[PATCH] KVM: x86: extend usage of RET_MMIO_PF_* constants

From: Paolo Bonzini
Date: Tue Oct 10 2017 - 05:11:10 EST


The x86 MMU if full of code that returns 0 and 1 for retry/emulate. Use
the existing RET_MMIO_PF_RETRY/RET_MMIO_PF_EMULATE enum, renaming it to
drop the MMIO part.

Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
---
arch/x86/kvm/mmu.c | 95 ++++++++++++++++++++++------------------------
arch/x86/kvm/paging_tmpl.h | 18 ++++-----
2 files changed, 55 insertions(+), 58 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 1706683d5fd7..c65593d7f23d 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -150,6 +150,20 @@ enum {
/* make pte_list_desc fit well in cache line */
#define PTE_LIST_EXT 3

+/*
+ * Return values of handle_mmio_page_fault and mmu.page_fault:
+ * RET_PF_RETRY: let CPU fault again on the address.
+ * RET_PF_EMULATE: mmio page fault, emulate the instruction directly.
+ *
+ * For handle_mmio_page_fault only:
+ * RET_PF_INVALID: the spte is invalid, let the real page fault path update it.
+ */
+enum {
+ RET_PF_RETRY = 0,
+ RET_PF_EMULATE = 1,
+ RET_PF_INVALID = 2,
+};
+
struct pte_list_desc {
u64 *sptes[PTE_LIST_EXT];
struct pte_list_desc *more;
@@ -2794,13 +2808,13 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep,
return ret;
}

-static bool mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, unsigned pte_access,
- int write_fault, int level, gfn_t gfn, kvm_pfn_t pfn,
- bool speculative, bool host_writable)
+static int mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, unsigned pte_access,
+ int write_fault, int level, gfn_t gfn, kvm_pfn_t pfn,
+ bool speculative, bool host_writable)
{
int was_rmapped = 0;
int rmap_count;
- bool emulate = false;
+ int ret = RET_PF_RETRY;

pgprintk("%s: spte %llx write_fault %d gfn %llx\n", __func__,
*sptep, write_fault, gfn);
@@ -2830,12 +2844,12 @@ static bool mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, unsigned pte_access,
if (set_spte(vcpu, sptep, pte_access, level, gfn, pfn, speculative,
true, host_writable)) {
if (write_fault)
- emulate = true;
+ ret = RET_PF_EMULATE;
kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
}

if (unlikely(is_mmio_spte(*sptep)))
- emulate = true;
+ ret = RET_PF_EMULATE;

pgprintk("%s: setting spte %llx\n", __func__, *sptep);
pgprintk("instantiating %s PTE (%s) at %llx (%llx) addr %p\n",
@@ -2855,7 +2869,7 @@ static bool mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, unsigned pte_access,

kvm_release_pfn_clean(pfn);

- return emulate;
+ return ret;
}

static kvm_pfn_t pte_prefetch_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn,
@@ -2994,14 +3008,13 @@ static int kvm_handle_bad_page(struct kvm_vcpu *vcpu, gfn_t gfn, kvm_pfn_t pfn)
* Do not cache the mmio info caused by writing the readonly gfn
* into the spte otherwise read access on readonly gfn also can
* caused mmio page fault and treat it as mmio access.
- * Return 1 to tell kvm to emulate it.
*/
if (pfn == KVM_PFN_ERR_RO_FAULT)
- return 1;
+ return RET_PF_EMULATE;

if (pfn == KVM_PFN_ERR_HWPOISON) {
kvm_send_hwpoison_signal(kvm_vcpu_gfn_to_hva(vcpu, gfn), current);
- return 0;
+ return RET_PF_RETRY;
}

return -EFAULT;
@@ -3286,13 +3299,13 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, u32 error_code,
}

if (fast_page_fault(vcpu, v, level, error_code))
- return 0;
+ return RET_PF_RETRY;

mmu_seq = vcpu->kvm->mmu_notifier_seq;
smp_rmb();

if (try_async_pf(vcpu, prefault, gfn, v, &pfn, write, &map_writable))
- return 0;
+ return RET_PF_RETRY;

if (handle_abnormal_pfn(vcpu, v, gfn, pfn, ACC_ALL, &r))
return r;
@@ -3312,7 +3325,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, u32 error_code,
out_unlock:
spin_unlock(&vcpu->kvm->mmu_lock);
kvm_release_pfn_clean(pfn);
- return 0;
+ return RET_PF_RETRY;
}


@@ -3659,54 +3672,38 @@ static bool mmio_info_in_cache(struct kvm_vcpu *vcpu, u64 addr, bool direct)
return reserved;
}

-/*
- * Return values of handle_mmio_page_fault:
- * RET_MMIO_PF_EMULATE: it is a real mmio page fault, emulate the instruction
- * directly.
- * RET_MMIO_PF_INVALID: invalid spte is detected then let the real page
- * fault path update the mmio spte.
- * RET_MMIO_PF_RETRY: let CPU fault again on the address.
- * RET_MMIO_PF_BUG: a bug was detected (and a WARN was printed).
- */
-enum {
- RET_MMIO_PF_EMULATE = 1,
- RET_MMIO_PF_INVALID = 2,
- RET_MMIO_PF_RETRY = 0,
- RET_MMIO_PF_BUG = -1
-};
-
static int handle_mmio_page_fault(struct kvm_vcpu *vcpu, u64 addr, bool direct)
{
u64 spte;
bool reserved;

if (mmio_info_in_cache(vcpu, addr, direct))
- return RET_MMIO_PF_EMULATE;
+ return RET_PF_EMULATE;

reserved = walk_shadow_page_get_mmio_spte(vcpu, addr, &spte);
if (WARN_ON(reserved))
- return RET_MMIO_PF_BUG;
+ return -EINVAL;

if (is_mmio_spte(spte)) {
gfn_t gfn = get_mmio_spte_gfn(spte);
unsigned access = get_mmio_spte_access(spte);

if (!check_mmio_spte(vcpu, spte))
- return RET_MMIO_PF_INVALID;
+ return RET_PF_INVALID;

if (direct)
addr = 0;

trace_handle_mmio_page_fault(addr, gfn, access);
vcpu_cache_mmio_info(vcpu, addr, gfn, access);
- return RET_MMIO_PF_EMULATE;
+ return RET_PF_EMULATE;
}

