Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/djm/tmem

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/djm/tmem:
  xen: cleancache shim to Xen Transcendent Memory
  ocfs2: add cleancache support
  ext4: add cleancache support
  btrfs: add cleancache support
  ext3: add cleancache support
  mm/fs: add hooks to support cleancache
  mm: cleancache core ops functions and config
  fs: add field to superblock to support cleancache
  mm/fs: cleancache documentation

Fix up trivial conflict in fs/btrfs/extent_io.c due to includes

Documentation/ABI/testing/sysfs-kernel-mm-cleancache

@@ -0,0 +1,11 @@
+What:		/sys/kernel/mm/cleancache/
+Date:		April 2011
+Contact:	Dan Magenheimer <dan.magenheimer@oracle.com>
+Description:
+		/sys/kernel/mm/cleancache/ contains a number of files which
+		record a count of various cleancache operations
+		(sum across all filesystems):
+			succ_gets
+			failed_gets
+			puts
+			flushes

Documentation/vm/cleancache.txt

@@ -0,0 +1,278 @@
+MOTIVATION
+
+Cleancache is a new optional feature provided by the VFS layer that
+potentially dramatically increases page cache effectiveness for
+many workloads in many environments at a negligible cost.
+
+Cleancache can be thought of as a page-granularity victim cache for clean
+pages that the kernel's pageframe replacement algorithm (PFRA) would like
+to keep around, but can't since there isn't enough memory.  So when the
+PFRA "evicts" a page, it first attempts to use cleancache code to
+put the data contained in that page into "transcendent memory", memory
+that is not directly accessible or addressable by the kernel and is
+of unknown and possibly time-varying size.
+
+Later, when a cleancache-enabled filesystem wishes to access a page
+in a file on disk, it first checks cleancache to see if it already
+contains it; if it does, the page of data is copied into the kernel
+and a disk access is avoided.
+
+Transcendent memory "drivers" for cleancache are currently implemented
+in Xen (using hypervisor memory) and zcache (using in-kernel compressed
+memory) and other implementations are in development.
+
+FAQs are included below.
+
+IMPLEMENTATION OVERVIEW
+
+A cleancache "backend" that provides transcendent memory registers itself
+to the kernel's cleancache "frontend" by calling cleancache_register_ops,
+passing a pointer to a cleancache_ops structure with funcs set appropriately.
+Note that cleancache_register_ops returns the previous settings so that
+chaining can be performed if desired. The functions provided must conform to
+certain semantics as follows:
+
+Most important, cleancache is "ephemeral".  Pages which are copied into
+cleancache have an indefinite lifetime which is completely unknowable
+by the kernel and so may or may not still be in cleancache at any later time.
+Thus, as its name implies, cleancache is not suitable for dirty pages.
+Cleancache has complete discretion over what pages to preserve and what
+pages to discard and when.
+
+Mounting a cleancache-enabled filesystem should call "init_fs" to obtain a
+pool id which, if positive, must be saved in the filesystem's superblock;
+a negative return value indicates failure.  A "put_page" will copy a
+(presumably about-to-be-evicted) page into cleancache and associate it with
+the pool id, a file key, and a page index into the file.  (The combination
+of a pool id, a file key, and an index is sometimes called a "handle".)
+A "get_page" will copy the page, if found, from cleancache into kernel memory.
+A "flush_page" will ensure the page no longer is present in cleancache;
+a "flush_inode" will flush all pages associated with the specified file;
+and, when a filesystem is unmounted, a "flush_fs" will flush all pages in
+all files specified by the given pool id and also surrender the pool id.
+
+An "init_shared_fs", like init_fs, obtains a pool id but tells cleancache
+to treat the pool as shared using a 128-bit UUID as a key.  On systems
+that may run multiple kernels (such as hard partitioned or virtualized
+systems) that may share a clustered filesystem, and where cleancache
+may be shared among those kernels, calls to init_shared_fs that specify the
+same UUID will receive the same pool id, thus allowing the pages to
+be shared.  Note that any security requirements must be imposed outside
+of the kernel (e.g. by "tools" that control cleancache).  Or a
+cleancache implementation can simply disable shared_init by always
+returning a negative value.
+
+If a get_page is successful on a non-shared pool, the page is flushed (thus
+making cleancache an "exclusive" cache).  On a shared pool, the page
+is NOT flushed on a successful get_page so that it remains accessible to
+other sharers.  The kernel is responsible for ensuring coherency between
+cleancache (shared or not), the page cache, and the filesystem, using
+cleancache flush operations as required.
+
+Note that cleancache must enforce put-put-get coherency and get-get
+coherency.  For the former, if two puts are made to the same handle but
+with different data, say AAA by the first put and BBB by the second, a
+subsequent get can never return the stale data (AAA).  For get-get coherency,
+if a get for a given handle fails, subsequent gets for that handle will
+never succeed unless preceded by a successful put with that handle.
+
+Last, cleancache provides no SMP serialization guarantees; if two
+different Linux threads are simultaneously putting and flushing a page
+with the same handle, the results are indeterminate.  Callers must
+lock the page to ensure serial behavior.
+
+CLEANCACHE PERFORMANCE METRICS
+
+Cleancache monitoring is done by sysfs files in the
+/sys/kernel/mm/cleancache directory.  The effectiveness of cleancache
+can be measured (across all filesystems) with:
+
+succ_gets	- number of gets that were successful
+failed_gets	- number of gets that failed
+puts		- number of puts attempted (all "succeed")
+flushes		- number of flushes attempted
+
+A backend implementatation may provide additional metrics.
+
+FAQ
+
+1) Where's the value? (Andrew Morton)
+
+Cleancache provides a significant performance benefit to many workloads
+in many environments with negligible overhead by improving the
+effectiveness of the pagecache.  Clean pagecache pages are
+saved in transcendent memory (RAM that is otherwise not directly
+addressable to the kernel); fetching those pages later avoids "refaults"
+and thus disk reads.
+
+Cleancache (and its sister code "frontswap") provide interfaces for
+this transcendent memory (aka "tmem"), which conceptually lies between
+fast kernel-directly-addressable RAM and slower DMA/asynchronous devices.
+Disallowing direct kernel or userland reads/writes to tmem
+is ideal when data is transformed to a different form and size (such
+as with compression) or secretly moved (as might be useful for write-
+balancing for some RAM-like devices).  Evicted page-cache pages (and
+swap pages) are a great use for this kind of slower-than-RAM-but-much-
+faster-than-disk transcendent memory, and the cleancache (and frontswap)
+"page-object-oriented" specification provides a nice way to read and
+write -- and indirectly "name" -- the pages.
+
+In the virtual case, the whole point of virtualization is to statistically
+multiplex physical resources across the varying demands of multiple
+virtual machines.  This is really hard to do with RAM and efforts to
+do it well with no kernel change have essentially failed (except in some
+well-publicized special-case workloads).  Cleancache -- and frontswap --
+with a fairly small impact on the kernel, provide a huge amount
+of flexibility for more dynamic, flexible RAM multiplexing.
