. In kvmppc_handle_pagefault(), if the access was allowed by the guest HPTE and we thus need to install a new host HPTE, we then need to remove the old host HPTE if there is one. This is done with a new function, kvmppc_mmu_unmap_page(), which uses kvmppc_mmu_pte_vflush() to find and remove the old host HPTE. Since the memslot-related functions require the KVM SRCU read lock to be held, this adds srcu_read_lock/unlock pairs around the calls to kvmppc_handle_pagefault(). Finally, this changes kvmppc_mmu_book3s_32_xlate_pte() to not ignore guest HPTEs that don't permit access, and to return -EPERM for accesses that are not permitted by the page protections. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:51 +00:00
if (!gpte->may_read || (iswrite && !gpte->may_write))
return -EPERM;
return 0;
no_page_found:
mutex_unlock(&vcpu->kvm->arch.hpt_mutex);
return -ENOENT;
no_seg_found:
dprintk("KVM MMU: Trigger segment fault\n");
return -EINVAL;
}
static void kvmppc_mmu_book3s_64_slbmte(struct kvm_vcpu *vcpu, u64 rs, u64 rb)
{
struct kvmppc_vcpu_book3s *vcpu_book3s;
u64 esid, esid_1t;
int slb_nr;
struct kvmppc_slb *slbe;
dprintk("KVM MMU: slbmte(0x%llx, 0x%llx)\n", rs, rb);
vcpu_book3s = to_book3s(vcpu);
esid = GET_ESID(rb);
esid_1t = GET_ESID_1T(rb);
slb_nr = rb & 0xfff;
if (slb_nr > vcpu->arch.slb_nr)
return;
slbe = &vcpu->arch.slb[slb_nr];
slbe->large = (rs & SLB_VSID_L) ? 1 : 0;
slbe->tb = (rs & SLB_VSID_B_1T) ? 1 : 0;
slbe->esid = slbe->tb ? esid_1t : esid;
slbe->vsid = (rs & ~SLB_VSID_B) >> (kvmppc_slb_sid_shift(slbe) - 16);
slbe->valid = (rb & SLB_ESID_V) ? 1 : 0;
slbe->Ks = (rs & SLB_VSID_KS) ? 1 : 0;
slbe->Kp = (rs & SLB_VSID_KP) ? 1 : 0;
slbe->nx = (rs & SLB_VSID_N) ? 1 : 0;
slbe->class = (rs & SLB_VSID_C) ? 1 : 0;
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
slbe->base_page_size = MMU_PAGE_4K;
if (slbe->large) {
if (vcpu->arch.hflags & BOOK3S_HFLAG_MULTI_PGSIZE) {
switch (rs & SLB_VSID_LP) {
case SLB_VSID_LP_00:
slbe->base_page_size = MMU_PAGE_16M;
break;
case SLB_VSID_LP_01:
slbe->base_page_size = MMU_PAGE_64K;
break;
}
} else
slbe->base_page_size = MMU_PAGE_16M;
}
slbe->orige = rb & (ESID_MASK | SLB_ESID_V);
slbe->origv = rs;
/* Map the new segment */
kvmppc_mmu_map_segment(vcpu, esid << SID_SHIFT);
}
static u64 kvmppc_mmu_book3s_64_slbmfee(struct kvm_vcpu *vcpu, u64 slb_nr)
{
struct kvmppc_slb *slbe;
if (slb_nr > vcpu->arch.slb_nr)
return 0;
slbe = &vcpu->arch.slb[slb_nr];
return slbe->orige;
}
static u64 kvmppc_mmu_book3s_64_slbmfev(struct kvm_vcpu *vcpu, u64 slb_nr)
{
struct kvmppc_slb *slbe;
if (slb_nr > vcpu->arch.slb_nr)
return 0;
slbe = &vcpu->arch.slb[slb_nr];
return slbe->origv;
}
static void kvmppc_mmu_book3s_64_slbie(struct kvm_vcpu *vcpu, u64 ea)
{
struct kvmppc_slb *slbe;
u64 seg_size;
dprintk("KVM MMU: slbie(0x%llx)\n", ea);
slbe = kvmppc_mmu_book3s_64_find_slbe(vcpu, ea);
if (!slbe)
return;
dprintk("KVM MMU: slbie(0x%llx, 0x%llx)\n", ea, slbe->esid);
slbe->valid = false;
slbe->orige = 0;
slbe->origv = 0;
seg_size = 1ull << kvmppc_slb_sid_shift(slbe);
kvmppc_mmu_flush_segment(vcpu, ea & ~(seg_size - 1), seg_size);
}
static void kvmppc_mmu_book3s_64_slbia(struct kvm_vcpu *vcpu)
{
int i;
dprintk("KVM MMU: slbia()\n");
for (i = 1; i < vcpu->arch.slb_nr; i++) {
vcpu->arch.slb[i].valid = false;
vcpu->arch.slb[i].orige = 0;
vcpu->arch.slb[i].origv = 0;
}
if (vcpu->arch.shared->msr & MSR_IR) {
kvmppc_mmu_flush_segments(vcpu);
kvmppc_mmu_map_segment(vcpu, kvmppc_get_pc(vcpu));
}
}
static void kvmppc_mmu_book3s_64_mtsrin(struct kvm_vcpu *vcpu, u32 srnum,
ulong value)
{
u64 rb = 0, rs = 0;
/*
* According to Book3 2.01 mtsrin is implemented as:
*
* The SLB entry specified by (RB)32:35 is loaded from register
* RS, as follows.
