aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--arch/powerpc/include/asm/kvm_book3s_asm.h3
-rw-r--r--arch/powerpc/include/asm/kvm_host.h10
-rw-r--r--arch/powerpc/include/asm/kvm_ppc.h5
-rw-r--r--arch/powerpc/include/asm/reg.h3
-rw-r--r--arch/powerpc/kvm/book3s_64_mmu_hv.c40
-rw-r--r--arch/powerpc/kvm/book3s_hv.c59
-rw-r--r--arch/powerpc/kvm/book3s_hv_nested.c137
-rw-r--r--arch/powerpc/kvm/book3s_hv_p9_entry.c15
-rw-r--r--arch/powerpc/kvm/book3s_hv_rm_xics.c2
-rw-r--r--arch/powerpc/kvm/book3s_hv_rmhandlers.S8
-rw-r--r--arch/powerpc/kvm/book3s_hv_uvmem.c8
-rw-r--r--arch/powerpc/kvm/book3s_xive.c11
-rw-r--r--arch/powerpc/kvm/e500mc.c1
-rw-r--r--arch/powerpc/kvm/powerpc.c30
-rw-r--r--arch/powerpc/mm/init_64.c3
15 files changed, 200 insertions, 135 deletions
diff --git a/arch/powerpc/include/asm/kvm_book3s_asm.h b/arch/powerpc/include/asm/kvm_book3s_asm.h
index b6d31bff5209..c8882d9b86c2 100644
--- a/arch/powerpc/include/asm/kvm_book3s_asm.h
+++ b/arch/powerpc/include/asm/kvm_book3s_asm.h
@@ -14,9 +14,6 @@
#define XICS_MFRR 0xc
#define XICS_IPI 2 /* interrupt source # for IPIs */
-/* LPIDs we support with this build -- runtime limit may be lower */
-#define KVMPPC_NR_LPIDS (LPID_RSVD + 1)
-
/* Maximum number of threads per physical core */
#define MAX_SMT_THREADS 8
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index faf301d0dec0..2909a88acd16 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -36,7 +36,12 @@
#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
#include <asm/kvm_book3s_asm.h> /* for MAX_SMT_THREADS */
#define KVM_MAX_VCPU_IDS (MAX_SMT_THREADS * KVM_MAX_VCORES)
-#define KVM_MAX_NESTED_GUESTS KVMPPC_NR_LPIDS
+
+/*
+ * Limit the nested partition table to 4096 entries (because that's what
+ * hardware supports). Both guest and host use this value.
+ */
+#define KVM_MAX_NESTED_GUESTS_SHIFT 12
#else
#define KVM_MAX_VCPU_IDS KVM_MAX_VCPUS
@@ -327,8 +332,7 @@ struct kvm_arch {
struct list_head uvmem_pfns;
struct mutex mmu_setup_lock; /* nests inside vcpu mutexes */
u64 l1_ptcr;
- int max_nested_lpid;
- struct kvm_nested_guest *nested_guests[KVM_MAX_NESTED_GUESTS];
+ struct idr kvm_nested_guest_idr;
/* This array can grow quite large, keep it at the end */
struct kvmppc_vcore *vcores[KVM_MAX_VCORES];
#endif
diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 838d4cb460b7..2f80191e5437 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h
@@ -685,7 +685,7 @@ extern int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq,
int level, bool line_status);
extern void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu);
extern void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu);
-extern void kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu);
+extern bool kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu);
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{
@@ -723,7 +723,7 @@ static inline int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 ir
int level, bool line_status) { return -ENODEV; }
static inline void kvmppc_xive_push_vcpu(struct kvm_vcpu *vcpu) { }
static inline void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu) { }
-static inline void kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu) { }
+static inline bool kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu) { return true; }
static inline int kvmppc_xive_enabled(struct kvm_vcpu *vcpu)
{ return 0; }
@@ -877,7 +877,6 @@ int kvm_vcpu_ioctl_dirty_tlb(struct kvm_vcpu *vcpu,
struct kvm_dirty_tlb *cfg);
long kvmppc_alloc_lpid(void);