+Specifically, the Xen Transcendent Memory backend allows otherwise
+"fallow" hypervisor-owned RAM to not only be "time-shared" between multiple
+virtual machines, but the pages can be compressed and deduplicated to
+optimize RAM utilization.  And when guest OS's are induced to surrender
+underutilized RAM (e.g. with "self-ballooning"), page cache pages
+are the first to go, and cleancache allows those pages to be
+saved and reclaimed if overall host system memory conditions allow.
+
+And the identical interface used for cleancache can be used in
+physical systems as well.  The zcache driver acts as a memory-hungry
+device that stores pages of data in a compressed state.  And
+the proposed "RAMster" driver shares RAM across multiple physical
+systems.
+
+2) Why does cleancache have its sticky fingers so deep inside the
+   filesystems and VFS? (Andrew Morton and Christoph Hellwig)
+
+The core hooks for cleancache in VFS are in most cases a single line
+and the minimum set are placed precisely where needed to maintain
+coherency (via cleancache_flush operations) between cleancache,
+the page cache, and disk.  All hooks compile into nothingness if
+cleancache is config'ed off and turn into a function-pointer-
+compare-to-NULL if config'ed on but no backend claims the ops
+functions, or to a compare-struct-element-to-negative if a
+backend claims the ops functions but a filesystem doesn't enable
+cleancache.
+
+Some filesystems are built entirely on top of VFS and the hooks
+in VFS are sufficient, so don't require an "init_fs" hook; the
+initial implementation of cleancache didn't provide this hook.
+But for some filesystems (such as btrfs), the VFS hooks are
+incomplete and one or more hooks in fs-specific code are required.
+And for some other filesystems, such as tmpfs, cleancache may
+be counterproductive.  So it seemed prudent to require a filesystem
+to "opt in" to use cleancache, which requires adding a hook in
+each filesystem.  Not all filesystems are supported by cleancache
+only because they haven't been tested.  The existing set should
+be sufficient to validate the concept, the opt-in approach means
+that untested filesystems are not affected, and the hooks in the
+existing filesystems should make it very easy to add more
+filesystems in the future.
+
+The total impact of the hooks to existing fs and mm files is only
+about 40 lines added (not counting comments and blank lines).
+
+3) Why not make cleancache asynchronous and batched so it can
+   more easily interface with real devices with DMA instead
+   of copying each individual page? (Minchan Kim)
+
+The one-page-at-a-time copy semantics simplifies the implementation
+on both the frontend and backend and also allows the backend to
+do fancy things on-the-fly like page compression and
+page deduplication.  And since the data is "gone" (copied into/out
+of the pageframe) before the cleancache get/put call returns,
+a great deal of race conditions and potential coherency issues
+are avoided.  While the interface seems odd for a "real device"
+or for real kernel-addressable RAM, it makes perfect sense for
+transcendent memory.
+
+4) Why is non-shared cleancache "exclusive"?  And where is the
+   page "flushed" after a "get"? (Minchan Kim)
+
+The main reason is to free up space in transcendent memory and
+to avoid unnecessary cleancache_flush calls.  If you want inclusive,
+the page can be "put" immediately following the "get".  If
+put-after-get for inclusive becomes common, the interface could
+be easily extended to add a "get_no_flush" call.
+
+The flush is done by the cleancache backend implementation.
+
+5) What's the performance impact?
+
+Performance analysis has been presented at OLS'09 and LCA'10.
+Briefly, performance gains can be significant on most workloads,
+especially when memory pressure is high (e.g. when RAM is
+overcommitted in a virtual workload); and because the hooks are
+invoked primarily in place of or in addition to a disk read/write,
+overhead is negligible even in worst case workloads.  Basically
+cleancache replaces I/O with memory-copy-CPU-overhead; on older
+single-core systems with slow memory-copy speeds, cleancache
+has little value, but in newer multicore machines, especially
+consolidated/virtualized machines, it has great value.
+
+6) How do I add cleancache support for filesystem X? (Boaz Harrash)
+
+Filesystems that are well-behaved and conform to certain
+restrictions can utilize cleancache simply by making a call to
+cleancache_init_fs at mount time.  Unusual, misbehaving, or
+poorly layered filesystems must either add additional hooks
+and/or undergo extensive additional testing... or should just
+not enable the optional cleancache.
+
+Some points for a filesystem to consider:
+
+- The FS should be block-device-based (e.g. a ram-based FS such
+  as tmpfs should not enable cleancache)
+- To ensure coherency/correctness, the FS must ensure that all
+  file removal or truncation operations either go through VFS or
+  add hooks to do the equivalent cleancache "flush" operations
+- To ensure coherency/correctness, either inode numbers must
+  be unique across the lifetime of the on-disk file OR the
+  FS must provide an "encode_fh" function.
+- The FS must call the VFS superblock alloc and deactivate routines
+  or add hooks to do the equivalent cleancache calls done there.
+- To maximize performance, all pages fetched from the FS should
+  go through the do_mpag_readpage routine or the FS should add
+  hooks to do the equivalent (cf. btrfs)
+- Currently, the FS blocksize must be the same as PAGESIZE.  This
+  is not an architectural restriction, but no backends currently
+  support anything different.
+- A clustered FS should invoke the "shared_init_fs" cleancache
+  hook to get best performance for some backends.
+
+7) Why not use the KVA of the inode as the key? (Christoph Hellwig)
+
+If cleancache would use the inode virtual address instead of
+inode/filehandle, the pool id could be eliminated.  But, this
+won't work because cleancache retains pagecache data pages
+persistently even when the inode has been pruned from the
+inode unused list, and only flushes the data page if the file
+gets removed/truncated.  So if cleancache used the inode kva,
+there would be potential coherency issues if/when the inode
+kva is reused for a different file.  Alternately, if cleancache
+flushed the pages when the inode kva was freed, much of the value
+of cleancache would be lost because the cache of pages in cleanache
+is potentially much larger than the kernel pagecache and is most
+useful if the pages survive inode cache removal.
+
+8) Why is a global variable required?
+
+The cleancache_enabled flag is checked in all of the frequently-used
+cleancache hooks.  The alternative is a function call to check a static
+variable. Since cleancache is enabled dynamically at runtime, systems
+that don't enable cleancache would suffer thousands (possibly
+tens-of-thousands) of unnecessary function calls per second.  So the
+global variable allows cleancache to be enabled by default at compile
+time, but have insignificant performance impact when cleancache remains
+disabled at runtime.
+
+9) Does cleanache work with KVM?
+
+The memory model of KVM is sufficiently different that a cleancache
+backend may have less value for KVM.  This remains to be tested,
+especially in an overcommitted system.
+
+10) Does cleancache work in userspace?  It sounds useful for
+   memory hungry caches like web browsers.  (Jamie Lokier)
+
+No plans yet, though we agree it sounds useful, at least for
+apps that bypass the page cache (e.g. O_DIRECT).
+
+Last updated: Dan Magenheimer, April 13 2011

