In fork() (or when we protect a page that is no longer a page table), we can
experience floods of writes to a page, which have to be emulated. This is
expensive.
So, if we detect such a flood, zap the page so subsequent writes can proceed
natively.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Index: linux-2.6/drivers/kvm/mmu.c
===================================================================
--- linux-2.6.orig/drivers/kvm/mmu.c
+++ linux-2.6/drivers/kvm/mmu.c
@@ -969,8 +969,17 @@ void kvm_mmu_pre_write(struct kvm_vcpu *
unsigned page_offset;
unsigned misaligned;
int level;
+ int flooded = 0;
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
+ if (gfn == vcpu->last_pt_write_gfn) {
+ ++vcpu->last_pt_write_count;
+ if (vcpu->last_pt_write_count >= 3)
+ flooded = 1;
+ } else {
+ vcpu->last_pt_write_gfn = gfn;
+ vcpu->last_pt_write_count = 1;
+ }
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
bucket = &vcpu->kvm->mmu_page_hash[index];
hlist_for_each_entry_safe(page, node, n, bucket, hash_link) {
@@ -978,11 +987,16 @@ void kvm_mmu_pre_write(struct kvm_vcpu *
continue;
pte_size = page->role.glevels == PT32_ROOT_LEVEL ? 4 : 8;
misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1);
- if (misaligned) {
+ if (misaligned || flooded) {
/*
* Misaligned accesses are too much trouble to fix
* up; also, they usually indicate a page is not used
* as a page table.
+ *
+ * If we're seeing too many writes to a page,
+ * it may no longer be a page table, or we may be
+ * forking, in which case it is better to unmap the
+ * page.
*/
pgprintk("misaligned: gpa %llx bytes %d role %x\n",
gpa, bytes, page->role.word);
Index: linux-2.6/drivers/kvm/kvm.h
===================================================================
--- linux-2.6.orig/drivers/kvm/kvm.h
+++ linux-2.6/drivers/kvm/kvm.h
@@ -238,6 +238,9 @@ struct kvm_vcpu {
struct kvm_mmu_page page_header_buf[KVM_NUM_MMU_PAGES];
struct kvm_mmu mmu;
+ gfn_t last_pt_write_gfn;
+ int last_pt_write_count;
+
struct kvm_guest_debug guest_debug;
char fx_buf[FX_BUF_SIZE];
-
