SeongJae Park [Tue, 22 Mar 2022 21:49:52 +0000 (14:49 -0700)]
selftests/damon: add a test for DAMON sysfs interface
This commit adds a selftest for DAMON sysfs interface. It tests the
functionality of 'nr' files and existence of files in each directory of
the hierarchy.
Link: https://lkml.kernel.org/r/20220228081314.5770-12-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Shuah Khan <skhan@linuxfoundation.org> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:49 +0000 (14:49 -0700)]
mm/damon/sysfs: support DAMOS stats
This commit makes DAMON sysfs interface supports the DAMOS stats feature.
Specifically, this commit adds 'stats' directory under each scheme
directory, and update the contents of the files under the directory
according to the latest monitoring results, when the user writes special
keyword, 'update_schemes_stats' to the 'state' file of the kdamond.
As a result, the files hierarchy becomes as below:
SeongJae Park [Tue, 22 Mar 2022 21:49:46 +0000 (14:49 -0700)]
mm/damon/sysfs: support DAMOS watermarks
This commit makes DAMON sysfs interface supports the DAMOS watermarks
feature. Specifically, this commit adds 'watermarks' directory under each
scheme directory and makes kdamond 'state' file writing respects the
contents in the directory.
As a result, the files hierarchy becomes as below:
SeongJae Park [Tue, 22 Mar 2022 21:49:43 +0000 (14:49 -0700)]
mm/damon/sysfs: support schemes prioritization
This commit makes DAMON sysfs interface supports the DAMOS' regions
prioritization weights feature under quotas limitation. Specifically,
this commit adds 'weights' directory under each scheme directory and makes
kdamond 'state' file writing respects the contents in the directory.
SeongJae Park [Tue, 22 Mar 2022 21:49:40 +0000 (14:49 -0700)]
mm/damon/sysfs: support DAMOS quotas
This commit makes DAMON sysfs interface supports the DAMOS quotas feature.
Specifically, this commit adds 'quotas' directory under each scheme
directory and makes kdamond 'state' file writing respects the contents in
the directory.
As a result, the files hierarchy becomes as below:
SeongJae Park [Tue, 22 Mar 2022 21:49:37 +0000 (14:49 -0700)]
mm/damon/sysfs: support DAMON-based Operation Schemes
This commit makes DAMON sysfs interface supports the DAMON-based operation
schemes (DAMOS) feature. Specifically, this commit adds 'schemes'
directory under each context direcotry, and makes kdamond 'state' file
writing respects the contents in the directory.
Note that this commit doesn't support all features of DAMOS but only the
target access pattern and action feature. Supports for quotas,
prioritization, watermarks will follow.
As a result, the files hierarchy becomes as below:
SeongJae Park [Tue, 22 Mar 2022 21:49:34 +0000 (14:49 -0700)]
mm/damon/sysfs: support the physical address space monitoring
This commit makes DAMON sysfs interface supports the physical address
space monitoring. Specifically, this commit adds support of the initial
monitoring regions set feature by adding 'regions' directory under each
target directory and makes context operations file to receive 'paddr' in
addition to 'vaddr'.
As a result, the files hierarchy becomes as below:
SeongJae Park [Tue, 22 Mar 2022 21:49:30 +0000 (14:49 -0700)]
mm/damon/sysfs: link DAMON for virtual address spaces monitoring
This commit links the DAMON sysfs interface to DAMON so that users can
control DAMON via the interface. In detail, this commit makes writing
'on' to 'state' file constructs DAMON contexts based on values that users
have written to relevant sysfs files and start the context. It supports
only virtual address spaces monitoring at the moment, though.
The files hierarchy of DAMON sysfs interface after this commit is shown
below. In the below figure, parents-children relations are represented
with indentations, each directory is having ``/`` suffix, and files in
each directory are separated by comma (",").
The usage is straightforward. Writing a number ('N') to each 'nr_*' file
makes directories named '0' to 'N-1'. Users can construct DAMON contexts
by writing proper values to the files in the straightforward manner and
start each kdamond by writing 'on' to 'kdamonds/<N>/state'.
Link: https://lkml.kernel.org/r/20220228081314.5770-5-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Shuah Khan <skhan@linuxfoundation.org> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:27 +0000 (14:49 -0700)]
mm/damon: implement a minimal stub for sysfs-based DAMON interface
DAMON's debugfs-based user interface served very well, so far. However,
it unnecessarily depends on debugfs, while DAMON is not aimed to be used
for only debugging. Also, the interface receives multiple values via one
file. For example, schemes file receives 18 values separated by white
spaces. As a result, it is ineffient, hard to be used, and difficult to
be extended. Especially, keeping backward compatibility of user space
tools is getting only challenging. It would be better to implement
another reliable and flexible interface and deprecate the debugfs
interface in long term.
To this end, this commit implements a stub of a part of the new user
interface of DAMON using sysfs. Specifically, this commit implements the
sysfs control parts for virtual address space monitoring.
More specifically, the idea of the new interface is, using directory
hierarchies and making one file for one value. The hierarchy that this
commit is introducing is as below. In the below figure, parents-children
relations are represented with indentations, each directory is having
``/`` suffix, and files in each directory are separated by comma (",").
Writing a number <N> to each 'nr' file makes directories of name <0> to
<N-1> in the directory of the 'nr' file. That's all this commit does.
Writing proper values to relevant files will construct the DAMON contexts,
and writing a special keyword, 'on', to 'state' files for each kdamond
will ask DAMON to start the constructed contexts.
For a short example, using below commands for monitoring virtual address
spaces of a given workload is imaginable:
Please note that this commit is implementing only the sysfs part stub as
abovely mentioned. This commit doesn't implement the special keywords for
'state' files. Following commits will do that.
[jiapeng.chong@linux.alibaba.com: fix missing error code in damon_sysfs_attrs_add_dirs()] Link: https://lkml.kernel.org/r/20220302111120.24984-1-jiapeng.chong@linux.alibaba.com Link: https://lkml.kernel.org/r/20220228081314.5770-4-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Cc: David Rientjes <rientjes@google.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Shuah Khan <skhan@linuxfoundation.org> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "Introduce DAMON sysfs interface", v3.