arch/x86/include/asm/xen/hypercall.h

@@ -447,6 +447,13 @@ HYPERVISOR_hvm_op(int op, void *arg)
        return _hypercall2(unsigned long, hvm_op, op, arg);
 }
 
+static inline int
+HYPERVISOR_tmem_op(
+	struct tmem_op *op)
+{
+	return _hypercall1(int, tmem_op, op);
+}
+
 static inline void
 MULTI_fpu_taskswitch(struct multicall_entry *mcl, int set)
 {

drivers/xen/Makefile

@@ -1,5 +1,6 @@
 obj-y	+= grant-table.o features.o events.o manage.o balloon.o
 obj-y	+= xenbus/
+obj-y	+= tmem.o
 
 nostackp := $(call cc-option, -fno-stack-protector)
 CFLAGS_features.o			:= $(nostackp)

drivers/xen/tmem.c

@@ -0,0 +1,264 @@
+/*
+ * Xen implementation for transcendent memory (tmem)
+ *
+ * Copyright (C) 2009-2010 Oracle Corp.  All rights reserved.
+ * Author: Dan Magenheimer
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/pagemap.h>
+#include <linux/cleancache.h>
+
+#include <xen/xen.h>
+#include <xen/interface/xen.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+#include <asm/xen/hypervisor.h>
+
+#define TMEM_CONTROL               0
+#define TMEM_NEW_POOL              1
+#define TMEM_DESTROY_POOL          2
+#define TMEM_NEW_PAGE              3
+#define TMEM_PUT_PAGE              4
+#define TMEM_GET_PAGE              5
+#define TMEM_FLUSH_PAGE            6
+#define TMEM_FLUSH_OBJECT          7
+#define TMEM_READ                  8
+#define TMEM_WRITE                 9
+#define TMEM_XCHG                 10
+
+/* Bits for HYPERVISOR_tmem_op(TMEM_NEW_POOL) */
+#define TMEM_POOL_PERSIST          1
+#define TMEM_POOL_SHARED           2
+#define TMEM_POOL_PAGESIZE_SHIFT   4
+#define TMEM_VERSION_SHIFT        24
+
+
+struct tmem_pool_uuid {
+	u64 uuid_lo;
+	u64 uuid_hi;
+};
+
+struct tmem_oid {
+	u64 oid[3];
+};
+
+#define TMEM_POOL_PRIVATE_UUID	{ 0, 0 }
+
+/* flags for tmem_ops.new_pool */
+#define TMEM_POOL_PERSIST          1
+#define TMEM_POOL_SHARED           2
+
+/* xen tmem foundation ops/hypercalls */
+
+static inline int xen_tmem_op(u32 tmem_cmd, u32 tmem_pool, struct tmem_oid oid,
+	u32 index, unsigned long gmfn, u32 tmem_offset, u32 pfn_offset, u32 len)
+{
+	struct tmem_op op;
+	int rc = 0;
+
+	op.cmd = tmem_cmd;
+	op.pool_id = tmem_pool;
+	op.u.gen.oid[0] = oid.oid[0];
+	op.u.gen.oid[1] = oid.oid[1];
+	op.u.gen.oid[2] = oid.oid[2];
+	op.u.gen.index = index;
+	op.u.gen.tmem_offset = tmem_offset;
+	op.u.gen.pfn_offset = pfn_offset;
+	op.u.gen.len = len;
+	set_xen_guest_handle(op.u.gen.gmfn, (void *)gmfn);
+	rc = HYPERVISOR_tmem_op(&op);
+	return rc;
+}
+
+static int xen_tmem_new_pool(struct tmem_pool_uuid uuid,
+				u32 flags, unsigned long pagesize)
+{
+	struct tmem_op op;
+	int rc = 0, pageshift;
+
+	for (pageshift = 0; pagesize != 1; pageshift++)
+		pagesize >>= 1;
+	flags |= (pageshift - 12) << TMEM_POOL_PAGESIZE_SHIFT;
+	flags |= TMEM_SPEC_VERSION << TMEM_VERSION_SHIFT;
+	op.cmd = TMEM_NEW_POOL;
+	op.u.new.uuid[0] = uuid.uuid_lo;
+	op.u.new.uuid[1] = uuid.uuid_hi;
+	op.u.new.flags = flags;
+	rc = HYPERVISOR_tmem_op(&op);
+	return rc;
+}
+
+/* xen generic tmem ops */
+
+static int xen_tmem_put_page(u32 pool_id, struct tmem_oid oid,
+			     u32 index, unsigned long pfn)
+{
+	unsigned long gmfn = xen_pv_domain() ? pfn_to_mfn(pfn) : pfn;
+
+	return xen_tmem_op(TMEM_PUT_PAGE, pool_id, oid, index,
+		gmfn, 0, 0, 0);
+}
+
+static int xen_tmem_get_page(u32 pool_id, struct tmem_oid oid,
+			     u32 index, unsigned long pfn)
+{
+	unsigned long gmfn = xen_pv_domain() ? pfn_to_mfn(pfn) : pfn;
+
+	return xen_tmem_op(TMEM_GET_PAGE, pool_id, oid, index,
+		gmfn, 0, 0, 0);
+}
+
+static int xen_tmem_flush_page(u32 pool_id, struct tmem_oid oid, u32 index)
+{
+	return xen_tmem_op(TMEM_FLUSH_PAGE, pool_id, oid, index,
+		0, 0, 0, 0);
+}
+
+static int xen_tmem_flush_object(u32 pool_id, struct tmem_oid oid)
+{
