linux-kernel - [PATCH 6/6] KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20120628110235.7410445e.yoshikawa.takuya@oss.ntt.co.jp>
Date:	Thu, 28 Jun 2012 11:02:35 +0900
From:	Takuya Yoshikawa <yoshikawa.takuya@....ntt.co.jp>
To:	avi@...hat.com, mtosatti@...hat.com
Cc:	agraf@...e.de, paulus@...ba.org, aarcange@...hat.com,
	kvm@...r.kernel.org, kvm-ppc@...r.kernel.org,
	linux-kernel@...r.kernel.org, takuya.yoshikawa@...il.com
Subject: [PATCH 6/6] KVM: MMU: Avoid handling same rmap_pde in
 kvm_handle_hva_range()

When we invalidate a THP page, we call the handler with the same
rmap_pde argument 512 times in the following loop:

  for each guest page in the range
    for each level
      unmap using rmap

This patch avoids these extra handler calls by changing the loop order
like this:

  for each level
    for each rmap in the range
      unmap using rmap

With the preceding patches in the patch series, this made THP page
invalidation more than 5 times faster on our x86 host: the host became
more responsive during swapping the guest's memory as a result.

Note: in the new code we could not use trace_kvm_age_page(), so we just
dropped the point from kvm_handle_hva_range().

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@....ntt.co.jp>
---
 arch/x86/kvm/mmu.c |   37 +++++++++++++++++++------------------
 1 files changed, 19 insertions(+), 18 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 477b3da..524f7c0 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -1192,8 +1192,7 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 					       unsigned long data))
 {
 	int j;
-	int ret;
-	int retval = 0;
+	int ret = 0;
 	struct kvm_memslots *slots;
 	struct kvm_memory_slot *memslot;
 
@@ -1201,7 +1200,7 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 
 	kvm_for_each_memslot(memslot, slots) {
 		unsigned long hva_start, hva_end;
-		gfn_t gfn, gfn_end;
+		gfn_t gfn_start, gfn_end;
 
 		hva_start = max(start, memslot->userspace_addr);
 		hva_end = min(end, memslot->userspace_addr +
@@ -1210,29 +1209,31 @@ static int kvm_handle_hva_range(struct kvm *kvm,
 			continue;
 		/*
 		 * {gfn(page) | page intersects with [hva_start, hva_end)} =
-		 * {gfn, gfn+1, ..., gfn_end-1}.
+		 * {gfn_start, gfn_start+1, ..., gfn_end-1}.
 		 */
-		gfn = hva_to_gfn_memslot(hva_start, memslot);
+		gfn_start = hva_to_gfn_memslot(hva_start, memslot);
 		gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot);
 
-		for (; gfn < gfn_end; ++gfn) {
-			ret = 0;
+		for (j = PT_PAGE_TABLE_LEVEL;
+		     j < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++j) {
+			unsigned long idx, idx_end;
+			unsigned long *rmapp;
 
-			for (j = PT_PAGE_TABLE_LEVEL;
-			     j < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++j) {
-				unsigned long *rmapp;
+			/*
+			 * {idx(page_j) | page_j intersects with
+			 *  [hva_start, hva_end)} ={idx, idx+1, ..., idx_end}.
+			 */
+			idx = gfn_to_index(gfn_start, memslot->base_gfn, j);
+			idx_end = gfn_to_index(gfn_end - 1, memslot->base_gfn, j);
 
-				rmapp = __gfn_to_rmap(gfn, j, memslot);
-				ret |= handler(kvm, rmapp, data);
-			}
-			trace_kvm_age_page(memslot->userspace_addr +
-					(gfn - memslot->base_gfn) * PAGE_SIZE,
-					memslot, ret);
-			retval |= ret;
+			rmapp = __gfn_to_rmap(gfn_start, j, memslot);
+
+			for (; idx <= idx_end; ++idx)
+				ret |= handler(kvm, rmapp++, data);
 		}
 	}
 
-	return retval;
+	return ret;
 }
 
 static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
-- 
1.7.5.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/