Introduction
============
DAMON's debugfs-based user interface (DAMON_DBGFS) served very well, so
far. However, it unnecessarily depends on debugfs, while DAMON is not
aimed to be used for only debugging. Also, the interface receives
multiple values via one file. For example, schemes file receives 18
values. As a result, it is inefficient, hard to be used, and difficult to
be extended. Especially, keeping backward compatibility of user space
tools is getting only challenging. It would be better to implement
another reliable and flexible interface and deprecate DAMON_DBGFS in long
term.
For the reason, this patchset introduces a sysfs-based new user interface
of DAMON. The idea of the new interface is, using directory hierarchies
and having one dedicated file for each value. For a short example, users
can do the virtual address monitoring via the interface as below:
A brief representation of the files hierarchy of DAMON sysfs interface is
as below. Childs are represented with indentation, directories are having
'/' suffix, and files in each directory are separated by comma.
Detailed usage of the files will be described in the final Documentation
patch of this patchset.
Main Difference Between DAMON_DBGFS and DAMON_SYSFS
---------------------------------------------------
At the moment, DAMON_DBGFS and DAMON_SYSFS provides same features. One
important difference between them is their exclusiveness. DAMON_DBGFS
works in an exclusive manner, so that no DAMON worker thread (kdamond) in
the system can run concurrently and interfere somehow. For the reason,
DAMON_DBGFS asks users to construct all monitoring contexts and start them
at once. It's not a big problem but makes the operation a little bit
complex and unflexible.
For more flexible usage, DAMON_SYSFS moves the responsibility of
preventing any possible interference to the admins and work in a
non-exclusive manner. That is, users can configure and start contexts one
by one. Note that DAMON respects both exclusive groups and non-exclusive
groups of contexts, in a manner similar to that of reader-writer locks.
That is, if any exclusive monitoring contexts (e.g., contexts that started
via DAMON_DBGFS) are running, DAMON_SYSFS does not start new contexts, and
vice versa.
Future Plan of DAMON_DBGFS Deprecation
======================================
Once this patchset is merged, DAMON_DBGFS development will be frozen.
That is, we will maintain it to work as is now so that no users will be
break. But, it will not be extended to provide any new feature of DAMON.
The support will be continued only until next LTS release. After that, we
will drop DAMON_DBGFS.
As DAMON_SYSFS provides all features of DAMON_DBGFS, all user space
tooling can move to DAMON_SYSFS. As we will continue supporting
DAMON_DBGFS until next LTS kernel release, user space tools would have
enough time to move to DAMON_SYSFS.
The official user space tool, damo[1], is already supporting both
DAMON_SYSFS and DAMON_DBGFS. Both correctness tests[2] and performance
tests[3] of DAMON using DAMON_SYSFS also passed.
First two patches (patches 1-2) make core changes for DAMON_SYSFS. The
first one (patch 1) allows non-exclusive DAMON contexts so that
DAMON_SYSFS can work in non-exclusive mode, while the second one (patch 2)
adds size of DAMON enum types so that DAMON API users can safely iterate
the enums.
Third patch (patch 3) implements basic sysfs stub for virtual address
spaces monitoring. Note that this implements only sysfs files and DAMON
is not linked. Fourth patch (patch 4) links the DAMON_SYSFS to DAMON so
that users can control DAMON using the sysfs files.
Following six patches (patches 5-10) implements other DAMON features that
DAMON_DBGFS supports one by one (physical address space monitoring,
DAMON-based operation schemes, schemes quotas, schemes prioritization
weights, schemes watermarks, and schemes stats).
Following patch (patch 11) adds a simple selftest for DAMON_SYSFS, and the
final one (patch 12) documents DAMON_SYSFS.
This patch (of 13):
To avoid interference between DAMON contexts monitoring overlapping memory
regions, damon_start() works in an exclusive manner. That is,
damon_start() does nothing bug fails if any context that started by
another instance of the function is still running. This makes its usage a
little bit restrictive. However, admins could aware each DAMON usage and
address such interferences on their own in some cases.
This commit hence implements non-exclusive mode of the function and allows
the callers to select the mode. Note that the exclusive groups and
non-exclusive groups of contexts will respect each other in a manner
similar to that of reader-writer locks. Therefore, this commit will not
cause any behavioral change to the exclusive groups.
Link: https://lkml.kernel.org/r/20220228081314.5770-1-sj@kernel.org Link: https://lkml.kernel.org/r/20220228081314.5770-2-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Shuah Khan <skhan@linuxfoundation.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:18 +0000 (14:49 -0700)]
Docs/damon: update outdated term 'regions update interval'
Before DAMON is merged in the mainline, the concept of 'regions update
interval' has generalized to be used as the time interval for update of
any monitoring operations related data structure, but the document has not
updated properly. This commit updates the document for better
consistency.
Link: https://lkml.kernel.org/r/20220222170100.17068-4-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In DAMON's early development stage before it be merged in the mainline, it
was first designed to work exclusively with Idle page tracking to avoid
any interference between each other. Later, but still before be merged in
the mainline, because Idle page tracking is fully under the control of
sysadmins, we made the resolving of conflict as the responsibility of
sysadmins. The document is not updated for the change, though. This
commit updates the document for that.
Link: https://lkml.kernel.org/r/20220222170100.17068-3-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:12 +0000 (14:49 -0700)]
Docs/vm/damon: call low level monitoring primitives the operations
Patch series "Docs/damon: Update documents for better consistency".
Some of DAMON document are not properly updated for latest version. This
patchset updates such parts.
This patch (of 3):
DAMON code calls the low level monitoring primitives implementations the
monitoring operations. The documentation would have no problem at still
calling those primitives implementation because there is no real
difference in the concepts, but making it more consistent with the code
would make it better. This commit therefore convert sentences in the doc
specifically pointing the implementations of the primitives to call it
monitoring operations.
Because DAMON debugfs interface and DAMON-based proactive reclaim are now
using monitoring operations via registration mechanism,
damon_{p,v}a_{target_valid,set_operations}() functions have no user. This
commit clean them up.
Link: https://lkml.kernel.org/r/20220215184603.1479-9-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:04 +0000 (14:49 -0700)]
mm/damon/dbgfs-test: fix is_target_id() change
DAMON kunit tests for DAMON debugfs interface fails because it still
assumes setting empty monitoring operations makes DAMON debugfs interface
believe the target of the context don't have pid. This commit fixes the
kunit test fails by explicitly setting the context's monitoring operations
with the operations for the physical address space, which let debugfs
knows the target will not have pid.