+	return xen_tmem_op(TMEM_FLUSH_OBJECT, pool_id, oid, 0, 0, 0, 0, 0);
+}
+
+static int xen_tmem_destroy_pool(u32 pool_id)
+{
+	struct tmem_oid oid = { { 0 } };
+
+	return xen_tmem_op(TMEM_DESTROY_POOL, pool_id, oid, 0, 0, 0, 0, 0);
+}
+
+int tmem_enabled;
+
+static int __init enable_tmem(char *s)
+{
+	tmem_enabled = 1;
+	return 1;
+}
+
+__setup("tmem", enable_tmem);
+
+/* cleancache ops */
+
+static void tmem_cleancache_put_page(int pool, struct cleancache_filekey key,
+				     pgoff_t index, struct page *page)
+{
+	u32 ind = (u32) index;
+	struct tmem_oid oid = *(struct tmem_oid *)&key;
+	unsigned long pfn = page_to_pfn(page);
+
+	if (pool < 0)
+		return;
+	if (ind != index)
+		return;
+	mb(); /* ensure page is quiescent; tmem may address it with an alias */
+	(void)xen_tmem_put_page((u32)pool, oid, ind, pfn);
+}
+
+static int tmem_cleancache_get_page(int pool, struct cleancache_filekey key,
+				    pgoff_t index, struct page *page)
+{
+	u32 ind = (u32) index;
+	struct tmem_oid oid = *(struct tmem_oid *)&key;
+	unsigned long pfn = page_to_pfn(page);
+	int ret;
+
+	/* translate return values to linux semantics */
+	if (pool < 0)
+		return -1;
+	if (ind != index)
+		return -1;
+	ret = xen_tmem_get_page((u32)pool, oid, ind, pfn);
+	if (ret == 1)
+		return 0;
+	else
+		return -1;
+}
+
+static void tmem_cleancache_flush_page(int pool, struct cleancache_filekey key,
+				       pgoff_t index)
+{
+	u32 ind = (u32) index;
+	struct tmem_oid oid = *(struct tmem_oid *)&key;
+
+	if (pool < 0)
+		return;
+	if (ind != index)
+		return;
+	(void)xen_tmem_flush_page((u32)pool, oid, ind);
+}
+
+static void tmem_cleancache_flush_inode(int pool, struct cleancache_filekey key)
+{
+	struct tmem_oid oid = *(struct tmem_oid *)&key;
+
+	if (pool < 0)
+		return;
+	(void)xen_tmem_flush_object((u32)pool, oid);
+}
+
+static void tmem_cleancache_flush_fs(int pool)
+{
+	if (pool < 0)
+		return;
+	(void)xen_tmem_destroy_pool((u32)pool);
+}
+
+static int tmem_cleancache_init_fs(size_t pagesize)
+{
+	struct tmem_pool_uuid uuid_private = TMEM_POOL_PRIVATE_UUID;
+
+	return xen_tmem_new_pool(uuid_private, 0, pagesize);
+}
+
+static int tmem_cleancache_init_shared_fs(char *uuid, size_t pagesize)
+{
+	struct tmem_pool_uuid shared_uuid;
+
+	shared_uuid.uuid_lo = *(u64 *)uuid;
+	shared_uuid.uuid_hi = *(u64 *)(&uuid[8]);
+	return xen_tmem_new_pool(shared_uuid, TMEM_POOL_SHARED, pagesize);
+}
+
+static int use_cleancache = 1;
+
+static int __init no_cleancache(char *s)
+{
+	use_cleancache = 0;
+	return 1;
+}
+
+__setup("nocleancache", no_cleancache);
+
+static struct cleancache_ops tmem_cleancache_ops = {
+	.put_page = tmem_cleancache_put_page,
+	.get_page = tmem_cleancache_get_page,
+	.flush_page = tmem_cleancache_flush_page,
+	.flush_inode = tmem_cleancache_flush_inode,
+	.flush_fs = tmem_cleancache_flush_fs,
+	.init_shared_fs = tmem_cleancache_init_shared_fs,
+	.init_fs = tmem_cleancache_init_fs
+};
+
+static int __init xen_tmem_init(void)
+{
+	struct cleancache_ops old_ops;
+
+	if (!xen_domain())
+		return 0;
+#ifdef CONFIG_CLEANCACHE
+	BUG_ON(sizeof(struct cleancache_filekey) != sizeof(struct tmem_oid));
+	if (tmem_enabled && use_cleancache) {
+		char *s = "";
+		old_ops = cleancache_register_ops(&tmem_cleancache_ops);
+		if (old_ops.init_fs != NULL)
+			s = " (WARNING: cleancache_ops overridden)";
+		printk(KERN_INFO "cleancache enabled, RAM provided by "
+				 "Xen Transcendent Memory%s\n", s);
+	}
+#endif
+	return 0;
+}
+
+module_init(xen_tmem_init)

fs/btrfs/extent_io.c

@@ -11,6 +11,7 @@
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
 #include <linux/prefetch.h>
+#include <linux/cleancache.h>
 #include "extent_io.h"
 #include "extent_map.h"
 #include "compat.h"
@@ -2016,6 +2017,13 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 
 	set_page_extent_mapped(page);
 