/*
* If the page table is zapped by other cpus, let CPU fault again on
* the address.
*/
- return RET_MMIO_PF_RETRY;
+ return RET_PF_RETRY;
}
EXPORT_SYMBOL_GPL(handle_mmio_page_fault);

@@ -3756,7 +3753,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
pgprintk("%s: gva %lx error %x\n", __func__, gva, error_code);

if (page_fault_handle_page_track(vcpu, error_code, gfn))
- return 1;
+ return RET_PF_EMULATE;

r = mmu_topup_memory_caches(vcpu);
if (r)
@@ -3875,7 +3872,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
MMU_WARN_ON(!VALID_PAGE(vcpu->arch.mmu.root_hpa));

if (page_fault_handle_page_track(vcpu, error_code, gfn))
- return 1;
+ return RET_PF_EMULATE;

r = mmu_topup_memory_caches(vcpu);
if (r)
@@ -3892,13 +3889,13 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
}

if (fast_page_fault(vcpu, gpa, level, error_code))
- return 0;
+ return RET_PF_RETRY;

mmu_seq = vcpu->kvm->mmu_notifier_seq;
smp_rmb();

if (try_async_pf(vcpu, prefault, gfn, gpa, &pfn, write, &map_writable))
- return 0;
+ return RET_PF_RETRY;

if (handle_abnormal_pfn(vcpu, 0, gfn, pfn, ACC_ALL, &r))
return r;
@@ -3918,7 +3915,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
out_unlock:
spin_unlock(&vcpu->kvm->mmu_lock);
kvm_release_pfn_clean(pfn);
- return 0;
+ return RET_PF_RETRY;
}

static void nonpaging_init_context(struct kvm_vcpu *vcpu,
@@ -4916,25 +4913,25 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u64 error_code,
vcpu->arch.gpa_val = cr2;
}

+ r = RET_PF_INVALID;
if (unlikely(error_code & PFERR_RSVD_MASK)) {
r = handle_mmio_page_fault(vcpu, cr2, direct);
- if (r == RET_MMIO_PF_EMULATE) {
+ if (r == RET_PF_EMULATE) {
emulation_type = 0;
goto emulate;
}
- if (r == RET_MMIO_PF_RETRY)
- return 1;
- if (r < 0)
- return r;
- /* Must be RET_MMIO_PF_INVALID. */
}

- r = vcpu->arch.mmu.page_fault(vcpu, cr2, lower_32_bits(error_code),
- false);
+ if (r == RET_PF_INVALID) {
+ r = vcpu->arch.mmu.page_fault(vcpu, cr2, lower_32_bits(error_code),
+ false);
+ WARN_ON(r == RET_PF_INVALID);
+ }
+
+ if (r == RET_PF_RETRY)
+ return 1;
if (r < 0)
return r;
- if (!r)
- return 1;

/*
* Before emulating the instruction, check if the error code
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 86b68dc5a649..7d6be951443e 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -592,7 +592,7 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
struct kvm_mmu_page *sp = NULL;
struct kvm_shadow_walk_iterator it;
unsigned direct_access, access = gw->pt_access;
- int top_level, emulate;
+ int top_level, ret;

direct_access = gw->pte_access;

@@ -658,15 +658,15 @@ static int FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
}

clear_sp_write_flooding_count(it.sptep);
- emulate = mmu_set_spte(vcpu, it.sptep, gw->pte_access, write_fault,
- it.level, gw->gfn, pfn, prefault, map_writable);
+ ret = mmu_set_spte(vcpu, it.sptep, gw->pte_access, write_fault,
+ it.level, gw->gfn, pfn, prefault, map_writable);
FNAME(pte_prefetch)(vcpu, gw, it.sptep);

- return emulate;
+ return ret;

out_gpte_changed:
kvm_release_pfn_clean(pfn);
- return 0;
+ return RET_PF_RETRY;
}

/*
@@ -761,12 +761,12 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
if (!prefault)
inject_page_fault(vcpu, &walker.fault);

- return 0;
+ return RET_PF_RETRY;
}

if (page_fault_handle_page_track(vcpu, error_code, walker.gfn)) {
shadow_page_table_clear_flood(vcpu, addr);
- return 1;
+ return RET_PF_EMULATE;
}

vcpu->arch.write_fault_to_shadow_pgtable = false;
@@ -788,7 +788,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,

if (try_async_pf(vcpu, prefault, walker.gfn, addr, &pfn, write_fault,
&map_writable))
- return 0;
+ return RET_PF_RETRY;

if (handle_abnormal_pfn(vcpu, addr, walker.gfn, pfn, walker.pte_access, &r))
return r;
@@ -833,7 +833,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
out_unlock:
spin_unlock(&vcpu->kvm->mmu_lock);
kvm_release_pfn_clean(pfn);
- return 0;
+ return RET_PF_EMULATE;
}

static gpa_t FNAME(get_level1_sp_gpa)(struct kvm_mmu_page *sp)
--
1.8.3.1