Link: https://lkml.kernel.org/r/20220215184603.1479-8-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:49:01 +0000 (14:49 -0700)]
mm/damon/dbgfs: use operations id for knowing if the target has pid
DAMON debugfs interface depends on monitoring operations for virtual
address spaces because it knows if the target has pid or not by seeing if
the context is configured to use one of the virtual address space
monitoring operation functions. We can replace that check with 'enum
damon_ops_id' now, to make it independent. This commit makes the change.
Link: https://lkml.kernel.org/r/20220215184603.1479-7-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:48:58 +0000 (14:48 -0700)]
mm/damon/dbgfs: use damon_select_ops() instead of damon_{v,p}a_set_operations()
This commit makes DAMON debugfs interface to select the registered
monitoring operations for the physical address space or virtual address
spaces depending on user requests instead of setting it on its own. Note
that DAMON debugfs interface is still dependent to DAMON_VADDR with this
change, because it is also using its symbol, 'damon_va_target_valid'.
Link: https://lkml.kernel.org/r/20220215184603.1479-6-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:48:55 +0000 (14:48 -0700)]
mm/damon/reclaim: use damon_select_ops() instead of damon_{v,p}a_set_operations()
This commit makes DAMON_RECLAIM to select the registered monitoring
operations for the physical address space instead of setting it on its
own. This allows DAMON_RECLAIM be independent of DAMON_PADDR, but leave
the dependency as is, because it's the only one monitoring operations it
use, and therefore it makes no sense to build DAMON_RECLAIM without
DAMON_PADDR.
Link: https://lkml.kernel.org/r/20220215184603.1479-5-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:48:52 +0000 (14:48 -0700)]
mm/damon/paddr,vaddr: register themselves to DAMON in subsys_initcall
This commit makes the monitoring operations for the physical address space
and virtual address spaces register themselves to DAMON in the
subsys_initcall step. Later, in-kernel DAMON user code can use them via
damon_select_ops() without have to unnecessarily depend on all possible
monitoring operations implementations.
Link: https://lkml.kernel.org/r/20220215184603.1479-4-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:48:49 +0000 (14:48 -0700)]
mm/damon: let monitoring operations can be registered and selected
In-kernel DAMON user code like DAMON debugfs interface should set 'struct
damon_operations' of its 'struct damon_ctx' on its own. Therefore, the
client code should depend on all supporting monitoring operations
implementations that it could use. For example, DAMON debugfs interface
depends on both vaddr and paddr, while some of the users are not always
interested in both.
To minimize such unnecessary dependencies, this commit makes the
monitoring operations can be registered by implementing code and then
dynamically selected by the user code without build-time dependency.
Link: https://lkml.kernel.org/r/20220215184603.1479-3-sj@kernel.org Signed-off-by: SeongJae Park <sj@kernel.org> Cc: David Rientjes <rientjes@google.com> Cc: Xin Hao <xhao@linux.alibaba.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SeongJae Park [Tue, 22 Mar 2022 21:48:46 +0000 (14:48 -0700)]
mm/damon: rename damon_primitives to damon_operations
Patch series "Allow DAMON user code independent of monitoring primitives".
In-kernel DAMON user code is required to configure the monitoring context
(struct damon_ctx) with proper monitoring primitives (struct
damon_primitive). This makes the user code dependent to all supporting
monitoring primitives. For example, DAMON debugfs interface depends on
both DAMON_VADDR and DAMON_PADDR, though some users have interest in only
one use case. As more monitoring primitives are introduced, the problem
will be bigger.
To minimize such unnecessary dependency, this patchset makes monitoring
primitives can be registered by the implemnting code and later dynamically
searched and selected by the user code.
In addition to that, this patchset renames monitoring primitives to
monitoring operations, which is more easy to intuitively understand what
it means and how it would be structed.
This patch (of 8):
DAMON has a set of callback functions called monitoring primitives and let
it can be configured with various implementations for easy extension for
different address spaces and usages. However, the word 'primitive' is not
so explicit. Meanwhile, many other structs resembles similar purpose
calls themselves 'operations'. To make the code easier to be understood,
this commit renames 'damon_primitives' to 'damon_operations' before it is
too late to rename.
Baolin Wang [Tue, 22 Mar 2022 21:48:43 +0000 (14:48 -0700)]
mm/damon: remove redundant page validation
It will never get a NULL page by pte_page() as discussed in thread [1],
thus remove the redundant page validation to fix below Smatch static
checker warning.
mm/damon/vaddr.c:405 damon_hugetlb_mkold()
warn: 'page' can't be NULL.
SeongJae Park [Tue, 22 Mar 2022 21:48:40 +0000 (14:48 -0700)]
mm/damon: remove the target id concept
DAMON asks each monitoring target ('struct damon_target') to have one
'unsigned long' integer called 'id', which should be unique among the
targets of same monitoring context. Meaning of it is, however, totally up
to the monitoring primitives that registered to the monitoring context.
For example, the virtual address spaces monitoring primitives treats the
id as a 'struct pid' pointer.
This makes the code flexible, but ugly, not well-documented, and
type-unsafe[1]. Also, identification of each target can be done via its
index. For the reason, this commit removes the concept and uses clear
type definition. For now, only 'struct pid' pointer is used for the
virtual address spaces monitoring. If DAMON is extended in future so that
we need to put another identifier field in the struct, we will use a union
for such primitives-dependent fields and document which primitives are
using which type.
SeongJae Park [Tue, 22 Mar 2022 21:48:37 +0000 (14:48 -0700)]
mm/damon/core: move damon_set_targets() into dbgfs
damon_set_targets() function is defined in the core for general use cases,
but called from only dbgfs. Also, because the function is for general use
cases, dbgfs does additional handling of pid type target id case. To make
the situation simpler, this commit moves the function into dbgfs and makes
it to do the pid type case handling on its own.
SeongJae Park [Tue, 22 Mar 2022 21:48:34 +0000 (14:48 -0700)]
Docs/admin-guide/mm/damon/usage: update for changed initail_regions file input
A previous commit made init_regions debugfs file to use target index
instead of target id for specifying the target of the init regions. This
commit updates the usage document to reflect the change.
SeongJae Park [Tue, 22 Mar 2022 21:48:31 +0000 (14:48 -0700)]
mm/damon/dbgfs/init_regions: use target index instead of target id
Patch series "Remove the type-unclear target id concept".
DAMON asks each monitoring target ('struct damon_target') to have one
'unsigned long' integer called 'id', which should be unique among the
targets of same monitoring context. Meaning of it is, however, totally up
to the monitoring primitives that registered to the monitoring context.
