KVM: Reduce misfirings of the fork detector

The kvm mmu tries to detects forks by looking for repeated writes to a
page table.  If it sees a fork, it unshadows the page table so the page
table copying can proceed at native speed instead of being emulated.

However, the detector also triggered on simple demand paging access patterns:
a linear walk of memory would of course cause repeated writes to the same
pagetable page, causing it to unshadow prematurely.

Fix by resetting the fork detector if we detect a demand fault.

Signed-off-by: Avi Kivity <avi@qumranet.com>

diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
index 73ffbff..bc64cce 100644 (file)
--- a/drivers/kvm/paging_tmpl.h
+++ b/drivers/kvm/paging_tmpl.h
@@ -421,6 +421,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
                pgprintk("%s: guest page fault\n", __FUNCTION__);
                inject_page_fault(vcpu, addr, walker.error_code);
                FNAME(release_walker)(&walker);
+               vcpu->last_pt_write_count = 0; /* reset fork detector */
                return 0;
        }
 
@@ -442,6 +443,9 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
 
        FNAME(release_walker)(&walker);
 
+       if (!write_pt)
+               vcpu->last_pt_write_count = 0; /* reset fork detector */
+
        /*
         * mmio: emulate if accessible, otherwise its a guest fault.
         */