-void kvmppc_claim_lpid(long lpid);
void kvmppc_free_lpid(long lpid);
void kvmppc_init_lpid(unsigned long nr_lpids);
diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h
index 2835f6363228..1e8b2e04e626 100644
--- a/arch/powerpc/include/asm/reg.h
+++ b/arch/powerpc/include/asm/reg.h
@@ -417,7 +417,6 @@
#define FSCR_DSCR __MASK(FSCR_DSCR_LG)
#define FSCR_INTR_CAUSE (ASM_CONST(0xFF) << 56) /* interrupt cause */
#define SPRN_HFSCR 0xbe /* HV=1 Facility Status & Control Register */
-#define HFSCR_PREFIX __MASK(FSCR_PREFIX_LG)
#define HFSCR_MSGP __MASK(FSCR_MSGP_LG)
#define HFSCR_TAR __MASK(FSCR_TAR_LG)
#define HFSCR_EBB __MASK(FSCR_EBB_LG)
@@ -474,8 +473,6 @@
#ifndef SPRN_LPID
#define SPRN_LPID 0x13F /* Logical Partition Identifier */
#endif
-#define LPID_RSVD_POWER7 0x3ff /* Reserved LPID for partn switching */
-#define LPID_RSVD 0xfff /* Reserved LPID for partn switching */
#define SPRN_HMER 0x150 /* Hypervisor maintenance exception reg */
#define HMER_DEBUG_TRIG (1ul << (63 - 17)) /* Debug trigger */
#define SPRN_HMEER 0x151 /* Hyp maintenance exception enable reg */
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 0aeb51738ca9..c036b1a22b00 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -256,26 +256,34 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot,
int kvmppc_mmu_hv_init(void)
{
- unsigned long host_lpid, rsvd_lpid;
+ unsigned long nr_lpids;
if (!mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE))
return -EINVAL;
- host_lpid = 0;
- if (cpu_has_feature(CPU_FTR_HVMODE))
- host_lpid = mfspr(SPRN_LPID);
+ if (cpu_has_feature(CPU_FTR_HVMODE)) {
+ if (WARN_ON(mfspr(SPRN_LPID) != 0))
+ return -EINVAL;
+ nr_lpids = 1UL << mmu_lpid_bits;
+ } else {
+ nr_lpids = 1UL << KVM_MAX_NESTED_GUESTS_SHIFT;
+ }
- /* POWER8 and above have 12-bit LPIDs (10-bit in POWER7) */
- if (cpu_has_feature(CPU_FTR_ARCH_207S))
- rsvd_lpid = LPID_RSVD;
- else
- rsvd_lpid = LPID_RSVD_POWER7;
+ if (!cpu_has_feature(CPU_FTR_ARCH_300)) {
+ /* POWER7 has 10-bit LPIDs, POWER8 has 12-bit LPIDs */
+ if (cpu_has_feature(CPU_FTR_ARCH_207S))
+ WARN_ON(nr_lpids != 1UL << 12);
+ else
+ WARN_ON(nr_lpids != 1UL << 10);
- kvmppc_init_lpid(rsvd_lpid + 1);
+ /*
+ * Reserve the last implemented LPID use in partition
+ * switching for POWER7 and POWER8.
+ */
+ nr_lpids -= 1;
+ }
- kvmppc_claim_lpid(host_lpid);
- /* rsvd_lpid is reserved for use in partition switching */
- kvmppc_claim_lpid(rsvd_lpid);
+ kvmppc_init_lpid(nr_lpids);
return 0;
}
@@ -879,7 +887,7 @@ static bool kvm_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot,
struct revmap_entry *rev = kvm->arch.hpt.rev;
unsigned long head, i, j;
__be64 *hptep;
- int ret = 0;
+ bool ret = false;
unsigned long *rmapp;
rmapp = &memslot->arch.rmap[gfn - memslot->base_gfn];
@@ -887,7 +895,7 @@ static bool kvm_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot,
lock_rmap(rmapp);
if (*rmapp & KVMPPC_RMAP_REFERENCED) {
*rmapp &= ~KVMPPC_RMAP_REFERENCED;
- ret = 1;
+ ret = true;
}
if (!(*rmapp & KVMPPC_RMAP_PRESENT)) {
unlock_rmap(rmapp);
@@ -919,7 +927,7 @@ static bool kvm_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot,
rev[i].guest_rpte |= HPTE_R_R;
note_hpte_modification(kvm, &rev[i]);
}
- ret = 1;
+ ret = true;
}
__unlock_hpte(hptep, be64_to_cpu(hptep[0]));
} while ((i = j) != head);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 6fa518f6501d..c5fd9b4657dd 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -2834,7 +2834,7 @@ static int kvmppc_core_vcpu_create_hv(struct kvm_vcpu *vcpu)
* to trap and then we emulate them.