+	if (!PageUptodate(page)) {
+		if (cleancache_get_page(page) == 0) {
+			BUG_ON(blocksize != PAGE_SIZE);
+			goto out;
+		}
+	}
+
 	end = page_end;
 	while (1) {
 		lock_extent(tree, start, end, GFP_NOFS);
@@ -2149,6 +2157,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 		cur = cur + iosize;
 		page_offset += iosize;
 	}
+out:
 	if (!nr) {
 		if (!PageError(page))
 			SetPageUptodate(page);

fs/btrfs/super.c

@@ -39,6 +39,7 @@
 #include <linux/miscdevice.h>
 #include <linux/magic.h>
 #include <linux/slab.h>
+#include <linux/cleancache.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -624,6 +625,7 @@ static int btrfs_fill_super(struct super_block *sb,
 	sb->s_root = root_dentry;
 
 	save_mount_options(sb, data);
+	cleancache_init_fs(sb);
 	return 0;
 
 fail_close:

fs/buffer.c

@@ -41,6 +41,7 @@
 #include <linux/bitops.h>
 #include <linux/mpage.h>
 #include <linux/bit_spinlock.h>
+#include <linux/cleancache.h>
 
 static int fsync_buffers_list(spinlock_t *lock, struct list_head *list);
 
@@ -269,6 +270,10 @@ void invalidate_bdev(struct block_device *bdev)
 	invalidate_bh_lrus();
 	lru_add_drain_all();	/* make sure all lru add caches are flushed */
 	invalidate_mapping_pages(mapping, 0, -1);
+	/* 99% of the time, we don't need to flush the cleancache on the bdev.
+	 * But, for the strange corners, lets be cautious
+	 */
+	cleancache_flush_inode(mapping);
 }
 EXPORT_SYMBOL(invalidate_bdev);
 

fs/ext3/super.c

@@ -36,6 +36,7 @@
 #include <linux/quotaops.h>
 #include <linux/seq_file.h>
 #include <linux/log2.h>
+#include <linux/cleancache.h>
 
 #include <asm/uaccess.h>
 
@@ -1367,6 +1368,7 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
 	} else {
 		ext3_msg(sb, KERN_INFO, "using internal journal");
 	}
+	cleancache_init_fs(sb);
 	return res;
 }
 

fs/ext4/super.c

@@ -38,6 +38,7 @@
 #include <linux/ctype.h>
 #include <linux/log2.h>
 #include <linux/crc16.h>
+#include <linux/cleancache.h>
 #include <asm/uaccess.h>
 
 #include <linux/kthread.h>
@@ -1948,6 +1949,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
 			EXT4_INODES_PER_GROUP(sb),
 			sbi->s_mount_opt, sbi->s_mount_opt2);
 
+	cleancache_init_fs(sb);
 	return res;
 }
 

fs/mpage.c

@@ -27,6 +27,7 @@
 #include <linux/writeback.h>
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
+#include <linux/cleancache.h>
 
 /*
  * I/O completion handler for multipage BIOs.
@@ -271,6 +272,12 @@ do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages,
 		SetPageMappedToDisk(page);
 	}
 
+	if (fully_mapped && blocks_per_page == 1 && !PageUptodate(page) &&
+	    cleancache_get_page(page) == 0) {
+		SetPageUptodate(page);
+		goto confused;
+	}
+
 	/*
 	 * This page will go to BIO.  Do we need to send this BIO off first?
 	 */

fs/ocfs2/super.c

@@ -41,6 +41,7 @@
 #include <linux/mount.h>
 #include <linux/seq_file.h>
 #include <linux/quotaops.h>
+#include <linux/cleancache.h>
 
 #define CREATE_TRACE_POINTS
 #include "ocfs2_trace.h"
@@ -2352,6 +2353,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
 		mlog_errno(status);
 		goto bail;
 	}
+	cleancache_init_shared_fs((char *)&uuid_net_key, sb);
 
 bail:
 	return status;

fs/super.c

@@ -31,6 +31,7 @@
 #include <linux/mutex.h>
 #include <linux/backing-dev.h>
 #include <linux/rculist_bl.h>
+#include <linux/cleancache.h>
 #include "internal.h"
 
 
@@ -112,6 +113,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		s->s_maxbytes = MAX_NON_LFS;
 		s->s_op = &default_op;
 		s->s_time_gran = 1000000000;
+		s->cleancache_poolid = -1;
 	}
 out:
 	return s;
@@ -177,6 +179,7 @@ void deactivate_locked_super(struct super_block *s)
 {
 	struct file_system_type *fs = s->s_type;
 	if (atomic_dec_and_test(&s->s_active)) {
+		cleancache_flush_fs(s);
 		fs->kill_sb(s);
 		/*
 		 * We need to call rcu_barrier so all the delayed rcu free