For example, the virtual address spaces monitoring primitives treats the
id as a 'struct pid' pointer.
This makes the code flexible but ugly, not well-documented, and
type-unsafe[1]. Also, identification of each target can be done via its
index. For the reason, this patchset removes the concept and uses clear
type definition.
Target id is a 'unsigned long' data, which can be interpreted differently
by each monitoring primitives. For example, it means 'struct pid *' for
the virtual address spaces monitoring, while it means nothing but an
integer to be displayed to debugfs interface users for the physical
address space monitoring. It's flexible but makes code ugly and
type-unsafe[1].
To be prepared for eventual removal of the concept, this commit removes a
use case of the concept in 'init_regions' debugfs file handling. In
detail, this commit replaces use of the id with the index of each target
in the context's targets list.
Marco Elver [Tue, 22 Mar 2022 21:48:25 +0000 (14:48 -0700)]
kfence: allow use of a deferrable timer
Allow the use of a deferrable timer, which does not force CPU wake-ups
when the system is idle. A consequence is that the sample interval
becomes very unpredictable, to the point that it is not guaranteed that
the KFENCE KUnit test still passes.
Nevertheless, on power-constrained systems this may be preferable, so
let's give the user the option should they accept the above trade-off.
Link: https://lkml.kernel.org/r/20220308141415.3168078-1-elver@google.com Signed-off-by: Marco Elver <elver@google.com> Reviewed-by: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Peng Liu [Tue, 22 Mar 2022 21:48:22 +0000 (14:48 -0700)]
kfence: test: try to avoid test_gfpzero trigger rcu_stall
When CONFIG_KFENCE_NUM_OBJECTS is set to a big number, kfence
kunit-test-case test_gfpzero will eat up nearly all the CPU's resources
and rcu_stall is reported as the following log which is cut from a
physical server.
Peng Liu [Tue, 22 Mar 2022 21:48:19 +0000 (14:48 -0700)]
kunit: make kunit_test_timeout compatible with comment
In function kunit_test_timeout, it is declared "300 * MSEC_PER_SEC"
represent 5min. However, it is wrong when dealing with arm64 whose
default HZ = 250, or some other situations. Use msecs_to_jiffies to fix
this, and kunit_test_timeout will work as desired.
Link: https://lkml.kernel.org/r/20220309083753.1561921-3-liupeng256@huawei.com Fixes: 087b16032720 ("kunit: test: add support for test abort") Signed-off-by: Peng Liu <liupeng256@huawei.com> Reviewed-by: Marco Elver <elver@google.com> Reviewed-by: Daniel Latypov <dlatypov@google.com> Reviewed-by: Brendan Higgins <brendanhiggins@google.com> Tested-by: Brendan Higgins <brendanhiggins@google.com> Cc: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Wang Kefeng <wangkefeng.wang@huawei.com> Cc: David Gow <davidgow@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Peng Liu [Tue, 22 Mar 2022 21:48:16 +0000 (14:48 -0700)]
kunit: fix UAF when run kfence test case test_gfpzero
Patch series "kunit: fix a UAF bug and do some optimization", v2.
This series is to fix UAF (use after free) when running kfence test case
test_gfpzero, which is time costly. This UAF bug can be easily triggered
by setting CONFIG_KFENCE_NUM_OBJECTS = 65535. Furthermore, some
optimization for kunit tests has been done.
This patch (of 3):
Kunit will create a new thread to run an actual test case, and the main
process will wait for the completion of the actual test thread until
overtime. The variable "struct kunit test" has local property in function
kunit_try_catch_run, and will be used in the test case thread. Task
kunit_try_catch_run will free "struct kunit test" when kunit runs
overtime, but the actual test case is still run and an UAF bug will be
triggered.
The above problem has been both observed in a physical machine and qemu
platform when running kfence kunit tests. The problem can be triggered
when setting CONFIG_KFENCE_NUM_OBJECTS = 65535. Under this setting, the
test case test_gfpzero will cost hours and kunit will run to overtime.
The follows show the panic log.
To solve this problem, the test case thread should be stopped when the
kunit frame runs overtime. The stop signal will send in function
kunit_try_catch_run, and test_gfpzero will handle it.
Link: https://lkml.kernel.org/r/20220309083753.1561921-1-liupeng256@huawei.com Link: https://lkml.kernel.org/r/20220309083753.1561921-2-liupeng256@huawei.com Signed-off-by: Peng Liu <liupeng256@huawei.com> Reviewed-by: Marco Elver <elver@google.com> Reviewed-by: Brendan Higgins <brendanhiggins@google.com> Tested-by: Brendan Higgins <brendanhiggins@google.com> Cc: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Wang Kefeng <wangkefeng.wang@huawei.com> Cc: Daniel Latypov <dlatypov@google.com> Cc: David Gow <davidgow@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Tianchen Ding [Tue, 22 Mar 2022 21:48:13 +0000 (14:48 -0700)]
kfence: alloc kfence_pool after system startup
Allow enabling KFENCE after system startup by allocating its pool via the
page allocator. This provides the flexibility to enable KFENCE even if it
wasn't enabled at boot time.
Link: https://lkml.kernel.org/r/20220307074516.6920-3-dtcccc@linux.alibaba.com Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com> Reviewed-by: Marco Elver <elver@google.com> Tested-by: Peng Liu <liupeng256@huawei.com> Cc: Alexander Potapenko <glider@google.com> Cc: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Tianchen Ding [Tue, 22 Mar 2022 21:48:10 +0000 (14:48 -0700)]
kfence: allow re-enabling KFENCE after system startup
Patch series "provide the flexibility to enable KFENCE", v3.
If CONFIG_CONTIG_ALLOC is not supported, we fallback to try
alloc_pages_exact(). Allocating pages in this way has limits about
MAX_ORDER (default 11). So we will not support allocating kfence pool
after system startup with a large KFENCE_NUM_OBJECTS.
When handling failures in kfence_init_pool_late(), we pair
free_pages_exact() to alloc_pages_exact() for compatibility consideration,
though it actually does the same as free_contig_range().
This patch (of 2):
If once KFENCE is disabled by:
echo 0 > /sys/module/kfence/parameters/sample_interval
KFENCE could never be re-enabled until next rebooting.
Allow re-enabling it by writing a positive num to sample_interval.