*/
vcpu->arch.hfscr = HFSCR_TAR | HFSCR_EBB | HFSCR_PM | HFSCR_BHRB |
- HFSCR_DSCR | HFSCR_VECVSX | HFSCR_FP | HFSCR_PREFIX;
+ HFSCR_DSCR | HFSCR_VECVSX | HFSCR_FP;
if (cpu_has_feature(CPU_FTR_HVMODE)) {
vcpu->arch.hfscr &= mfspr(SPRN_HFSCR);
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
@@ -3967,6 +3967,7 @@ static int kvmhv_vcpu_entry_p9_nested(struct kvm_vcpu *vcpu, u64 time_limit, uns
kvmhv_save_hv_regs(vcpu, &hvregs);
hvregs.lpcr = lpcr;
+ hvregs.amor = ~0;
vcpu->arch.regs.msr = vcpu->arch.shregs.msr;
hvregs.version = HV_GUEST_STATE_VERSION;
if (vcpu->arch.nested) {
@@ -4029,6 +4030,8 @@ static int kvmhv_vcpu_entry_p9_nested(struct kvm_vcpu *vcpu, u64 time_limit, uns
static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
unsigned long lpcr, u64 *tb)
{
+ struct kvm *kvm = vcpu->kvm;
+ struct kvm_nested_guest *nested = vcpu->arch.nested;
u64 next_timer;
int trap;
@@ -4048,34 +4051,61 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
trap = kvmhv_vcpu_entry_p9_nested(vcpu, time_limit, lpcr, tb);
/* H_CEDE has to be handled now, not later */
- if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
+ if (trap == BOOK3S_INTERRUPT_SYSCALL && !nested &&
kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
kvmppc_cede(vcpu);
kvmppc_set_gpr(vcpu, 3, 0);
trap = 0;
}
- } else {
- struct kvm *kvm = vcpu->kvm;
+ } else if (nested) {
+ __this_cpu_write(cpu_in_guest, kvm);
+ trap = kvmhv_vcpu_entry_p9(vcpu, time_limit, lpcr, tb);
+ __this_cpu_write(cpu_in_guest, NULL);
+ } else {
kvmppc_xive_push_vcpu(vcpu);
__this_cpu_write(cpu_in_guest, kvm);
trap = kvmhv_vcpu_entry_p9(vcpu, time_limit, lpcr, tb);
__this_cpu_write(cpu_in_guest, NULL);
- if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
+ if (trap == BOOK3S_INTERRUPT_SYSCALL &&
!(vcpu->arch.shregs.msr & MSR_PR)) {
unsigned long req = kvmppc_get_gpr(vcpu, 3);
- /* H_CEDE has to be handled now, not later */
+ /*
+ * XIVE rearm and XICS hcalls must be handled
+ * before xive context is pulled (is this
+ * true?)
+ */
if (req == H_CEDE) {
+ /* H_CEDE has to be handled now */
kvmppc_cede(vcpu);
- kvmppc_xive_rearm_escalation(vcpu); /* may un-cede */
+ if (!kvmppc_xive_rearm_escalation(vcpu)) {
+ /*
+ * Pending escalation so abort
+ * the cede.
+ */
+ vcpu->arch.ceded = 0;
+ }
kvmppc_set_gpr(vcpu, 3, 0);
trap = 0;
- /* XICS hcalls must be handled before xive is pulled */
+ } else if (req == H_ENTER_NESTED) {
+ /*
+ * L2 should not run with the L1
+ * context so rearm and pull it.
+ */
+ if (!kvmppc_xive_rearm_escalation(vcpu)) {
+ /*
+ * Pending escalation so abort
+ * H_ENTER_NESTED.
+ */
+ kvmppc_set_gpr(vcpu, 3, 0);
+ trap = 0;
+ }
+
} else if (hcall_is_xics(req)) {
int ret;
@@ -4519,9 +4549,14 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
if (!nested) {
kvmppc_core_prepare_to_enter(vcpu);
- if (test_bit(BOOK3S_IRQPRIO_EXTERNAL,
- &vcpu->arch.pending_exceptions))
+ if (vcpu->arch.shregs.msr & MSR_EE) {
+ if (xive_interrupt_pending(vcpu))
+ kvmppc_inject_interrupt_hv(vcpu,
+ BOOK3S_INTERRUPT_EXTERNAL, 0);
+ } else if (test_bit(BOOK3S_IRQPRIO_EXTERNAL,
+ &vcpu->arch.pending_exceptions)) {
lpcr |= LPCR_MER;
+ }
} else if (vcpu->arch.pending_exceptions ||
vcpu->arch.doorbell_request ||
xive_interrupt_pending(vcpu)) {
@@ -5283,6 +5318,10 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm)
kvm->arch.host_lpcr = lpcr = mfspr(SPRN_LPCR);
lpcr &= LPCR_PECE | LPCR_LPES;
} else {
+ /*
+ * The L2 LPES mode will be set by the L0 according to whether
+ * or not it needs to take external interrupts in HV mode.