include/linux/cleancache.h

@@ -0,0 +1,122 @@
+#ifndef _LINUX_CLEANCACHE_H
+#define _LINUX_CLEANCACHE_H
+
+#include <linux/fs.h>
+#include <linux/exportfs.h>
+#include <linux/mm.h>
+
+#define CLEANCACHE_KEY_MAX 6
+
+/*
+ * cleancache requires every file with a page in cleancache to have a
+ * unique key unless/until the file is removed/truncated.  For some
+ * filesystems, the inode number is unique, but for "modern" filesystems
+ * an exportable filehandle is required (see exportfs.h)
+ */
+struct cleancache_filekey {
+	union {
+		ino_t ino;
+		__u32 fh[CLEANCACHE_KEY_MAX];
+		u32 key[CLEANCACHE_KEY_MAX];
+	} u;
+};
+
+struct cleancache_ops {
+	int (*init_fs)(size_t);
+	int (*init_shared_fs)(char *uuid, size_t);
+	int (*get_page)(int, struct cleancache_filekey,
+			pgoff_t, struct page *);
+	void (*put_page)(int, struct cleancache_filekey,
+			pgoff_t, struct page *);
+	void (*flush_page)(int, struct cleancache_filekey, pgoff_t);
+	void (*flush_inode)(int, struct cleancache_filekey);
+	void (*flush_fs)(int);
+};
+
+extern struct cleancache_ops
+	cleancache_register_ops(struct cleancache_ops *ops);
+extern void __cleancache_init_fs(struct super_block *);
+extern void __cleancache_init_shared_fs(char *, struct super_block *);
+extern int  __cleancache_get_page(struct page *);
+extern void __cleancache_put_page(struct page *);
+extern void __cleancache_flush_page(struct address_space *, struct page *);
+extern void __cleancache_flush_inode(struct address_space *);
+extern void __cleancache_flush_fs(struct super_block *);
+extern int cleancache_enabled;
+
+#ifdef CONFIG_CLEANCACHE
+static inline bool cleancache_fs_enabled(struct page *page)
+{
+	return page->mapping->host->i_sb->cleancache_poolid >= 0;
+}
+static inline bool cleancache_fs_enabled_mapping(struct address_space *mapping)
+{
+	return mapping->host->i_sb->cleancache_poolid >= 0;
+}
+#else
+#define cleancache_enabled (0)
+#define cleancache_fs_enabled(_page) (0)
+#define cleancache_fs_enabled_mapping(_page) (0)
+#endif
+
+/*
+ * The shim layer provided by these inline functions allows the compiler
+ * to reduce all cleancache hooks to nothingness if CONFIG_CLEANCACHE
+ * is disabled, to a single global variable check if CONFIG_CLEANCACHE
+ * is enabled but no cleancache "backend" has dynamically enabled it,
+ * and, for the most frequent cleancache ops, to a single global variable
+ * check plus a superblock element comparison if CONFIG_CLEANCACHE is enabled
+ * and a cleancache backend has dynamically enabled cleancache, but the
+ * filesystem referenced by that cleancache op has not enabled cleancache.
+ * As a result, CONFIG_CLEANCACHE can be enabled by default with essentially
+ * no measurable performance impact.
+ */
+
+static inline void cleancache_init_fs(struct super_block *sb)
+{
+	if (cleancache_enabled)
+		__cleancache_init_fs(sb);
+}
+
+static inline void cleancache_init_shared_fs(char *uuid, struct super_block *sb)
+{
+	if (cleancache_enabled)
+		__cleancache_init_shared_fs(uuid, sb);
+}
+
+static inline int cleancache_get_page(struct page *page)
+{
+	int ret = -1;
+
+	if (cleancache_enabled && cleancache_fs_enabled(page))
+		ret = __cleancache_get_page(page);
+	return ret;
+}
+
+static inline void cleancache_put_page(struct page *page)
+{
+	if (cleancache_enabled && cleancache_fs_enabled(page))
+		__cleancache_put_page(page);
+}
+
+static inline void cleancache_flush_page(struct address_space *mapping,
+					struct page *page)
+{
+	/* careful... page->mapping is NULL sometimes when this is called */
+	if (cleancache_enabled && cleancache_fs_enabled_mapping(mapping))
+		__cleancache_flush_page(mapping, page);
+}
+
+static inline void cleancache_flush_inode(struct address_space *mapping)
+{
+	if (cleancache_enabled && cleancache_fs_enabled_mapping(mapping))
+		__cleancache_flush_inode(mapping);
+}
+
+static inline void cleancache_flush_fs(struct super_block *sb)
+{
+	if (cleancache_enabled)
+		__cleancache_flush_fs(sb);
+}
+
+#endif /* _LINUX_CLEANCACHE_H */

include/linux/fs.h

@@ -1428,6 +1428,11 @@ struct super_block {
 	 */
 	char __rcu *s_options;
 	const struct dentry_operations *s_d_op; /* default d_op for dentries */
+
+	/*
+	 * Saved pool identifier for cleancache (-1 means none)
+	 */
+	int cleancache_poolid;
 };
 
 extern struct timespec current_fs_time(struct super_block *sb);

include/xen/interface/xen.h

@@ -58,6 +58,7 @@
 #define __HYPERVISOR_event_channel_op     32
 #define __HYPERVISOR_physdev_op           33
 #define __HYPERVISOR_hvm_op               34
+#define __HYPERVISOR_tmem_op              38
 
 /* Architecture-specific hypercall definitions. */
 #define __HYPERVISOR_arch_0               48
@@ -461,6 +462,27 @@ typedef uint8_t xen_domain_handle_t[16];
 #define __mk_unsigned_long(x) x ## UL
 #define mk_unsigned_long(x) __mk_unsigned_long(x)
 
+#define TMEM_SPEC_VERSION 1
+
+struct tmem_op {
+	uint32_t cmd;
+	int32_t pool_id;
+	union {
+		struct {  /* for cmd == TMEM_NEW_POOL */
+			uint64_t uuid[2];
+			uint32_t flags;
+		} new;
+		struct {
+			uint64_t oid[3];
+			uint32_t index;
+			uint32_t tmem_offset;
+			uint32_t pfn_offset;
+			uint32_t len;
+			GUEST_HANDLE(void) gmfn; /* guest machine page frame */
+		} gen;
+	} u;
+};
+
 #else /* __ASSEMBLY__ */
 
 /* In assembly code we cannot use C numeric constant suffixes. */

mm/Kconfig

@@ -347,3 +347,26 @@ config NEED_PER_CPU_KM
 	depends on !SMP
 	bool
 	default y
+
+config CLEANCACHE
+	bool "Enable cleancache driver to cache clean pages if tmem is present"
+	default n
+	help
+	  Cleancache can be thought of as a page-granularity victim cache
+	  for clean pages that the kernel's pageframe replacement algorithm
+	  (PFRA) would like to keep around, but can't since there isn't enough
+	  memory.  So when the PFRA "evicts" a page, it first attempts to use
+	  cleancacne code to put the data contained in that page into
+	  "transcendent memory", memory that is not directly accessible or
+	  addressable by the kernel and is of unknown and possibly
+	  time-varying size.  And when a cleancache-enabled
+	  filesystem wishes to access a page in a file on disk, it first
+	  checks cleancache to see if it already contains it; if it does,
+	  the page is copied into the kernel and a disk access is avoided.
+	  When a transcendent memory driver is available (such as zcache or
+	  Xen transcendent memory), a significant I/O reduction
+	  may be achieved.  When none is available, all cleancache calls
+	  are reduced to a single pointer-compare-against-NULL resulting
+	  in a negligible performance hit.
+
+	  If unsure, say Y to enable cleancache