Ira Weiny [Tue, 22 Mar 2022 21:47:58 +0000 (14:47 -0700)]
highmem: document kunmap_local()
Some users of kmap() add an offset to the kmap() address to be used
during the mapping.
When converting to kmap_local_page() the base address does not need to
be stored because any address within the page can be used in
kunmap_local(). However, this was not clear from the documentation and
cause some questions.[1]
Document that any address in the page can be used in kunmap_local() to
clarify this for future users.
The mm/ directory can almost fully be built with W=1, which would help
in local development. One remaining issue is missing prototype for
early_memremap_pgprot_adjust().
Thus add a declaration for this function. Use mm/internal.h instead of
asm/early_ioremap.h to avoid missing type definitions and unnecessary
exposure.
Link: https://lkml.kernel.org/r/20220314165724.16071-2-vbabka@suse.cz Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: David Hildenbrand <david@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Randy Dunlap [Tue, 22 Mar 2022 21:47:52 +0000 (14:47 -0700)]
mm/usercopy: return 1 from hardened_usercopy __setup() handler
__setup() handlers should return 1 if the command line option is handled
and 0 if not (or maybe never return 0; it just pollutes init's
environment). This prevents:
Unknown kernel command line parameters \
"BOOT_IMAGE=/boot/bzImage-517rc5 hardened_usercopy=off", will be \
passed to user space.
Run /sbin/init as init process
with arguments:
/sbin/init
with environment:
HOME=/
TERM=linux
BOOT_IMAGE=/boot/bzImage-517rc5
hardened_usercopy=off
or
hardened_usercopy=on
but when "hardened_usercopy=foo" is used, there is no Unknown kernel
command line parameter.
Return 1 to indicate that the boot option has been handled.
Print a warning if strtobool() returns an error on the option string,
but do not mark this as in unknown command line option and do not cause
init's environment to be polluted with this string.
Link: https://lkml.kernel.org/r/20220222034249.14795-1-rdunlap@infradead.org
Link: lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru Fixes: 4c217a5f0d841 ("usercopy: Allow boot cmdline disabling of hardening") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru> Acked-by: Chris von Recklinghausen <crecklin@redhat.com> Cc: Kees Cook <keescook@chromium.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Christophe Leroy [Tue, 22 Mar 2022 21:47:49 +0000 (14:47 -0700)]
mm: uninline copy_overflow()
While building a small config with CONFIG_CC_OPTIMISE_FOR_SIZE, I ended
up with more than 50 times the following function in vmlinux because GCC
doesn't honor the 'inline' keyword:
mm/zswap.c: allow handling just same-value filled pages
Zswap has an ability to efficiently store same-value filled pages, which
can be turned on and off using the "same_filled_pages_enabled"
parameter.
However, there is currently no way to enable just this (lightweight)
functionality, while not making use of the whole compressed page storage
machinery.
Add a "non_same_filled_pages_enabled" parameter which allows disabling
handling of pages that aren't same-value filled. This way zswap can be
run in such lightweight same-value filled pages only mode.
Hugh Dickins [Tue, 22 Mar 2022 21:47:40 +0000 (14:47 -0700)]
mm/thp: ClearPageDoubleMap in first page_add_file_rmap()
PageDoubleMap is maintained differently for anon and for shmem+file: the
shmem+file one was never cleared, because a safe place to do so could
not be found; so it would blight future use of the cached hugepage until
evicted.
See https://lore.kernel.org/lkml/1571938066-29031-1-git-send-email-yang.shi@linux.alibaba.com/
But page_add_file_rmap() does provide a safe place to do so (though later
than one might wish): allowing testing to return to an initial state
without a damaging drop_caches.
Link: https://lkml.kernel.org/r/61c5cf99-a962-9a25-597a-53ab1bd8fbc0@google.com Fixes: 97bfeea9c9fa ("thp, mlock: do not mlock PTE-mapped file huge pages") Signed-off-by: Hugh Dickins <hughd@google.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Oscar Salvador [Tue, 22 Mar 2022 21:47:37 +0000 (14:47 -0700)]
mm: only re-generate demotion targets when a numa node changes its N_CPU state
Abhishek reported that after patch [1], hotplug operations are taking
roughly double the expected time. [2]
The reason behind is that the CPU callbacks that
migrate_on_reclaim_init() sets always call set_migration_target_nodes()
whenever a CPU is brought up/down.
But we only care about numa nodes going from having cpus to become
cpuless, and vice versa, as that influences the demotion_target order.
We do already have two CPU callbacks (vmstat_cpu_online() and
vmstat_cpu_dead()) that check exactly that, so get rid of the CPU
callbacks in migrate_on_reclaim_init() and only call
set_migration_target_nodes() from vmstat_cpu_{dead,online}() whenever a
numa node change its N_CPU state.
drivers/base/memory: clarify adding and removing of memory blocks
Let's make it clearer at which places we actually add and remove memory
blocks -- streamlining the terminology -- and highlight which memory block
start out online and which start out as offline.
drivers/base/memory: determine and store zone for single-zone memory blocks
test_pages_in_a_zone() is just another nasty PFN walker that can easily
stumble over ZONE_DEVICE memory ranges falling into the same memory block
as ordinary system RAM: the memmap of parts of these ranges might possibly
be uninitialized. In fact, we observed (on an older kernel) with UBSAN:
UBSAN: Undefined behaviour in ./include/linux/mm.h:1133:50
index 7 is out of range for type 'zone [5]'
CPU: 121 PID: 35603 Comm: read_all Kdump: loaded Tainted: [...]
Hardware name: Dell Inc. PowerEdge R7425/08V001, BIOS 1.12.2 11/15/2019
Call Trace:
dump_stack+0x9a/0xf0
ubsan_epilogue+0x9/0x7a
__ubsan_handle_out_of_bounds+0x13a/0x181
test_pages_in_a_zone+0x3c4/0x500
show_valid_zones+0x1fa/0x380
dev_attr_show+0x43/0xb0
sysfs_kf_seq_show+0x1c5/0x440
seq_read+0x49d/0x1190
vfs_read+0xff/0x300
ksys_read+0xb8/0x170
do_syscall_64+0xa5/0x4b0
entry_SYSCALL_64_after_hwframe+0x6a/0xdf
RIP: 0033:0x7f01f4439b52
We seem to stumble over a memmap that contains a garbage zone id. While
we could try inserting pfn_to_online_page() calls, it will just make
memory offlining slower, because we use test_pages_in_a_zone() to make
sure we're offlining pages that all belong to the same zone.
