linux-kernel - [PATCH RFC 1/4] mm: Introduce PTE

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: <20210807032521.7591-2-peterx@redhat.com>
Date:   Fri,  6 Aug 2021 23:25:18 -0400
From:   Peter Xu <peterx@...hat.com>
To:     linux-kernel@...r.kernel.org, linux-mm@...ck.org
Cc:     Alistair Popple <apopple@...dia.com>,
        Tiberiu Georgescu <tiberiu.georgescu@...anix.com>,
        ivan.teterevkov@...anix.com,
        Mike Rapoport <rppt@...ux.vnet.ibm.com>,
        Hugh Dickins <hughd@...gle.com>, peterx@...hat.com,
        Matthew Wilcox <willy@...radead.org>,
        Andrea Arcangeli <aarcange@...hat.com>,
        David Hildenbrand <david@...hat.com>,
        "Kirill A . Shutemov" <kirill@...temov.name>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Mike Kravetz <mike.kravetz@...cle.com>
Subject: [PATCH RFC 1/4] mm: Introduce PTE_MARKER swap entry

This patch introduces a new swap entry type called PTE_MARKER.  It can be
installed for any pte that maps a file-backed memory when the pte is
temporarily zapped, so as to maintain per-pte information.

The information that kept in the pte is called a "marker".  Here we define the
marker as "unsigned long" just to match pgoff_t, however it will only work if
it still fits in swp_offset(), which is e.g. currently 58 bits on x86_64.

The first marker bit that is introduced together with the new swap pte is the
PAGEOUT marker.  When that bit is set, it means this pte used to point to a
page which got swapped out.  It's mostly a definition so the swap type is not
totally nothing, however the functions are not implemented yet to handle the
new swap type.

A new config CONFIG_PTE_MARKER is introduced too; it's by default off.

Signed-off-by: Peter Xu <peterx@...hat.com>
---
 include/linux/swap.h    | 14 ++++++++++++-
 include/linux/swapops.h | 45 +++++++++++++++++++++++++++++++++++++++++
 mm/Kconfig              | 17 ++++++++++++++++
 3 files changed, 75 insertions(+), 1 deletion(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 6f5a43251593..545dc8e0b0fb 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -55,6 +55,18 @@ static inline int current_is_kswapd(void)
  * actions on faults.
  */
 
+/*
+ * PTE markers are used to persist information onto PTEs that are mapped with
+ * file-backed memories.
+ */
+#ifdef CONFIG_PTE_MARKER
+#define SWP_PTE_MARKER_NUM 1
+#define SWP_PTE_MARKER     (MAX_SWAPFILES + SWP_HWPOISON_NUM + \
+			    SWP_MIGRATION_NUM + SWP_DEVICE_NUM)
+#else
+#define SWP_PTE_MARKER_NUM 0
+#endif
+
 /*
  * Unaddressable device memory support. See include/linux/hmm.h and
  * Documentation/vm/hmm.rst. Short description is we need struct pages for
@@ -100,7 +112,7 @@ static inline int current_is_kswapd(void)
 
 #define MAX_SWAPFILES \
 	((1 << MAX_SWAPFILES_SHIFT) - SWP_DEVICE_NUM - \
-	SWP_MIGRATION_NUM - SWP_HWPOISON_NUM)
+	SWP_MIGRATION_NUM - SWP_HWPOISON_NUM - SWP_PTE_MARKER_NUM)
 
 /*
  * Magic header for a swap area. The first part of the union is
diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index d356ab4047f7..3fec83449e1e 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -247,6 +247,51 @@ static inline int is_writable_migration_entry(swp_entry_t entry)
 
 #endif
 
+#ifdef CONFIG_PTE_MARKER
+
+#ifdef CONFIG_PTE_MARKER_PAGEOUT
+/* When this bit is set, it means this page is swapped out previously */
+#define  PTE_MARKER_PAGEOUT  (1UL << 0)
+#else
+#define  PTE_MARKER_PAGEOUT  0
+#endif
+
+#define  PTE_MARKER_MASK     (PTE_MARKER_PAGEOUT)
+
+static inline swp_entry_t make_pte_marker_entry(unsigned long marker)
+{
+	return swp_entry(SWP_PTE_MARKER, marker);
+}
+
+static inline bool is_pte_marker_entry(swp_entry_t entry)
+{
+	return swp_type(entry) == SWP_PTE_MARKER;
+}
+
+static inline unsigned long pte_marker_get(swp_entry_t entry)
+{
+	return swp_offset(entry) & PTE_MARKER_MASK;
+}
+
+#else /* CONFIG_PTE_MARKER */
+
+static inline swp_entry_t make_pte_marker_entry(unsigned long marker)
+{
+	return swp_entry(0, 0);
+}
+
+static inline bool is_pte_marker_entry(swp_entry_t entry)
+{
+	return false;
+}
+
+static inline unsigned long pte_marker_get(swp_entry_t entry)
+{
+	return 0;
+}
+
+#endif /* CONFIG_PTE_MARKER */
+
 static inline struct page *pfn_swap_entry_to_page(swp_entry_t entry)
 {
 	struct page *p = pfn_to_page(swp_offset(entry));
diff --git a/mm/Kconfig b/mm/Kconfig
index 40a9bfcd5062..6043d8f1c066 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -889,4 +889,21 @@ config IO_MAPPING
 config SECRETMEM
 	def_bool ARCH_HAS_SET_DIRECT_MAP && !EMBEDDED
 
+config PTE_MARKER
+        def_bool n
+        bool "Marker PTEs support"
+
+	help
+	  Allows to create marker PTEs for file-backed memory.
+
+config PTE_MARKER_PAGEOUT
+        def_bool n
+        depends on PTE_MARKER
+        bool "Shmem pagemap PM_SWAP support"
+
+	help
+	  Allows to create marker PTEs for file-backed memory when the page is
+	  swapped out.  It's required for pagemap to work correctly with shmem
+	  on page swapping.
+
 endmenu
-- 
2.32.0