mm/Makefile

@@ -49,3 +49,4 @@ obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o
 obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o
 obj-$(CONFIG_DEBUG_KMEMLEAK) += kmemleak.o
 obj-$(CONFIG_DEBUG_KMEMLEAK_TEST) += kmemleak-test.o
+obj-$(CONFIG_CLEANCACHE) += cleancache.o

mm/cleancache.c

@@ -0,0 +1,244 @@
+/*
+ * Cleancache frontend
+ *
+ * This code provides the generic "frontend" layer to call a matching
+ * "backend" driver implementation of cleancache.  See
+ * Documentation/vm/cleancache.txt for more information.
+ *
+ * Copyright (C) 2009-2010 Oracle Corp. All rights reserved.
+ * Author: Dan Magenheimer
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/exportfs.h>
+#include <linux/mm.h>
+#include <linux/cleancache.h>
+
+/*
+ * This global enablement flag may be read thousands of times per second
+ * by cleancache_get/put/flush even on systems where cleancache_ops
+ * is not claimed (e.g. cleancache is config'ed on but remains
+ * disabled), so is preferred to the slower alternative: a function
+ * call that checks a non-global.
+ */
+int cleancache_enabled;
+EXPORT_SYMBOL(cleancache_enabled);
+
+/*
+ * cleancache_ops is set by cleancache_ops_register to contain the pointers
+ * to the cleancache "backend" implementation functions.
+ */
+static struct cleancache_ops cleancache_ops;
+
+/* useful stats available in /sys/kernel/mm/cleancache */
+static unsigned long cleancache_succ_gets;
+static unsigned long cleancache_failed_gets;
+static unsigned long cleancache_puts;
+static unsigned long cleancache_flushes;
+
+/*
+ * register operations for cleancache, returning previous thus allowing
+ * detection of multiple backends and possible nesting
+ */
+struct cleancache_ops cleancache_register_ops(struct cleancache_ops *ops)
+{
+	struct cleancache_ops old = cleancache_ops;
+
+	cleancache_ops = *ops;
+	cleancache_enabled = 1;
+	return old;
+}
+EXPORT_SYMBOL(cleancache_register_ops);
+
+/* Called by a cleancache-enabled filesystem at time of mount */
+void __cleancache_init_fs(struct super_block *sb)
+{
+	sb->cleancache_poolid = (*cleancache_ops.init_fs)(PAGE_SIZE);
+}
+EXPORT_SYMBOL(__cleancache_init_fs);
+
+/* Called by a cleancache-enabled clustered filesystem at time of mount */
+void __cleancache_init_shared_fs(char *uuid, struct super_block *sb)
+{
+	sb->cleancache_poolid =
+		(*cleancache_ops.init_shared_fs)(uuid, PAGE_SIZE);
+}
+EXPORT_SYMBOL(__cleancache_init_shared_fs);
+
+/*
+ * If the filesystem uses exportable filehandles, use the filehandle as
+ * the key, else use the inode number.
+ */
+static int cleancache_get_key(struct inode *inode,
+			      struct cleancache_filekey *key)
+{
+	int (*fhfn)(struct dentry *, __u32 *fh, int *, int);
+	int len = 0, maxlen = CLEANCACHE_KEY_MAX;
+	struct super_block *sb = inode->i_sb;
+
+	key->u.ino = inode->i_ino;
+	if (sb->s_export_op != NULL) {
+		fhfn = sb->s_export_op->encode_fh;
+		if  (fhfn) {
+			struct dentry d;
+			d.d_inode = inode;
+			len = (*fhfn)(&d, &key->u.fh[0], &maxlen, 0);
+			if (len <= 0 || len == 255)
+				return -1;
+			if (maxlen > CLEANCACHE_KEY_MAX)
+				return -1;
+		}
+	}
+	return 0;
+}
+
+/*
+ * "Get" data from cleancache associated with the poolid/inode/index
+ * that were specified when the data was put to cleanache and, if
+ * successful, use it to fill the specified page with data and return 0.
+ * The pageframe is unchanged and returns -1 if the get fails.
+ * Page must be locked by caller.
+ */
+int __cleancache_get_page(struct page *page)
+{
+	int ret = -1;
+	int pool_id;
+	struct cleancache_filekey key = { .u.key = { 0 } };
+
+	VM_BUG_ON(!PageLocked(page));
+	pool_id = page->mapping->host->i_sb->cleancache_poolid;
+	if (pool_id < 0)
+		goto out;
+
+	if (cleancache_get_key(page->mapping->host, &key) < 0)
+		goto out;
+
+	ret = (*cleancache_ops.get_page)(pool_id, key, page->index, page);
+	if (ret == 0)
+		cleancache_succ_gets++;
+	else
+		cleancache_failed_gets++;
+out:
+	return ret;
+}
+EXPORT_SYMBOL(__cleancache_get_page);
+
+/*
+ * "Put" data from a page to cleancache and associate it with the
+ * (previously-obtained per-filesystem) poolid and the page's,
+ * inode and page index.  Page must be locked.  Note that a put_page
+ * always "succeeds", though a subsequent get_page may succeed or fail.
+ */
+void __cleancache_put_page(struct page *page)
+{
+	int pool_id;
+	struct cleancache_filekey key = { .u.key = { 0 } };
+
+	VM_BUG_ON(!PageLocked(page));
+	pool_id = page->mapping->host->i_sb->cleancache_poolid;
+	if (pool_id >= 0 &&
+	      cleancache_get_key(page->mapping->host, &key) >= 0) {
+		(*cleancache_ops.put_page)(pool_id, key, page->index, page);
+		cleancache_puts++;
+	}
+}
+EXPORT_SYMBOL(__cleancache_put_page);
+
+/*
+ * Flush any data from cleancache associated with the poolid and the
+ * page's inode and page index so that a subsequent "get" will fail.
+ */
+void __cleancache_flush_page(struct address_space *mapping, struct page *page)
+{
+	/* careful... page->mapping is NULL sometimes when this is called */
+	int pool_id = mapping->host->i_sb->cleancache_poolid;
+	struct cleancache_filekey key = { .u.key = { 0 } };
+
+	if (pool_id >= 0) {
+		VM_BUG_ON(!PageLocked(page));
+		if (cleancache_get_key(mapping->host, &key) >= 0) {
+			(*cleancache_ops.flush_page)(pool_id, key, page->index);
+			cleancache_flushes++;
+		}
+	}
+}
+EXPORT_SYMBOL(__cleancache_flush_page);
+
+/*
+ * Flush all data from cleancache associated with the poolid and the
+ * mappings's inode so that all subsequent gets to this poolid/inode
+ * will fail.
+ */
+void __cleancache_flush_inode(struct address_space *mapping)
+{
+	int pool_id = mapping->host->i_sb->cleancache_poolid;
+	struct cleancache_filekey key = { .u.key = { 0 } };
+
+	if (pool_id >= 0 && cleancache_get_key(mapping->host, &key) >= 0)
+		(*cleancache_ops.flush_inode)(pool_id, key);
+}
+EXPORT_SYMBOL(__cleancache_flush_inode);
+
+/*
+ * Called by any cleancache-enabled filesystem at time of unmount;
+ * note that pool_id is surrendered and may be reutrned by a subsequent
+ * cleancache_init_fs or cleancache_init_shared_fs
+ */
+void __cleancache_flush_fs(struct super_block *sb)
+{
+	if (sb->cleancache_poolid >= 0) {
+		int old_poolid = sb->cleancache_poolid;
+		sb->cleancache_poolid = -1;
+		(*cleancache_ops.flush_fs)(old_poolid);
+	}
+}
+EXPORT_SYMBOL(__cleancache_flush_fs);
+
+#ifdef CONFIG_SYSFS
+
+/* see Documentation/ABI/xxx/sysfs-kernel-mm-cleancache */
+
+#define CLEANCACHE_SYSFS_RO(_name) \
+	static ssize_t cleancache_##_name##_show(struct kobject *kobj, \
+				struct kobj_attribute *attr, char *buf) \
+	{ \
+		return sprintf(buf, "%lu\n", cleancache_##_name); \
+	} \
+	static struct kobj_attribute cleancache_##_name##_attr = { \
+		.attr = { .name = __stringify(_name), .mode = 0444 }, \
+		.show = cleancache_##_name##_show, \
+	}
+
+CLEANCACHE_SYSFS_RO(succ_gets);
+CLEANCACHE_SYSFS_RO(failed_gets);
+CLEANCACHE_SYSFS_RO(puts);
+CLEANCACHE_SYSFS_RO(flushes);
+
+static struct attribute *cleancache_attrs[] = {
+	&cleancache_succ_gets_attr.attr,
+	&cleancache_failed_gets_attr.attr,
+	&cleancache_puts_attr.attr,
+	&cleancache_flushes_attr.attr,
+	NULL,
+};
+
+static struct attribute_group cleancache_attr_group = {
+	.attrs = cleancache_attrs,
+	.name = "cleancache",
+};
+
+#endif /* CONFIG_SYSFS */
+
+static int __init init_cleancache(void)
+{
+#ifdef CONFIG_SYSFS
+	int err;
+
+	err = sysfs_create_group(mm_kobj, &cleancache_attr_group);
+#endif /* CONFIG_SYSFS */
+	return 0;
+}
+module_init(init_cleancache)

