Skip to contentSkip to conversation

KVM: MMU: Don't flush shadow when enabling dirty tracking

PATCH Applied Avi Kivity 3 participants 5 messages Dec 27, 2010 – Jan 4, 2011

Subsystems (3)
lore
Share & tools
lore

Avi Kivity

author
#0 lore raw

PATCH v2 1/1 KVM: MMU: Don't flush shadow when enabling dirty tracking

Instead, drop large mappings, which were the reason we dropped shadow.

---

v2: maintain largepage stats

 arch/x86/kvm/mmu.c  |   12 ++++++++----
 virt/kvm/kvm_main.c |    7 +------
 2 files changed, 9 insertions(+), 10 deletions(-)
arch/x86/kvm/mmu.c +8 -4
index 43bd5e3..dc36088 100644
@@ -3444,14 +3444,18 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
 		if (!test_bit(slot, sp->slot_bitmap))
 			continue;
 
-		if (sp->role.level != PT_PAGE_TABLE_LEVEL)
-			continue;
-
 		pt = sp->spt;
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
+		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+			if (sp->role.level != PT_PAGE_TABLE_LEVEL
+			    && is_large_pte(pt[i])) {
+				drop_spte(kvm, &pt[i],
+					  shadow_trap_nonpresent_pte);
+				--kvm->stat.lpages;
+			}
 			/* avoid RMW */
 			if (is_writable_pte(pt[i]))
 				update_spte(&pt[i], pt[i] & ~PT_WRITABLE_MASK);
+		}
 	}
 	kvm_flush_remote_tlbs(kvm);
 }
virt/kvm/kvm_main.c +1 -6
index b1b6cbb..b3bfeb8 100644
@@ -586,7 +586,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 			    struct kvm_userspace_memory_region *mem,
 			    int user_alloc)
 {
-	int r, flush_shadow = 0;
+	int r;
 	gfn_t base_gfn;
 	unsigned long npages;
 	unsigned long i;
@@ -706,8 +706,6 @@ skip_lpage:
 		if (kvm_create_dirty_bitmap(&new) < 0)
 			goto out_free;
 		/* destroy any largepage mappings for dirty tracking */
-		if (old.npages)
-			flush_shadow = 1;
 	}
 #else  /* not defined CONFIG_S390 */
 	new.user_alloc = user_alloc;
@@ -778,9 +776,6 @@ skip_lpage:
 	kvm_free_physmem_slot(&old, &new);
 	kfree(old_memslots);
 
-	if (flush_shadow)
-		kvm_arch_flush_shadow(kvm);
-
 	return 0;
 
 out_free:
Signature · 1 line
1.7.1
#1 lore raw
Quoted 24 lines · On Mon, 27 Dec 2010 12:08:45 +0200 Avi Kivity <[email protected]> wrote:

Instead, drop large mappings, which were the reason we dropped shadow.

---

v2: maintain largepage stats

 arch/x86/kvm/mmu.c  |   12 ++++++++----
 virt/kvm/kvm_main.c |    7 +------
 2 files changed, 9 insertions(+), 10 deletions(-)
arch/x86/kvm/mmu.c +0 -4
index 43bd5e3..dc36088 100644
@@ -3444,14 +3444,18 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
 		if (!test_bit(slot, sp->slot_bitmap))
 			continue;
 
-		if (sp->role.level != PT_PAGE_TABLE_LEVEL)
-			continue;
-
 		pt = sp->spt;
-		for (i = 0; i < PT64_ENT_PER_PAGE; ++i)

arch/x86/kvm/mmu.c void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) · line 3457

Quoted 10 lines
arch/x86/kvm/mmu.c +7 -0
+		for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+			if (sp->role.level != PT_PAGE_TABLE_LEVEL
+			    && is_large_pte(pt[i])) {
+				drop_spte(kvm, &pt[i],
+					  shadow_trap_nonpresent_pte);
+				--kvm->stat.lpages;
+			}
 			/* avoid RMW */
 			if (is_writable_pte(pt[i]))
 				update_spte(&pt[i], pt[i] & ~PT_WRITABLE_MASK);

I'm a bit confused here.
rmap_write_protect() does drop_spte() but update_spte() in /* check for huge page mappings */ loop.

I may be able to learn something from your share code later on.

