A typical demand page/copy on write pattern is:
- page fault on vaddr
- kvm propagates fault to guest
- guest handles fault, updates pte
- kvm traps write, clears shadow pte, resumes guest
- guest returns to userspace, re-faults on same vaddr
- kvm installs shadow pte, resumes guest
- guest continues
So, three vmexits for a single guest page fault. But if instead of clearing
the page table entry, we update to correspond to the value that the guest
has just written, we eliminate the third vmexit.
This patch does exactly that, reducing kbuild time by about 10%.
Signed-off-by: Avi Kivity <[email protected]>
---
drivers/kvm/mmu.c | 15 +++++++++++++++
drivers/kvm/paging_tmpl.h | 15 +++++++++++++++
2 files changed, 30 insertions(+), 0 deletions(-)
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 23dc461..9ec3df9 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
@@ -1137,6 +1137,20 @@ static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu,
*spte = 0;
}
+static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu,
+ struct kvm_mmu_page *page,
+ u64 *spte,
+ const void *new, int bytes)
+{
+ if (page->role.level != PT_PAGE_TABLE_LEVEL)
+ return;
+
+ if (page->role.glevels == PT32_ROOT_LEVEL)
+ paging32_update_pte(vcpu, page, spte, new, bytes);
+ else
+ paging64_update_pte(vcpu, page, spte, new, bytes);
+}
+
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
const u8 *old, const u8 *new, int bytes)
{
@@ -1212,6 +1226,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
spte += page_offset / sizeof(*spte);
while (npte--) {
mmu_pte_write_zap_pte(vcpu, page, spte);
+ mmu_pte_write_new_pte(vcpu, page, spte, new, bytes);
++spte;
}
}
diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
index bc64cce..10ba0a8 100644
--- a/drivers/kvm/paging_tmpl.h
+++ b/drivers/kvm/paging_tmpl.h
@@ -202,6 +202,21 @@ static void FNAME(set_pte)(struct kvm_vcpu *vcpu, u64 guest_pte,
guest_pte & PT_DIRTY_MASK, access_bits, gfn);
}
+static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
+ u64 *spte, const void *pte, int bytes)
+{
+ pt_element_t gpte;
+
+ if (bytes < sizeof(pt_element_t))
+ return;
+ gpte = *(const pt_element_t *)pte;
+ if (~gpte & (PT_PRESENT_MASK | PT_ACCESSED_MASK))
+ return;
+ pgprintk("%s: gpte %llx spte %p\n", __FUNCTION__, (u64)gpte, spte);
+ FNAME(set_pte)(vcpu, gpte, spte, 6,
+ (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT);
+}
+
static void FNAME(set_pde)(struct kvm_vcpu *vcpu, u64 guest_pde,
u64 *shadow_pte, u64 access_bits, gfn_t gfn)
{
--
1.5.0.6
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