*
* SLBE Bit Source SLB Field
*
* 0:31 0x0000_0000 ESID-0:31
* 32:35 (RB)32:35 ESID-32:35
* 36 0b1 V
* 37:61 0x00_0000|| 0b0 VSID-0:24
* 62:88 (RS)37:63 VSID-25:51
* 89:91 (RS)33:35 Ks Kp N
* 92 (RS)36 L ((RS)36 must be 0b0)
* 93 0b0 C
*/
dprintk("KVM MMU: mtsrin(0x%x, 0x%lx)\n", srnum, value);
/* ESID = srnum */
rb |= (srnum & 0xf) << 28;
/* Set the valid bit */
rb |= 1 << 27;
/* Index = ESID */
rb |= srnum;
/* VSID = VSID */
rs |= (value & 0xfffffff) << 12;
/* flags = flags */
rs |= ((value >> 28) & 0x7) << 9;
kvmppc_mmu_book3s_64_slbmte(vcpu, rs, rb);
}
static void kvmppc_mmu_book3s_64_tlbie(struct kvm_vcpu *vcpu, ulong va,
bool large)
{
u64 mask = 0xFFFFFFFFFULL;
long i;
struct kvm_vcpu *v;
dprintk("KVM MMU: tlbie(0x%lx)\n", va);
KVM: PPC: Book3S PR: Allow guest to use 64k pages This adds the code to interpret 64k HPTEs in the guest hashed page table (HPT), 64k SLB entries, and to tell the guest about 64k pages in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed by 4k pages. This also adds another hash table to the four we have already in book3s_mmu_hpte.c to allow us to find all the PTEs that we have instantiated that match a given 64k guest page. The tlbie instruction changed starting with POWER6 to use a bit in the RB operand to indicate large page invalidations, and to use other RB bits to indicate the base and actual page sizes and the segment size. 64k pages came in slightly earlier, with POWER5++. We use one bit in vcpu->arch.hflags to indicate that the emulated cpu supports 64k pages, and another to indicate that it has the new tlbie definition. The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because the MMU capabilities depend on which CPU model we're emulating, but it is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU fd. In addition, commonly-used userspace (QEMU) calls it before setting the PVR for any VCPU. Therefore, as a best effort we look at the first vcpu in the VM and return 64k pages or not depending on its capabilities. We also make the PVR default to the host PVR on recent CPUs that support 1TB segments (and therefore multiple page sizes as well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB segment support on those CPUs. Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
2013-09-20 04:52:44 +00:00
/*
* The tlbie instruction changed behaviour starting with
* POWER6. POWER6 and later don't have the large page flag
* in the instruction but in the RB value, along with bits
* indicating page and segment sizes.