+ */
lpcr = 0;
}
lpcr |= (4UL << LPCR_DPFD_SH) | LPCR_HDICE |
diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c
index 9d373f8963ee..ce08573fc58f 100644
--- a/arch/powerpc/kvm/book3s_hv_nested.c
+++ b/arch/powerpc/kvm/book3s_hv_nested.c
@@ -261,8 +261,7 @@ static void load_l2_hv_regs(struct kvm_vcpu *vcpu,
/*
* Don't let L1 change LPCR bits for the L2 except these:
*/
- mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD |
- LPCR_LPES | LPCR_MER;
+ mask = LPCR_DPFD | LPCR_ILE | LPCR_TC | LPCR_AIL | LPCR_LD | LPCR_MER;
/*
* Additional filtering is required depending on hardware
@@ -439,10 +438,11 @@ long kvmhv_nested_init(void)
if (!radix_enabled())
return -ENODEV;
- /* find log base 2 of KVMPPC_NR_LPIDS, rounding up */
- ptb_order = __ilog2(KVMPPC_NR_LPIDS - 1) + 1;
- if (ptb_order < 8)
- ptb_order = 8;
+ /* Partition table entry is 1<<4 bytes in size, hence the 4. */
+ ptb_order = KVM_MAX_NESTED_GUESTS_SHIFT + 4;
+ /* Minimum partition table size is 1<<12 bytes */
+ if (ptb_order < 12)
+ ptb_order = 12;
pseries_partition_tb = kmalloc(sizeof(struct patb_entry) << ptb_order,
GFP_KERNEL);
if (!pseries_partition_tb) {
@@ -450,7 +450,7 @@ long kvmhv_nested_init(void)
return -ENOMEM;
}
- ptcr = __pa(pseries_partition_tb) | (ptb_order - 8);
+ ptcr = __pa(pseries_partition_tb) | (ptb_order - 12);
rc = plpar_hcall_norets(H_SET_PARTITION_TABLE, ptcr);
if (rc != H_SUCCESS) {
pr_err("kvm-hv: Parent hypervisor does not support nesting (rc=%ld)\n",
@@ -521,11 +521,6 @@ static void kvmhv_set_nested_ptbl(struct kvm_nested_guest *gp)
kvmhv_set_ptbl_entry(gp->shadow_lpid, dw0, gp->process_table);
}
-void kvmhv_vm_nested_init(struct kvm *kvm)
-{
- kvm->arch.max_nested_lpid = -1;
-}
-
/*
* Handle the H_SET_PARTITION_TABLE hcall.
* r4 = guest real address of partition table + log_2(size) - 12
@@ -539,16 +534,14 @@ long kvmhv_set_partition_table(struct kvm_vcpu *vcpu)
long ret = H_SUCCESS;
srcu_idx = srcu_read_lock(&kvm->srcu);
- /*
- * Limit the partition table to 4096 entries (because that's what
- * hardware supports), and check the base address.
- */
- if ((ptcr & PRTS_MASK) > 12 - 8 ||
+ /* Check partition size and base address. */
+ if ((ptcr & PRTS_MASK) + 12 - 4 > KVM_MAX_NESTED_GUESTS_SHIFT ||
!kvm_is_visible_gfn(vcpu->kvm, (ptcr & PRTB_MASK) >> PAGE_SHIFT))
ret = H_PARAMETER;
srcu_read_unlock(&kvm->srcu, srcu_idx);
if (ret == H_SUCCESS)
kvm->arch.l1_ptcr = ptcr;
+
return ret;
}
@@ -644,7 +637,7 @@ static void kvmhv_update_ptbl_cache(struct kvm_nested_guest *gp)
ret = -EFAULT;
ptbl_addr = (kvm->arch.l1_ptcr & PRTB_MASK) + (gp->l1_lpid << 4);
- if (gp->l1_lpid < (1ul << ((kvm->arch.l1_ptcr & PRTS_MASK) + 8))) {
+ if (gp->l1_lpid < (1ul << ((kvm->arch.l1_ptcr & PRTS_MASK) + 12 - 4))) {
int srcu_idx = srcu_read_lock(&kvm->srcu);
ret = kvm_read_guest(kvm, ptbl_addr,
&ptbl_entry, sizeof(ptbl_entry));
@@ -660,6 +653,35 @@ static void kvmhv_update_ptbl_cache(struct kvm_nested_guest *gp)
kvmhv_set_nested_ptbl(gp);
}
+void kvmhv_vm_nested_init(struct kvm *kvm)
+{
+ idr_init(&kvm->arch.kvm_nested_guest_idr);
+}
+
+static struct kvm_nested_guest *__find_nested(struct kvm *kvm, int lpid)