Let's just get rid of this PFN walker and determine the single zone of a
memory block -- if any -- for early memory blocks during boot. For memory
onlining, we know the single zone already. Let's avoid any additional
memmap scanning and just rely on the zone information available during
boot.
For memory hot(un)plug, we only really care about memory blocks that:
* span a single zone (and, thereby, a single node)
* are completely System RAM (IOW, no holes, no ZONE_DEVICE)
If one of these conditions is not met, we reject memory offlining.
Hotplugged memory blocks (starting out offline), always meet both
conditions.
There are three scenarios to handle:
(1) Memory hot(un)plug
A memory block with zone == NULL cannot be offlined, corresponding to
our previous test_pages_in_a_zone() check.
After successful memory onlining/offlining, we simply set the zone
accordingly.
* Memory onlining: set the zone we just used for onlining
* Memory offlining: set zone = NULL
So a hotplugged memory block starts with zone = NULL. Once memory
onlining is done, we set the proper zone.
(2) Boot memory with !CONFIG_NUMA
We know that there is just a single pgdat, so we simply scan all zones
of that pgdat for an intersection with our memory block PFN range when
adding the memory block. If more than one zone intersects (e.g., DMA and
DMA32 on x86 for the first memory block) we set zone = NULL and
consequently mimic what test_pages_in_a_zone() used to do.
(3) Boot memory with CONFIG_NUMA
At the point in time we create the memory block devices during boot, we
don't know yet which nodes *actually* span a memory block. While we could
scan all zones of all nodes for intersections, overlapping nodes complicate
the situation and scanning all nodes is possibly expensive. But that
problem has already been solved by the code that sets the node of a memory
block and creates the link in the sysfs --
do_register_memory_block_under_node().
So, we hook into the code that sets the node id for a memory block. If
we already have a different node id set for the memory block, we know
that multiple nodes *actually* have PFNs falling into our memory block:
we set zone = NULL and consequently mimic what test_pages_in_a_zone() used
to do. If there is no node id set, we do the same as (2) for the given
node.
Note that the call order in driver_init() is:
-> memory_dev_init(): create memory block devices
-> node_dev_init(): link memory block devices to the node and set the
node id
So in summary, we detect if there is a single zone responsible for this
memory block and we consequently store the zone in that case in the
memory block, updating it during memory onlining/offlining.
Link: https://lkml.kernel.org/r/20220210184359.235565-3-david@redhat.com Signed-off-by: David Hildenbrand <david@redhat.com> Reported-by: Rafael Parra <rparrazo@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Rafael Parra <rparrazo@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/base/node: rename link_mem_sections() to register_memory_block_under_node()
Patch series "drivers/base/memory: determine and store zone for single-zone memory blocks", v2.
I remember talking to Michal in the past about removing
test_pages_in_a_zone(), which we use for:
* verifying that a memory block we intend to offline is really only managed
by a single zone. We don't support offlining of memory blocks that are
managed by multiple zones (e.g., multiple nodes, DMA and DMA32)
* exposing that zone to user space via
/sys/devices/system/memory/memory*/valid_zones
Now that I identified some more cases where test_pages_in_a_zone() might
go wrong, and we received an UBSAN report (see patch #3), let's get rid of
this PFN walker.
So instead of detecting the zone at runtime with test_pages_in_a_zone() by
scanning the memmap, let's determine and remember for each memory block if
it's managed by a single zone. The stored zone can then be used for the
above two cases, avoiding a manual lookup using test_pages_in_a_zone().
This avoids eventually stumbling over uninitialized memmaps in corner
cases, especially when ZONE_DEVICE ranges partly fall into memory block
(that are responsible for managing System RAM).
Handling memory onlining is easy, because we online to exactly one zone.
Handling boot memory is more tricky, because we want to avoid scanning all
zones of all nodes to detect possible zones that overlap with the physical
memory region of interest. Fortunately, we already have code that
determines the applicable nodes for a memory block, to create sysfs links
-- we'll hook into that.
Patch #1 is a simple cleanup I had laying around for a longer time.
Patch #2 contains the main logic to remove test_pages_in_a_zone() and
further details.
Let's adjust the stale terminology, making it match
unregister_memory_block_under_nodes() and
do_register_memory_block_under_node(). We're dealing with memory block
devices, which span 1..X memory sections.
Link: https://lkml.kernel.org/r/20220210184359.235565-1-david@redhat.com Link: https://lkml.kernel.org/r/20220210184359.235565-2-david@redhat.com Signed-off-by: David Hildenbrand <david@redhat.com> Acked-by: Oscar Salvador <osalvador@suse.de> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Michal Hocko <mhocko@suse.com> Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Rafael Parra <rparrazo@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Miaohe Lin [Tue, 22 Mar 2022 21:47:24 +0000 (14:47 -0700)]
mm/memory_hotplug: fix misplaced comment in offline_pages
It's misplaced since commit 09be3157421f ("mm, memory_hotplug: print
reason for the offlining failure"). Move it to the right place.
Link: https://lkml.kernel.org/r/20220207133643.23427-5-linmiaohe@huawei.com Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Miaohe Lin [Tue, 22 Mar 2022 21:47:22 +0000 (14:47 -0700)]
mm/memory_hotplug: clean up try_offline_node
We can use helper macro node_spanned_pages to check whether node spans
pages. And we can change the parameter of check_cpu_on_node to nid as
that's what it really cares. Thus we can further get rid of the local
variable pgdat and improve the readability a bit.
Link: https://lkml.kernel.org/r/20220207133643.23427-4-linmiaohe@huawei.com Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Miaohe Lin [Tue, 22 Mar 2022 21:47:19 +0000 (14:47 -0700)]
mm/memory_hotplug: avoid calling zone_intersects() for ZONE_NORMAL
If zid reaches ZONE_NORMAL, the caller will always get the NORMAL zone no
matter what zone_intersects() returns. So we can save some possible cpu
cycles by avoid calling zone_intersects() for ZONE_NORMAL.
Link: https://lkml.kernel.org/r/20220207133643.23427-3-linmiaohe@huawei.com Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/base/node: consolidate node device subsystem initialization in node_dev_init()
... and call node_dev_init() after memory_dev_init() from driver_init(),
so before any of the existing arch/subsys calls. All online nodes should
be known at that point: early during boot, arch code determines node and
zone ranges and sets the relevant nodes online; usually this happens in
setup_arch().