*/
if (vcpu->arch.hflags & BOOK3S_HFLAG_NEW_TLBIE) {
/* POWER6 or later */
if (va & 1) { /* L bit */
if ((va & 0xf000) == 0x1000)
mask = 0xFFFFFFFF0ULL; /* 64k page */
else
mask = 0xFFFFFF000ULL; /* 16M page */
}
} else {
/* older processors, e.g. PPC970 */
if (large)
mask = 0xFFFFFF000ULL;
}
/* flush this VA on all vcpus */
kvm_for_each_vcpu(i, v, vcpu->kvm)
kvmppc_mmu_pte_vflush(v, va >> 12, mask);
}
#ifdef CONFIG_PPC_64K_PAGES
static int segment_contains_magic_page(struct kvm_vcpu *vcpu, ulong esid)
{
ulong mp_ea = vcpu->arch.magic_page_ea;
return mp_ea && !(vcpu->arch.shared->msr & MSR_PR) &&
(mp_ea >> SID_SHIFT) == esid;
}
#endif
static int kvmppc_mmu_book3s_64_esid_to_vsid(struct kvm_vcpu *vcpu, ulong esid,
u64 *vsid)
{
ulong ea = esid << SID_SHIFT;
struct kvmppc_slb *slb;
u64 gvsid = esid;
ulong mp_ea = vcpu->arch.magic_page_ea;
int pagesize = MMU_PAGE_64K;
if (vcpu->arch.shared->msr & (MSR_DR|MSR_IR)) {
slb = kvmppc_mmu_book3s_64_find_slbe(vcpu, ea);
if (slb) {
gvsid = slb->vsid;
pagesize = slb->base_page_size;
if (slb->tb) {
gvsid <<= SID_SHIFT_1T - SID_SHIFT;
gvsid |= esid & ((1ul << (SID_SHIFT_1T - SID_SHIFT)) - 1);
gvsid |= VSID_1T;
}
}
}
switch (vcpu->arch.shared->msr & (MSR_DR|MSR_IR)) {
case 0:
gvsid = VSID_REAL | esid;
break;
case MSR_IR:
gvsid |= VSID_REAL_IR;
break;
case MSR_DR:
gvsid |= VSID_REAL_DR;
break;
case MSR_DR|MSR_IR:
if (!slb)
goto no_slb;
break;
default:
BUG();
break;
}
#ifdef CONFIG_PPC_64K_PAGES
/*
* Mark this as a 64k segment if the host is using
* 64k pages, the host MMU supports 64k pages and
* the guest segment page size is >= 64k,
* but not if this segment contains the magic page.
*/
if (pagesize >= MMU_PAGE_64K &&
mmu_psize_defs[MMU_PAGE_64K].shift &&
!segment_contains_magic_page(vcpu, esid))
gvsid |= VSID_64K;
#endif
if (vcpu->arch.shared->msr & MSR_PR)
gvsid |= VSID_PR;
*vsid = gvsid;
return 0;
no_slb:
/* Catch magic page case */
if (unlikely(mp_ea) &&
unlikely(esid == (mp_ea >> SID_SHIFT)) &&
!(vcpu->arch.shared->msr & MSR_PR)) {
*vsid = VSID_REAL | esid;
return 0;
}
return -EINVAL;
}
static bool kvmppc_mmu_book3s_64_is_dcbz32(struct kvm_vcpu *vcpu)
{
return (to_book3s(vcpu)->hid[5] & 0x80);
}
void kvmppc_mmu_book3s_64_init(struct kvm_vcpu *vcpu)
{
struct kvmppc_mmu *mmu = &vcpu->arch.mmu;
mmu->mfsrin = NULL;
mmu->mtsrin = kvmppc_mmu_book3s_64_mtsrin;
mmu->slbmte = kvmppc_mmu_book3s_64_slbmte;
mmu->slbmfee = kvmppc_mmu_book3s_64_slbmfee;
mmu->slbmfev = kvmppc_mmu_book3s_64_slbmfev;
mmu->slbie = kvmppc_mmu_book3s_64_slbie;
mmu->slbia = kvmppc_mmu_book3s_64_slbia;
mmu->xlate = kvmppc_mmu_book3s_64_xlate;
mmu->reset_msr = kvmppc_mmu_book3s_64_reset_msr;
mmu->tlbie = kvmppc_mmu_book3s_64_tlbie;
mmu->esid_to_vsid = kvmppc_mmu_book3s_64_esid_to_vsid;
mmu->ea_to_vp = kvmppc_mmu_book3s_64_ea_to_vp;
mmu->is_dcbz32 = kvmppc_mmu_book3s_64_is_dcbz32;
vcpu->arch.hflags |= BOOK3S_HFLAG_SLB;
}