ocfs2: Fix cleancache initialization call to correctly pass uuid
As reported by Steven Whitehouse in https://lkml.org/lkml/2011/5/27/221
the ocfs2 volume UUID is incorrectly passed to cleancache.
As a result, shared-ephemeral tmem pools will not actually
be created; instead they will be private (unshared) which
misses out on a major benefit of tmem.
Reported-by: Steven Whitehouse <[email protected]>
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Joel Becker <[email protected]>
The balloon driver's "current_pages" is very different from
totalram_pages. Self-ballooning needs to be driven by
the latter. Also, Committed_AS doesn't account for pages
used by the kernel so:
1) Add totalreserve_pages to Committed_AS for the normal target.
2) Enforce a floor for when there are little or no user-space threads
using memory (e.g. single-user mode) to avoid OOMs. The floor
function includes a "min_usable_mb" tuneable in case we discover
later that the floor function is still too aggressive in some
workloads, though likely it will not be needed.
Changes since version 4:
- change floor calculation so that it is not as aggressive; this version
uses a piecewise linear function similar to minimum_target in the 2.6.18
balloon driver, but modified to add to totalreserve_pages instead of
subtract from max_pfn, the 2.6.18 version causes OOMs on recent kernels
because the kernel has expanded over time
- change safety_margin to min_usable_mb and comment on its use
- since committed_as does NOT include kernel space (and other reserved
pages), totalreserve_pages is now added to committed_as. The result is
less aggressive self-ballooning, but theoretically more appropriate.
Changes since version 3:
- missing include causes compile problem when CONFIG_FRONTSWAP is disabled
- add comments after includes
Changes since version 2:
- missing include causes compile problem only on 32-bit
Changes since version 1:
- tuneable safety margin added
[v5: [email protected]: still too aggressive, seeing some OOMs]
[v4: [email protected]: fix compile when CONFIG_FRONTSWAP is disabled]
[v3: [email protected]: fix 32-bit compile]
[v2: [email protected]: make safety margin tuneable]
Signed-off-by: Dan Magenheimer <[email protected]>
[v1: Altered description and added an extra include]
Signed-off-by: Konrad Rzeszutek Wilk <[email protected]>
This is version 3 of an update to zcache, incorporating feedback from the list.
This patch adds support to the in-kernel transcendent memory ("tmem") code
and the zcache driver for multiple clients, which will be needed for both
RAMster and KVM support. It also adds additional tmem callbacks to support
RAMster and corresponding no-op stubs in the zcache driver. In v2, I've
also taken the liberty of adding some additional sysfs variables to
both surface information and allow policy control. Those experimenting
with zcache should find them useful. V3 clarifies some code walking
and declaring arrays.
Signed-off-by: Dan Magenheimer <[email protected]>
[v3: [email protected]: fix array bounds/walking]
[v2: [email protected]: fix bools, add check for NULL, fix a comment]
[v2: [email protected]: add info/tunables for poor compression]
[v2: [email protected]: add tunable for max persistent pages]
Acked-by: Dan Carpenter <[email protected]>
Cc: Nitin Gupta <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
This patch introduces two in-kernel drivers for Xen transcendent memory
("tmem") functionality that complement cleancache and frontswap. Both
use control theory to dynamically adjust and optimize memory utilization.
Selfballooning controls the in-kernel Xen balloon driver, targeting a goal
value (vm_committed_as), thus pushing less frequently used clean
page cache pages (through the cleancache code) into Xen tmem where
Xen can balance needs across all VMs residing on the physical machine.
Frontswap-selfshrinking controls the number of pages in frontswap,
driving it towards zero (effectively doing a partial swapoff) when
in-kernel memory pressure subsides, freeing up RAM for other VMs.
More detail is provided in the header comment of xen-selfballooning.c.
Signed-off-by: Dan Magenheimer <[email protected]>
[v8: [email protected]: set default enablement depending on frontswap]
[v7: [email protected]: fix capitalization and punctuation in comments]
[v6: fix frontswap-selfshrinking initialization]
[v6: [email protected]: fix init pr_infos; add comments about swap]
[v5: [email protected]: add NULL to attr list; move inits up to decls]
[v4: [email protected]: use strict_strtoul plus a few syntactic nits]
[v3: [email protected]: fix potential divides-by-zero]
[v3: [email protected]: add many more comments, fix nits]
[v2: rebased to linux-3.0-rc1]
[v2: [email protected]: reorganize as new file (xen-selfballoon.c)]
[v2: [email protected]: proper access to vm_committed_as]
[v2: [email protected]: accounting fixes]
Cc: Jan Beulich <[email protected]>
Cc: Jeremy Fitzhardinge <[email protected]>
Cc: <[email protected]>
Provide the shim code for frontswap for Xen tmem even if the
frontswap patchset is not present yet. (The egg is before
the chicken.)