+{
+ return idr_find(&kvm->arch.kvm_nested_guest_idr, lpid);
+}
+
+static bool __prealloc_nested(struct kvm *kvm, int lpid)
+{
+ if (idr_alloc(&kvm->arch.kvm_nested_guest_idr,
+ NULL, lpid, lpid + 1, GFP_KERNEL) != lpid)
+ return false;
+ return true;
+}
+
+static void __add_nested(struct kvm *kvm, int lpid, struct kvm_nested_guest *gp)
+{
+ if (idr_replace(&kvm->arch.kvm_nested_guest_idr, gp, lpid))
+ WARN_ON(1);
+}
+
+static void __remove_nested(struct kvm *kvm, int lpid)
+{
+ idr_remove(&kvm->arch.kvm_nested_guest_idr, lpid);
+}
+
static struct kvm_nested_guest *kvmhv_alloc_nested(struct kvm *kvm, unsigned int lpid)
{
struct kvm_nested_guest *gp;
@@ -720,13 +742,8 @@ static void kvmhv_remove_nested(struct kvm_nested_guest *gp)
long ref;
spin_lock(&kvm->mmu_lock);
- if (gp == kvm->arch.nested_guests[lpid]) {
- kvm->arch.nested_guests[lpid] = NULL;
- if (lpid == kvm->arch.max_nested_lpid) {
- while (--lpid >= 0 && !kvm->arch.nested_guests[lpid])
- ;
- kvm->arch.max_nested_lpid = lpid;
- }
+ if (gp == __find_nested(kvm, lpid)) {
+ __remove_nested(kvm, lpid);
--gp->refcnt;
}
ref = gp->refcnt;
@@ -743,24 +760,22 @@ static void kvmhv_remove_nested(struct kvm_nested_guest *gp)
*/
void kvmhv_release_all_nested(struct kvm *kvm)
{
- int i;
+ int lpid;
struct kvm_nested_guest *gp;
struct kvm_nested_guest *freelist = NULL;
struct kvm_memory_slot *memslot;
int srcu_idx, bkt;
spin_lock(&kvm->mmu_lock);
- for (i = 0; i <= kvm->arch.max_nested_lpid; i++) {
- gp = kvm->arch.nested_guests[i];
- if (!gp)
- continue;
- kvm->arch.nested_guests[i] = NULL;
+ idr_for_each_entry(&kvm->arch.kvm_nested_guest_idr, gp, lpid) {
+ __remove_nested(kvm, lpid);
if (--gp->refcnt == 0) {
gp->next = freelist;
freelist = gp;
}
}
- kvm->arch.max_nested_lpid = -1;
+ idr_destroy(&kvm->arch.kvm_nested_guest_idr);
+ /* idr is empty and may be reused at this point */
spin_unlock(&kvm->mmu_lock);
while ((gp = freelist) != NULL) {
freelist = gp->next;
@@ -792,12 +807,11 @@ struct kvm_nested_guest *kvmhv_get_nested(struct kvm *kvm, int l1_lpid,
{
struct kvm_nested_guest *gp, *newgp;
- if (l1_lpid >= KVM_MAX_NESTED_GUESTS ||
- l1_lpid >= (1ul << ((kvm->arch.l1_ptcr & PRTS_MASK) + 12 - 4)))
+ if (l1_lpid >= (1ul << ((kvm->arch.l1_ptcr & PRTS_MASK) + 12 - 4)))
return NULL;
spin_lock(&kvm->mmu_lock);
- gp = kvm->arch.nested_guests[l1_lpid];
+ gp = __find_nested(kvm, l1_lpid);
if (gp)
++gp->refcnt;
spin_unlock(&kvm->mmu_lock);
@@ -808,17 +822,19 @@ struct kvm_nested_guest *kvmhv_get_nested(struct kvm *kvm, int l1_lpid,
newgp = kvmhv_alloc_nested(kvm, l1_lpid);
if (!newgp)
return NULL;
+
+ if (!__prealloc_nested(kvm, l1_lpid)) {
+ kvmhv_release_nested(newgp);
+ return NULL;
+ }
+
spin_lock(&kvm->mmu_lock);
- if (kvm->arch.nested_guests[l1_lpid]) {
- /* someone else beat us to it */
- gp = kvm->arch.nested_guests[l1_lpid];
- } else {
- kvm->arch.nested_guests[l1_lpid] = newgp;
+ gp = __find_nested(kvm, l1_lpid);
+ if (!gp) {
+ __add_nested(kvm, l1_lpid, newgp);
++newgp->refcnt;
gp = newgp;
newgp = NULL;
- if (l1_lpid > kvm->arch.max_nested_lpid)
- kvm->arch.max_nested_lpid = l1_lpid;