This is in line with memory_dev_init(), which initializes the memory
device subsystem and creates all memory block devices.
Similar to memory_dev_init(), panic() if anything goes wrong, we don't
want to continue with such basic initialization errors.
The important part is that node_dev_init() gets called after
memory_dev_init() and after cpu_dev_init(), but before any of the relevant
archs call register_cpu() to register the new cpu device under the node
device. The latter should be the case for the current users of
topology_init().
Link: https://lkml.kernel.org/r/20220203105212.30385-1-david@redhat.com Signed-off-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Tested-by: Anatoly Pugachev <matorola@gmail.com> (sparc64) Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Mike Rapoport <rppt@kernel.org> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Heiko Carstens <hca@linux.ibm.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Rich Felker <dalias@libc.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: "Rafael J. Wysocki" <rafael@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/base/memory: add memory block to memory group after registration succeeded
If register_memory() fails, we freed the memory block but already added
the memory block to the group list, not good. Let's defer adding the
block to the memory group to after registering the memory block device.
We do handle it properly during unregister_memory(), but that's not
called when the registration fails.
Link: https://lkml.kernel.org/r/20220128144540.153902-1-david@redhat.com Fixes: c7f55f08510a ("drivers/base/memory: introduce "memory groups" to logically group memory blocks") Signed-off-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Rafael J. Wysocki" <rafael@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Wei Yang [Tue, 22 Mar 2022 21:47:06 +0000 (14:47 -0700)]
memcg: do not tweak node in alloc_mem_cgroup_per_node_info
alloc_mem_cgroup_per_node_info is allocated for each possible node and
this used to be a problem because !node_online nodes didn't have
appropriate data structure allocated. This has changed by "mm: handle
uninitialized numa nodes gracefully" so we can drop the special casing
here.
Link: https://lkml.kernel.org/r/20220127085305.20890-7-mhocko@kernel.org Signed-off-by: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Michal Hocko <mhocko@suse.com> Cc: David Hildenbrand <david@redhat.com> Cc: Alexey Makhalov <amakhalov@vmware.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Christoph Lameter <cl@linux.com> Cc: Nico Pache <npache@redhat.com> Cc: Wei Yang <richard.weiyang@gmail.com> Cc: Mike Rapoport <rppt@linux.ibm.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Rafael Aquini <raquini@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko [Tue, 22 Mar 2022 21:47:03 +0000 (14:47 -0700)]
mm: make free_area_init_node aware of memory less nodes
free_area_init_node is also called from memory less node initialization
path (free_area_init_memoryless_node). It doesn't really make much sense
to display the physical memory range for those nodes: Initmem setup node
XX [mem 0x0000000000000000-0x0000000000000000]
Instead be explicit that the node is memoryless: Initmem setup node XX as
memoryless
Link: https://lkml.kernel.org/r/20220127085305.20890-6-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Rafael Aquini <raquini@redhat.com> Acked-by: David Hildenbrand <david@redhat.com> Reviewed-by: Mike Rapoport <rppt@linux.ibm.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: Alexey Makhalov <amakhalov@vmware.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Nico Pache <npache@redhat.com> Cc: Tejun Heo <tj@kernel.org> Cc: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko [Tue, 22 Mar 2022 21:47:00 +0000 (14:47 -0700)]
mm, memory_hotplug: reorganize new pgdat initialization
When a !node_online node is brought up it needs a hotplug specific
initialization because the node could be either uninitialized yet or it
could have been recycled after previous hotremove. hotadd_init_pgdat is
responsible for that.
Internal pgdat state is initialized at two places currently
- hotadd_init_pgdat
- free_area_init_core_hotplug
There is no real clear cut what should go where but this patch's chosen to
move the whole internal state initialization into
free_area_init_core_hotplug. hotadd_init_pgdat is still responsible to
pull all the parts together - most notably to initialize zonelists because
those depend on the overall topology.
This patch doesn't introduce any functional change.
Link: https://lkml.kernel.org/r/20220127085305.20890-5-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Rafael Aquini <raquini@redhat.com> Acked-by: David Hildenbrand <david@redhat.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: Alexey Makhalov <amakhalov@vmware.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Mike Rapoport <rppt@linux.ibm.com> Cc: Nico Pache <npache@redhat.com> Cc: Tejun Heo <tj@kernel.org> Cc: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko [Tue, 22 Mar 2022 21:46:57 +0000 (14:46 -0700)]
mm, memory_hotplug: drop arch_free_nodedata
Prior to "mm: handle uninitialized numa nodes gracefully" memory hotplug
used to allocate pgdat when memory has been added to a node
(hotadd_init_pgdat) arch_free_nodedata has been only used in the failure
path because once the pgdat is exported (to be visible by NODA_DATA(nid))
it cannot really be freed because there is no synchronization available
for that.
pgdat is allocated for each possible nodes now so the memory hotplug
doesn't need to do the ever use arch_free_nodedata so drop it.
This patch doesn't introduce any functional change.
Link: https://lkml.kernel.org/r/20220127085305.20890-4-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Rafael Aquini <raquini@redhat.com> Acked-by: David Hildenbrand <david@redhat.com> Acked-by: Mike Rapoport <rppt@linux.ibm.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: Alexey Makhalov <amakhalov@vmware.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Nico Pache <npache@redhat.com> Cc: Tejun Heo <tj@kernel.org> Cc: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko [Tue, 22 Mar 2022 21:46:54 +0000 (14:46 -0700)]
mm: handle uninitialized numa nodes gracefully
We have had several reports [1][2][3] that page allocator blows up when an
allocation from a possible node is requested. The underlying reason is
that NODE_DATA for the specific node is not allocated.
NUMA specific initialization is arch specific and it can vary a lot. E.g.
x86 tries to initialize all nodes that have some cpu affinity (see
init_cpu_to_node) but this can be insufficient because the node might be
cpuless for example.
One way to address this problem would be to check for !node_online nodes
when trying to get a zonelist and silently fall back to another node.
That is unfortunately adding a branch into allocator hot path and it
doesn't handle any other potential NODE_DATA users.
This patch takes a different approach (following a lead of [3]) and it pre
allocates pgdat for all possible nodes in an arch indipendent code -
free_area_init. All uninitialized nodes are treated as memoryless nodes.
node_state of the node is not changed because that would lead to other
side effects - e.g. sysfs representation of such a node and from past
discussions [4] it is known that some tools might have problems digesting
that.