Signed-off-by: Dan Magenheimer <[email protected]>
Reviewed-by: Konrad Wilk <[email protected]>
This eighth patch of eight in this cleancache series "opts-in"
cleancache for ocfs2. Clustered filesystems must explicitly enable
cleancache by calling cleancache_init_shared_fs anytime an instance
of the filesystem is mounted. Ocfs2 is currently the only user of
the clustered filesystem interface but nevertheless, the cleancache
hooks in the VFS layer are sufficient for ocfs2 including the matching
cleancache_flush_fs hook which must be called on unmount.
Details and a FAQ can be found in Documentation/vm/cleancache.txt
[v8: trivial merge conflict update]
[v5: [email protected]: simplify init hook and any future fs init changes]
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Joel Becker <[email protected]>
Reviewed-by: Jeremy Fitzhardinge <[email protected]>
Reviewed-by: Konrad Rzeszutek Wilk <[email protected]>
Cc: Mark Fasheh <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik Van Riel <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Chris Mason <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Ted Tso <[email protected]>
Cc: Nitin Gupta <[email protected]>
This sixth patch of eight in this cleancache series "opts-in"
cleancache for btrfs. Filesystems must explicitly enable
cleancache by calling cleancache_init_fs anytime an instance
of the filesystem is mounted. Btrfs uses its own readpage
which must be hooked, but all other cleancache hooks are in
the VFS layer including the matching cleancache_flush_fs hook
which must be called on unmount.
Details and a FAQ can be found in Documentation/vm/cleancache.txt
[v6-v8: no changes]
[v5: [email protected]: simplify init hook and any future fs init changes]
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Reviewed-by: Jeremy Fitzhardinge <[email protected]>
Reviewed-by: Konrad Rzeszutek Wilk <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik Van Riel <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Nitin Gupta <[email protected]>
This fourth patch of eight in this cleancache series provides the
core hooks in VFS for: initializing cleancache per filesystem;
capturing clean pages reclaimed by page cache; attempting to get
pages from cleancache before filesystem read; and ensuring coherency
between pagecache, disk, and cleancache. Note that the placement
of these hooks was stable from 2.6.18 to 2.6.38; a minor semantic
change was required due to a patchset in 2.6.39.
All hooks become no-ops if CONFIG_CLEANCACHE is unset, or become
a check of a boolean global if CONFIG_CLEANCACHE is set but no
cleancache "backend" has claimed cleancache_ops.
Details and a FAQ can be found in Documentation/vm/cleancache.txt
[v8: [email protected]: adapt to new remove_from_page_cache function]
Signed-off-by: Chris Mason <[email protected]>
Signed-off-by: Dan Magenheimer <[email protected]>
Reviewed-by: Jeremy Fitzhardinge <[email protected]>
Reviewed-by: Konrad Rzeszutek Wilk <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik Van Riel <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Nitin Gupta <[email protected]>
This third patch of eight in this cleancache series provides
the core code for cleancache that interfaces between the hooks in
VFS and individual filesystems and a cleancache backend. It also
includes build and config patches.
Two new files are added: mm/cleancache.c and include/linux/cleancache.h.
Note that CONFIG_CLEANCACHE can default to on; in systems that do
not provide a cleancache backend, all hooks devolve to a simple
check of a global enable flag, so performance impact should
be negligible but can be reduced to zero impact if config'ed off.
However for this first commit, it defaults to off.
Details and a FAQ can be found in Documentation/vm/cleancache.txt
Credits: Cleancache_ops design derived from Jeremy Fitzhardinge
design for tmem
[v8: [email protected]: fix exportfs call affecting btrfs]
[v8: [email protected]: use static inline function, not macro]
[v7: [email protected]: cleanup sysfs and remove cleancache prefix]
[v6: [email protected]: robustly handle buggy fs encode_fh actor definition]
[v5: [email protected]: clean up global usage and static var names]
[v5: [email protected]: simplify init hook and any future fs init changes]
[v5: [email protected]: cleaner non-global interface for ops registration]
[v4: [email protected]: interface must support exportfs FS's]
[v4: [email protected]: interface must support 64-bit FS on 32-bit kernel]
[v3: [email protected]: use one ops struct to avoid pointer hops]
[v3: [email protected]: document and ensure PageLocked reqts are met]
[v3: [email protected]: fix success/fail codes, change funcs to void]
[v2: [email protected]: use sane types]
Signed-off-by: Dan Magenheimer <[email protected]>
Reviewed-by: Jeremy Fitzhardinge <[email protected]>
Reviewed-by: Konrad Rzeszutek Wilk <[email protected]>
Acked-by: Al Viro <[email protected]>
Acked-by: Andrew Morton <[email protected]>
Acked-by: Nitin Gupta <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Acked-by: Andreas Dilger <[email protected]>
Acked-by: Jan Beulich <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik Van Riel <[email protected]>
Cc: Chris Mason <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
This patchset introduces cleancache, an optional new feature exposed
by the VFS layer that potentially dramatically increases page cache
effectiveness for many workloads in many environments at a negligible
cost. It does this by providing an interface to transcendent memory,
which is memory/storage that is not otherwise visible to and/or directly
addressable by the kernel.