}
++gp->refcnt;
spin_unlock(&kvm->mmu_lock);
@@ -841,20 +857,13 @@ void kvmhv_put_nested(struct kvm_nested_guest *gp)
kvmhv_release_nested(gp);
}
-static struct kvm_nested_guest *kvmhv_find_nested(struct kvm *kvm, int lpid)
-{
- if (lpid > kvm->arch.max_nested_lpid)
- return NULL;
- return kvm->arch.nested_guests[lpid];
-}
-
pte_t *find_kvm_nested_guest_pte(struct kvm *kvm, unsigned long lpid,
unsigned long ea, unsigned *hshift)
{
struct kvm_nested_guest *gp;
pte_t *pte;
- gp = kvmhv_find_nested(kvm, lpid);
+ gp = __find_nested(kvm, lpid);
if (!gp)
return NULL;
@@ -960,7 +969,7 @@ static void kvmhv_remove_nest_rmap(struct kvm *kvm, u64 n_rmap,
gpa = n_rmap & RMAP_NESTED_GPA_MASK;
lpid = (n_rmap & RMAP_NESTED_LPID_MASK) >> RMAP_NESTED_LPID_SHIFT;
- gp = kvmhv_find_nested(kvm, lpid);
+ gp = __find_nested(kvm, lpid);
if (!gp)
return;
@@ -1152,16 +1161,13 @@ static void kvmhv_emulate_tlbie_all_lpid(struct kvm_vcpu *vcpu, int ric)
{
struct kvm *kvm = vcpu->kvm;
struct kvm_nested_guest *gp;
- int i;
+ int lpid;
spin_lock(&kvm->mmu_lock);
- for (i = 0; i <= kvm->arch.max_nested_lpid; i++) {
- gp = kvm->arch.nested_guests[i];
- if (gp) {
- spin_unlock(&kvm->mmu_lock);
- kvmhv_emulate_tlbie_lpid(vcpu, gp, ric);
- spin_lock(&kvm->mmu_lock);
- }
+ idr_for_each_entry(&kvm->arch.kvm_nested_guest_idr, gp, lpid) {
+ spin_unlock(&kvm->mmu_lock);
+ kvmhv_emulate_tlbie_lpid(vcpu, gp, ric);
+ spin_lock(&kvm->mmu_lock);
}
spin_unlock(&kvm->mmu_lock);
}
@@ -1313,7 +1319,7 @@ long do_h_rpt_invalidate_pat(struct kvm_vcpu *vcpu, unsigned long lpid,
* H_ENTER_NESTED call. Since we can't differentiate this case from
* the invalid case, we ignore such flush requests and return success.
*/
- if (!kvmhv_find_nested(vcpu->kvm, lpid))
+ if (!__find_nested(vcpu->kvm, lpid))
return H_SUCCESS;
/*
@@ -1657,15 +1663,12 @@ long int kvmhv_nested_page_fault(struct kvm_vcpu *vcpu)
int kvmhv_nested_next_lpid(struct kvm *kvm, int lpid)
{
- int ret = -1;
+ int ret = lpid + 1;
spin_lock(&kvm->mmu_lock);
- while (++lpid <= kvm->arch.max_nested_lpid) {
- if (kvm->arch.nested_guests[lpid]) {
- ret = lpid;
- break;
- }
- }
+ if (!idr_get_next(&kvm->arch.kvm_nested_guest_idr, &ret))
+ ret = -1;
spin_unlock(&kvm->mmu_lock);
+
return ret;
}
diff --git a/arch/powerpc/kvm/book3s_hv_p9_entry.c b/arch/powerpc/kvm/book3s_hv_p9_entry.c
index a28e5b3daabd..9dba3e3f65a0 100644
--- a/arch/powerpc/kvm/book3s_hv_p9_entry.c
+++ b/arch/powerpc/kvm/book3s_hv_p9_entry.c
@@ -539,8 +539,10 @@ static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u6
{
struct kvm_nested_guest *nested = vcpu->arch.nested;
u32 lpid;
+ u32 pid;
lpid = nested ? nested->shadow_lpid : kvm->arch.lpid;
+ pid = vcpu->arch.pid;
/*
* Prior memory accesses to host PID Q3 must be completed before we
@@ -551,7 +553,7 @@ static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u6
isync();
mtspr(SPRN_LPID, lpid);
mtspr(SPRN_LPCR, lpcr);
- mtspr(SPRN_PID, vcpu->arch.pid);
+ mtspr(SPRN_PID, pid);
/*
* isync not required here because we are HRFID'ing to guest before
* any guest context access, which is context synchronising.