Thanks,
Takuya

Quoted 43 lines
arch/x86/kvm/mmu.c +1 -0
+		}
 	}
 	kvm_flush_remote_tlbs(kvm);
 }
virt/kvm/kvm_main.c +1 -6
index b1b6cbb..b3bfeb8 100644
@@ -586,7 +586,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 			    struct kvm_userspace_memory_region *mem,
 			    int user_alloc)
 {
-	int r, flush_shadow = 0;
+	int r;
 	gfn_t base_gfn;
 	unsigned long npages;
 	unsigned long i;
@@ -706,8 +706,6 @@ skip_lpage:
 		if (kvm_create_dirty_bitmap(&new) < 0)
 			goto out_free;
 		/* destroy any largepage mappings for dirty tracking */
-		if (old.npages)
-			flush_shadow = 1;
 	}
 #else  /* not defined CONFIG_S390 */
 	new.user_alloc = user_alloc;
@@ -778,9 +776,6 @@ skip_lpage:
 	kvm_free_physmem_slot(&old, &new);
 	kfree(old_memslots);
 
-	if (flush_shadow)
-		kvm_arch_flush_shadow(kvm);
-
 	return 0;
 
 out_free:
Signature · 6 lines
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 1 line
Takuya Yoshikawa <[email protected]>
author
#2 lore raw
Quoted 38 lines · Takuya Yoshikawa
Quoted 19 lines · On Mon, 27 Dec 2010 12:08:45 +0200 Avi Kivity<[email protected]> wrote:

Instead, drop large mappings, which were the reason we dropped shadow.

---

v2: maintain largepage stats

  arch/x86/kvm/mmu.c  |   12 ++++++++----
  virt/kvm/kvm_main.c |    7 +------
  2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 43bd5e3..dc36088 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -3444,14 +3444,18 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
if (!test_bit(slot, sp->slot_bitmap))
continue;

Quoted 15 lines

- if (sp->role.level != PT_PAGE_TABLE_LEVEL)
- continue;
-
pt = sp->spt;
- for (i = 0; i< PT64_ENT_PER_PAGE; ++i)
+ for (i = 0; i< PT64_ENT_PER_PAGE; ++i) {
+ if (sp->role.level != PT_PAGE_TABLE_LEVEL
+ && is_large_pte(pt[i])) {
+ drop_spte(kvm,&pt[i],
+ shadow_trap_nonpresent_pte);
+ --kvm->stat.lpages;
+ }
/* avoid RMW */
if (is_writable_pte(pt[i]))
update_spte(&pt[i], pt[i]& ~PT_WRITABLE_MASK);

Quoted 4 lines

I'm a bit confused here.
rmap_write_protect() does drop_spte() but update_spte() in /* check for huge page mappings */ loop.

rmap_write_protect() protects a 4k page, so it uses update_spte() to drop the W bit, and drops large mappings that point to the same 4k page. We can't drop the W bit from the large page, because that will write protect 511 unrelated pages.

Quoted 2 lines

I may be able to learn something from your share code later on.

Unlikely - I was thinking of sharing the

Quoted 3 lines

+ drop_spte(kvm,&pt[i],
+ shadow_trap_nonpresent_pte);
+ --kvm->stat.lpages;

bits. Now I'm thinking of dropping kvm->stat, I think we have all the information it provides covered by tracepoints.

Signature · 1 line
error compiling committee.c: too many arguments to function
#3 lore raw
Quoted 4 lines · On Mon, 27 Dec 2010 12:43:59 +0200 Avi Kivity <[email protected]> wrote:

rmap_write_protect() protects a 4k page, so it uses update_spte() to drop the W bit, and drops large mappings that point to the same 4k page. We can't drop the W bit from the large page, because that will write protect 511 unrelated pages.

I see, I'll reread these from that sense.

Thank you for the explanation.

Takuya

Quoted 15 lines
Quoted 3 lines

I may be able to learn something from your share code later on.

Unlikely - I was thinking of sharing the

Quoted 3 lines

+ drop_spte(kvm,&pt[i],
+ shadow_trap_nonpresent_pte);
+ --kvm->stat.lpages;

bits. Now I'm thinking of dropping kvm->stat, I think we have all the information it provides covered by tracepoints.

#4 lore raw
In reply to Avi Kivity
Quoted 11 lines · Avi Kivity

Instead, drop large mappings, which were the reason we dropped shadow.

---

v2: maintain largepage stats

 arch/x86/kvm/mmu.c  |   12 ++++++++----
 virt/kvm/kvm_main.c |    7 +------
 2 files changed, 9 insertions(+), 10 deletions(-)

Applied, thanks.