Instead of being discarded, hooks in the reclaim code "put" clean
pages to cleancache. Filesystems that "opt-in" may "get" pages
from cleancache that were previously put, but pages in cleancache are
"ephemeral", meaning they may disappear at any time. And the size
of cleancache is entirely dynamic and unknowable to the kernel.
Filesystems currently supported by this patchset include ext3, ext4,
btrfs, and ocfs2. Other filesystems (especially those built entirely
on VFS) should be easy to add, but should first be thoroughly tested to
ensure coherency.
Details and a FAQ are provided in Documentation/vm/cleancache.txt
This first patch of eight in this cleancache series only adds two
new documentation files.
[v8: minor documentation changes by author]
[v3: [email protected]: document sysfs API]
[v3: [email protected]: move detailed description to Documentation/vm]
Signed-off-by: Dan Magenheimer <[email protected]>
Reviewed-by: Jeremy Fitzhardinge <[email protected]>
Reviewed-by: Konrad Rzeszutek Wilk <[email protected]>
Acked-by: Andrew Morton <[email protected]>
Acked-by: Randy Dunlap <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Matthew Wilcox <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Rik Van Riel <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Chris Mason <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Ted Ts'o <[email protected]>
Cc: Mark Fasheh <[email protected]>
Cc: Joel Becker <[email protected]>
Cc: Nitin Gupta <[email protected]>
[PATCH V2 3/3] drivers/staging: zcache: misc build/config
Makefiles and Kconfigs to build zcache in drivers/staging
There is a dependency on xvmalloc.* which in 2.6.37 resides
in drivers/staging/zram. Should this move or disappear,
some Makefile/Kconfig changes will be required.
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Nitin Gupta <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
[PATCH V2 2/3] drivers/staging: zcache: host services and PAM services
Zcache provides host services (memory allocation) for tmem,
a "shim" to interface cleancache and frontswap to tmem, and
two different page-addressable memory implemenations using
lzo1x compression. The first, "compression buddies" ("zbud")
compresses pairs of pages and supplies a shrinker interface
that allows entire pages to be reclaimed. The second is
a shim to xvMalloc which is more space-efficient but
less receptive to page reclamation. The first is used
for ephemeral pools and the second for persistent pools.
All ephemeral pools share the same memory, that is, even
pages from different pools can share the same page.
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Nitin Gupta <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
[PATCH V2 1/3] drivers/staging: zcache: in-kernel tmem code
Transcendent memory ("tmem") is a clean API/ABI that provides
for an efficient address translation and a set of highly
concurrent access methods to copy data between a page-oriented
data source (e.g. cleancache or frontswap) and a page-addressable
memory ("PAM") data store. Of critical importance, the PAM data
store is of unknown (and possibly varying) size so any individual
access may succeed or fail as defined by the API/ABI.
Tmem exports a basic set of access methods (e.g. put, get,
flush, flush object, new pool, and destroy pool) which are
normally called from a "host" (e.g. zcache).
To be functional, two sets of "ops" must be registered by the
host, one to provide "host services" (memory allocation) and
one to provide page-addressable memory ("PAM") hooks.
Tmem supports one or more "clients", each which can provide
a set of "pools" to partition pages. Each pool contains
a set of "objects"; each object holds pointers to some number
of PAM page descriptors ("pampd"), indexed by an "index" number.
This triple <pool id, object id, index> is sometimes referred
to as a "handle". Tmem's primary function is to essentially
provide address translation of handles into pampds and move
data appropriately.
As an example, for cleancache, a pool maps to a filesystem,
an object maps to a file, and the index is the page offset
into the file. And in this patch, zcache is the host and
each PAM descriptor points to a compressed page of data.
Tmem supports two kinds of pages: "ephemeral" and "persistent".
Ephemeral pages may be asynchronously reclaimed "bottoms up"
so the data structures and concurrency model must allow for
this. For example, each pampd must retain sufficient information
to invalidate tmem's handle-to-pampd translation.
its containing object so that, on reclaim, all tmem data
structures can be made consistent.
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Fix this:
> It appears that in the upstream balloon driver,
> the call to HYPERVISOR_update_va_mapping is missing
> from decrease_reservation. I think as a result,
> the balloon driver is eating memory but not
> releasing it to Xen, thus rendering the balloon
> driver essentially useless. (Can be observed via xentop.)
Signed-off-by: Dan Magenheimer <[email protected]>
Signed-off-by: Jeremy Fitzhardinge <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Dan Magenheimer