@@ -561,9 +563,11 @@ static void switch_mmu_to_guest_radix(struct kvm *kvm, struct kvm_vcpu *vcpu, u6
static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64 lpcr)
{
u32 lpid;
+ u32 pid;
int i;
lpid = kvm->arch.lpid;
+ pid = vcpu->arch.pid;
/*
* See switch_mmu_to_guest_radix. ptesync should not be required here
@@ -574,7 +578,7 @@ static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64
isync();
mtspr(SPRN_LPID, lpid);
mtspr(SPRN_LPCR, lpcr);
- mtspr(SPRN_PID, vcpu->arch.pid);
+ mtspr(SPRN_PID, pid);
for (i = 0; i < vcpu->arch.slb_max; i++)
mtslb(vcpu->arch.slb[i].orige, vcpu->arch.slb[i].origv);
@@ -585,6 +589,9 @@ static void switch_mmu_to_guest_hpt(struct kvm *kvm, struct kvm_vcpu *vcpu, u64
static void switch_mmu_to_host(struct kvm *kvm, u32 pid)
{
+ u32 lpid = kvm->arch.host_lpid;
+ u64 lpcr = kvm->arch.host_lpcr;
+
/*
* The guest has exited, so guest MMU context is no longer being
* non-speculatively accessed, but a hwsync is needed before the
@@ -594,8 +601,8 @@ static void switch_mmu_to_host(struct kvm *kvm, u32 pid)
asm volatile("hwsync" ::: "memory");
isync();
mtspr(SPRN_PID, pid);
- mtspr(SPRN_LPID, kvm->arch.host_lpid);
- mtspr(SPRN_LPCR, kvm->arch.host_lpcr);
+ mtspr(SPRN_LPID, lpid);
+ mtspr(SPRN_LPCR, lpcr);
/*
* isync is not required after the switch, because mtmsrd with L=0
* is performed after this switch, which is context synchronising.
diff --git a/arch/powerpc/kvm/book3s_hv_rm_xics.c b/arch/powerpc/kvm/book3s_hv_rm_xics.c
index 587c33fc4564..6e16bd751c84 100644
--- a/arch/powerpc/kvm/book3s_hv_rm_xics.c
+++ b/arch/powerpc/kvm/book3s_hv_rm_xics.c
@@ -883,7 +883,7 @@ long kvmppc_deliver_irq_passthru(struct kvm_vcpu *vcpu,
/* --- Non-real mode XICS-related built-in routines --- */
-/**
+/*
* Host Operations poked by RM KVM
*/
static void rm_host_ipi_action(int action, void *data)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index d185dee26026..0c552885a032 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -51,6 +51,14 @@
#define STACK_SLOT_FSCR (SFS-96)
/*
+ * Use the last LPID (all implemented LPID bits = 1) for partition switching.
+ * This is reserved in the LPID allocator. POWER7 only implements 0x3ff, but
+ * we write 0xfff into the LPID SPR anyway, which seems to work and just
+ * ignores the top bits.
+ */
+#define LPID_RSVD 0xfff
+
+/*
* Call kvmppc_hv_entry in real mode.
* Must be called with interrupts hard-disabled.
*
diff --git a/arch/powerpc/kvm/book3s_hv_uvmem.c b/arch/powerpc/kvm/book3s_hv_uvmem.c
index 45c993dd05f5..36f2314c58e5 100644
--- a/arch/powerpc/kvm/book3s_hv_uvmem.c
+++ b/arch/powerpc/kvm/book3s_hv_uvmem.c
@@ -361,13 +361,15 @@ static bool kvmppc_gfn_is_uvmem_pfn(unsigned long gfn, struct kvm *kvm,
static bool kvmppc_next_nontransitioned_gfn(const struct kvm_memory_slot *memslot,
struct kvm *kvm, unsigned long *gfn)
{
- struct kvmppc_uvmem_slot *p;
+ struct kvmppc_uvmem_slot *p = NULL, *iter;
bool ret = false;
unsigned long i;
- list_for_each_entry(p, &kvm->arch.uvmem_pfns, list)
- if (*gfn >= p->base_pfn && *gfn < p->base_pfn + p->nr_pfns)
+ list_for_each_entry(iter, &kvm->arch.uvmem_pfns, list)
+ if (*gfn >= iter->base_pfn && *gfn < iter->base_pfn + iter->nr_pfns) {
+ p = iter;
break;
+ }
if (!p)
return ret;
/*
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index c0ce5531d9bc..ee4be73649e5 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -179,12 +179,13 @@ void kvmppc_xive_pull_vcpu(struct kvm_vcpu *vcpu)
}
EXPORT_SYMBOL_GPL(kvmppc_xive_pull_vcpu);
-void kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu)
+bool kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu)
{
void __iomem *esc_vaddr = (void __iomem *)vcpu->arch.xive_esc_vaddr;
+ bool ret = true;
if (!esc_vaddr)
- return;
+ return ret;
/* we are using XIVE with single escalation */
@@ -197,7 +198,7 @@ void kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu)
* we also don't want to set xive_esc_on to 1 here in
* case we race with xive_esc_irq().