mm/filemap.c

@@ -34,6 +34,7 @@
 #include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */
 #include <linux/memcontrol.h>
 #include <linux/mm_inline.h> /* for page_is_file_cache() */
+#include <linux/cleancache.h>
 #include "internal.h"
 
 /*
@@ -118,6 +119,16 @@ void __delete_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
 
+	/*
+	 * if we're uptodate, flush out into the cleancache, otherwise
+	 * invalidate any existing cleancache entries.  We can't leave
+	 * stale data around in the cleancache once our page is gone
+	 */
+	if (PageUptodate(page) && PageMappedToDisk(page))
+		cleancache_put_page(page);
+	else
+		cleancache_flush_page(mapping, page);
+
 	radix_tree_delete(&mapping->page_tree, page->index);
 	page->mapping = NULL;
 	mapping->nrpages--;

mm/truncate.c

@@ -19,6 +19,7 @@
 #include <linux/task_io_accounting_ops.h>
 #include <linux/buffer_head.h>	/* grr. try_to_release_page,
 				   do_invalidatepage */
+#include <linux/cleancache.h>
 #include "internal.h"
 
 
@@ -51,6 +52,7 @@ void do_invalidatepage(struct page *page, unsigned long offset)
 static inline void truncate_partial_page(struct page *page, unsigned partial)
 {
 	zero_user_segment(page, partial, PAGE_CACHE_SIZE);
+	cleancache_flush_page(page->mapping, page);
 	if (page_has_private(page))
 		do_invalidatepage(page, partial);
 }
@@ -214,6 +216,7 @@ void truncate_inode_pages_range(struct address_space *mapping,
 	pgoff_t next;
 	int i;
 
+	cleancache_flush_inode(mapping);
 	if (mapping->nrpages == 0)
 		return;
 
@@ -291,6 +294,7 @@ void truncate_inode_pages_range(struct address_space *mapping,
 		pagevec_release(&pvec);
 		mem_cgroup_uncharge_end();
 	}
+	cleancache_flush_inode(mapping);
 }
 EXPORT_SYMBOL(truncate_inode_pages_range);
 
@@ -440,6 +444,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping,
 	int did_range_unmap = 0;
 	int wrapped = 0;
 
+	cleancache_flush_inode(mapping);
 	pagevec_init(&pvec, 0);
 	next = start;
 	while (next <= end && !wrapped &&
@@ -498,6 +503,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping,
 		mem_cgroup_uncharge_end();
 		cond_resched();
 	}
+	cleancache_flush_inode(mapping);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range);