Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1728847

[PATCH 4.9 25/32] s390/mm: avoid empty zero pages for KVM guests to avoid postcopy hangs

Show all headers | View raw

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Christian Borntraeger <borntraeger@de.ibm.com>

commit fa41ba0d08de7c975c3e94d0067553f9b934221f upstream.

Right now there is a potential hang situation for postcopy migrations,
if the guest is enabling storage keys on the target system during the
postcopy process.

For storage key virtualization, we have to forbid the empty zero page as
the storage key is a property of the physical page frame.  As we enable
storage key handling lazily we then drop all mappings for empty zero
pages for lazy refaulting later on.

This does not work with the postcopy migration, which relies on the
empty zero page never triggering a fault again in the future. The reason
is that postcopy migration will simply read a page on the target system
if that page is a known zero page to fault in an empty zero page.  At
the same time postcopy remembers that this page was already transferred
- so any future userfault on that page will NOT be retransmitted again
to avoid races.

If now the guest enters the storage key mode while in postcopy, we will
break this assumption of postcopy.

The solution is to disable the empty zero page for KVM guests early on
and not during storage key enablement. With this change, the postcopy
migration process is guaranteed to start after no zero pages are left.

As guest pages are very likely not empty zero pages anyway the memory
overhead is also pretty small.

While at it this also adds proper page table locking to the zero page
removal.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/s390/include/asm/pgtable.h |    2 +-
 arch/s390/mm/gmap.c             |   39 ++++++++++++++++++++++++++++++++-------
 2 files changed, 33 insertions(+), 8 deletions(-)

--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -480,7 +480,7 @@ static inline int mm_alloc_pgste(struct
  * In the case that a guest uses storage keys
  * faults should no longer be backed by zero pages
  */
-#define mm_forbids_zeropage mm_use_skey
+#define mm_forbids_zeropage mm_has_pgste
 static inline int mm_use_skey(struct mm_struct *mm)
 {
 #ifdef CONFIG_PGSTE
--- a/arch/s390/mm/gmap.c
+++ b/arch/s390/mm/gmap.c
@@ -2125,6 +2125,37 @@ static inline void thp_split_mm(struct m
 }
 
 /*
+ * Remove all empty zero pages from the mapping for lazy refaulting
+ * - This must be called after mm->context.has_pgste is set, to avoid
+ *   future creation of zero pages
+ * - This must be called after THP was enabled
+ */
+static int __zap_zero_pages(pmd_t *pmd, unsigned long start,
+			   unsigned long end, struct mm_walk *walk)
+{
+	unsigned long addr;
+
+	for (addr = start; addr != end; addr += PAGE_SIZE) {
+		pte_t *ptep;
+		spinlock_t *ptl;
+
+		ptep = pte_offset_map_lock(walk->mm, pmd, addr, &ptl);
+		if (is_zero_pfn(pte_pfn(*ptep)))
+			ptep_xchg_direct(walk->mm, addr, ptep, __pte(_PAGE_INVALID));
+		pte_unmap_unlock(ptep, ptl);
+	}
+	return 0;
+}
+
+static inline void zap_zero_pages(struct mm_struct *mm)
+{
+	struct mm_walk walk = { .pmd_entry = __zap_zero_pages };
+
+	walk.mm = mm;
+	walk_page_range(0, TASK_SIZE, &walk);
+}
+
+/*
  * switch on pgstes for its userspace process (for kvm)
  */
 int s390_enable_sie(void)
@@ -2141,6 +2172,7 @@ int s390_enable_sie(void)
 	mm->context.has_pgste = 1;
 	/* split thp mappings and disable thp for future mappings */
 	thp_split_mm(mm);
+	zap_zero_pages(mm);
 	up_write(&mm->mmap_sem);
 	return 0;
 }
@@ -2153,13 +2185,6 @@ EXPORT_SYMBOL_GPL(s390_enable_sie);
 static int __s390_enable_skey(pte_t *pte, unsigned long addr,
 			      unsigned long next, struct mm_walk *walk)
 {
-	/*
-	 * Remove all zero page mappings,
-	 * after establishing a policy to forbid zero page mappings
-	 * following faults for that page will get fresh anonymous pages
-	 */
-	if (is_zero_pfn(pte_pfn(*pte)))
-		ptep_xchg_direct(walk->mm, addr, pte, __pte(_PAGE_INVALID));
 	/* Clear storage key */
 	ptep_zap_key(walk->mm, addr, pte);
 	return 0;

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 4.9 00/32] 4.9.49-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 12/32] iio: adc: ti-ads1015: dont return invalid value from buffer setup callbacks Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 28/32] cs5536: add support for IDE controller variant Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 03/32] usb: Add device quirk for Logitech HD Pro Webcam C920-C Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 25/32] s390/mm: avoid empty zero pages for KVM guests to avoid postcopy hangs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 04/32] usb:xhci:Fix regression when ATI chipsets detected Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 18/32] Input: trackpoint - assume 3 buttons when buttons detection fails Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 09/32] iio: adc: ti-ads1015: fix scale information for ADS1115 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 30/32] scsi: sg: recheck MMAP_IO request length with lock held Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 16/32] intel_th: pci: Add Cannon Lake PCH-LP support Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 26/32] drm/nouveau/pci/msi: disable MSI on big-endian platforms by default Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 07/32] staging/rts5208: fix incorrect shift to extract upper nybble Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 27/32] workqueue: Fix flag collision Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 29/32] scsi: sg: protect against races between mmap() and SG_SET_RESERVED_SIZE Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:30 +0200
  [PATCH 4.9 20/32] Bluetooth: Add support of 13d3:3494 RTL8723BE device Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 17/32] ath10k: fix memory leak in rx ring buffer allocation Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 14/32] driver core: bus: Fix a potential double free Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 11/32] iio: adc: ti-ads1015: avoid getting stale result after runtime resume Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 15/32] intel_th: pci: Add Cannon Lake PCH-H support Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 13/32] iio: adc: ti-ads1015: add adequate wait time to get correct conversion Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 21/32] iwlwifi: pci: add new PCI ID for 7265D Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 10/32] iio: adc: ti-ads1015: enable conversion when CONFIG_PM is not set Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 19/32] rtlwifi: rtl_pci_probe: Fix fail path of _rtl_pci_find_adapter Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  [PATCH 4.9 01/32] usb: quirks: add delay init quirk for Corsair Strafe RGB keyboard Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-09-08 15:40 +0200
  Re: [PATCH 4.9 00/32] 4.9.49-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2017-09-08 20:40 +0200
  Re: [PATCH 4.9 00/32] 4.9.49-stable review Guenter Roeck <linux@roeck-us.net> - 2017-09-09 15:50 +0200

csiph-web