*/
- vcpu->arch.ceded = 0;
+ ret = false;
/*
* The escalation interrupts are special as we don't EOI them.
* There is no need to use the load-after-store ordering offset
@@ -210,6 +211,8 @@ void kvmppc_xive_rearm_escalation(struct kvm_vcpu *vcpu)
__raw_readq(esc_vaddr + XIVE_ESB_SET_PQ_00);
}
mb();
+
+ return ret;
}
EXPORT_SYMBOL_GPL(kvmppc_xive_rearm_escalation);
@@ -238,7 +241,7 @@ static irqreturn_t xive_esc_irq(int irq, void *data)
vcpu->arch.irq_pending = 1;
smp_mb();
- if (vcpu->arch.ceded)
+ if (vcpu->arch.ceded || vcpu->arch.nested)
kvmppc_fast_vcpu_kick(vcpu);
/* Since we have the no-EOI flag, the interrupt is effectively
diff --git a/arch/powerpc/kvm/e500mc.c b/arch/powerpc/kvm/e500mc.c
index fa0d8dbbe484..2b9ad8984ea1 100644
--- a/arch/powerpc/kvm/e500mc.c
+++ b/arch/powerpc/kvm/e500mc.c
@@ -399,7 +399,6 @@ static int __init kvmppc_e500mc_init(void)
* allocator.
*/
kvmppc_init_lpid(KVMPPC_NR_LPIDS/threads_per_core);
- kvmppc_claim_lpid(0); /* host */
r = kvm_init(NULL, sizeof(struct kvmppc_vcpu_e500), 0, THIS_MODULE);
if (r)
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 875c30c12db0..32561192cf12 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -2496,41 +2496,37 @@ out:
return r;
}
-static unsigned long lpid_inuse[BITS_TO_LONGS(KVMPPC_NR_LPIDS)];
+static DEFINE_IDA(lpid_inuse);
static unsigned long nr_lpids;
long kvmppc_alloc_lpid(void)
{
- long lpid;
+ int lpid;
- do {
- lpid = find_first_zero_bit(lpid_inuse, KVMPPC_NR_LPIDS);
- if (lpid >= nr_lpids) {
+ /* The host LPID must always be 0 (allocation starts at 1) */
+ lpid = ida_alloc_range(&lpid_inuse, 1, nr_lpids - 1, GFP_KERNEL);
+ if (lpid < 0) {
+ if (lpid == -ENOMEM)
+ pr_err("%s: Out of memory\n", __func__);
+ else
pr_err("%s: No LPIDs free\n", __func__);
- return -ENOMEM;
- }
- } while (test_and_set_bit(lpid, lpid_inuse));
+ return -ENOMEM;
+ }
return lpid;
}
EXPORT_SYMBOL_GPL(kvmppc_alloc_lpid);
-void kvmppc_claim_lpid(long lpid)
-{
- set_bit(lpid, lpid_inuse);
-}
-EXPORT_SYMBOL_GPL(kvmppc_claim_lpid);
-
void kvmppc_free_lpid(long lpid)
{
- clear_bit(lpid, lpid_inuse);
+ ida_free(&lpid_inuse, lpid);
}
EXPORT_SYMBOL_GPL(kvmppc_free_lpid);
+/* nr_lpids_param includes the host LPID */
void kvmppc_init_lpid(unsigned long nr_lpids_param)
{
- nr_lpids = min_t(unsigned long, KVMPPC_NR_LPIDS, nr_lpids_param);
- memset(lpid_inuse, 0, sizeof(lpid_inuse));
+ nr_lpids = nr_lpids_param;
}
EXPORT_SYMBOL_GPL(kvmppc_init_lpid);
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index 83c0ee9fbf05..0f2608679067 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -372,6 +372,9 @@ void register_page_bootmem_memmap(unsigned long section_nr,
#ifdef CONFIG_PPC_BOOK3S_64
unsigned int mmu_lpid_bits;
+#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
+EXPORT_SYMBOL_GPL(mmu_lpid_bits);
+#endif
unsigned int mmu_pid_bits;
static bool disable_radix = !IS_ENABLED(CONFIG_PPC_RADIX_MMU_DEFAULT);
Powered by cgit, Gruvboxed by Blaster4385.