Newly allocated pgdat only gets a minimal initialization and the rest of
the work is expected to be done by the memory hotplug - hotadd_new_pgdat
(renamed to hotadd_init_pgdat).
generic_alloc_nodedata is changed to use the memblock allocator because
neither page nor slab allocators are available at the stage when all
pgdats are allocated. Hotplug doesn't allocate pgdat anymore so we can
use the early boot allocator. The only arch specific implementation is
ia64 and that is changed to use the early allocator as well.
Michal Hocko [Tue, 22 Mar 2022 21:46:51 +0000 (14:46 -0700)]
mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG
Patch series "mm, memory_hotplug: handle unitialized numa node gracefully".
The core of the fix is patch 2 which also links existing bug reports. The
high level goal is to have all possible numa nodes have their pgdat
allocated and initialized so
for_each_possible_node(nid)
NODE_DATA(nid)
will never return garbage. This has proven to be problem in several
places when an offline numa node is used for an allocation just to realize
that node_data and therefore allocation fallback zonelists are not
initialized and such an allocation request blows up.
There were attempts to address that by checking node_online in several
places including the page allocator. This patchset approaches the problem
from a different perspective and instead of special casing, which just
adds a runtime overhead, it allocates pglist_data for each possible node.
This can add some memory overhead for platforms with high number of
possible nodes if they do not contain any memory. This should be a rather
rare configuration though.
How to test this? David has provided and excellent howto:
http://lkml.kernel.org/r/6e5ebc19-890c-b6dd-1924-9f25c441010d@redhat.com
Patches 1 and 3-6 are mostly cleanups. The patchset has been reviewed by
Rafael (thanks!) and the core fix tested by Rafael and Alexey (thanks to
both). David has tested as per instructions above and hasn't found any
fallouts in the memory hotplug scenarios.
This patch (of 6):
This is a preparatory patch and it doesn't introduce any functional
change. It merely pulls out arch_alloc_nodedata (and co) outside of
CONFIG_MEMORY_HOTPLUG because the following patch will need to call this
from the generic MM code.
Link: https://lkml.kernel.org/r/20220127085305.20890-1-mhocko@kernel.org Link: https://lkml.kernel.org/r/20220127085305.20890-2-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Rafael Aquini <raquini@redhat.com> Acked-by: David Hildenbrand <david@redhat.com> Acked-by: Mike Rapoport <rppt@linux.ibm.com> Reviewed-by: Oscar Salvador <osalvador@suse.de> Reviewed-by: Wei Yang <richard.weiyang@gmail.com> Cc: Alexey Makhalov <amakhalov@vmware.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Nico Pache <npache@redhat.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: madvise: skip unmapped vma holes passed to process_madvise
The process_madvise() system call is expected to skip holes in vma passed
through 'struct iovec' vector list. But do_madvise, which
process_madvise() calls for each vma, returns ENOMEM in case of unmapped
holes, despite the VMA is processed.
Thus process_madvise() should treat ENOMEM as expected and consider the
VMA passed to as processed and continue processing other vma's in the
vector list. Returning -ENOMEM to user, despite the VMA is processed,
will be unable to figure out where to start the next madvise.
Link: https://lkml.kernel.org/r/4f091776142f2ebf7b94018146de72318474e686.1647008754.git.quic_charante@quicinc.com Fixes: c5a294b9e6e1("mm/madvise: introduce process_madvise() syscall: an external memory hinting API") Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com> Cc: David Rientjes <rientjes@google.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Nadav Amit <nadav.amit@gmail.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: madvise: return correct bytes advised with process_madvise
Patch series "mm: madvise: return correct bytes processed with
process_madvise", v2. With the process_madvise(), always choose to return
non zero processed bytes over an error. This can help the user to know on
which VMA, passed in the 'struct iovec' vector list, is failed to advise
thus can take the decission of retrying/skipping on that VMA.
This patch (of 2):
The process_madvise() system call returns error even after processing some
VMA's passed in the 'struct iovec' vector list which leaves the user
confused to know where to restart the advise next. It is also against
this syscall man page[1] documentation where it mentions that "return
value may be less than the total number of requested bytes, if an error
occurred after some iovec elements were already processed.".
Consider a user passed 10 VMA's in the 'struct iovec' vector list of which
9 are processed but one. Then it just returns the error caused on that
failed VMA despite the first 9 VMA's processed, leaving the user confused
about on which VMA it is failed. Returning the number of bytes processed
here can help the user to know which VMA it is failed on and thus can
retry/skip the advise on that VMA.
Yang Yang [Tue, 22 Mar 2022 21:46:33 +0000 (14:46 -0700)]
mm/vmstat: add event for ksm swapping in copy
When faults in from swap what used to be a KSM page and that page had been
swapped in before, system has to make a copy, and leaves remerging the
pages to a later pass of ksmd.
That is not good for performace, we'd better to reduce this kind of copy.
There are some ways to reduce it, for example lessen swappiness or
madvise(, , MADV_MERGEABLE) range. So add this event to support doing
this tuning. Just like this patch: "mm, THP, swap: add THP swapping out
fallback counting".
Link: https://lkml.kernel.org/r/20220113023839.758845-1-yang.yang29@zte.com.cn Signed-off-by: Yang Yang <yang.yang29@zte.com.cn> Reviewed-by: Ran Xiaokai <ran.xiaokai@zte.com.cn> Cc: Hugh Dickins <hughd@google.com> Cc: Yang Shi <yang.shi@linux.alibaba.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Saravanan D <saravanand@fb.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Weiner [Tue, 22 Mar 2022 21:46:30 +0000 (14:46 -0700)]
mm: page_io: fix psi memory pressure error on cold swapins
Once upon a time, all swapins counted toward memory pressure[1]. Then
Joonsoo introduced workingset detection for anonymous pages and we gained
the ability to distinguish hot from cold swapins[2][3]. But we failed to
update swap_readpage() accordingly, and now we account partial memory
pressure in the swapin path of cold memory.
Not for all situations - which adds more inconsistency: paths using the
conventional submit_bio() and lock_page() route will not see much pressure
- unless storage itself is heavily congested and the bio submissions
stall. ZRAM and ZSWAP do most of the work directly from swap_readpage()
and will see all swapins reflected as pressure.
IOW, a workload doing cold swapins could see little to no pressure
reported with on-disk swap, but potentially high pressure with a zram or
zswap backend. That confuses any psi-based health monitoring, load
shedding, proactive reclaim, or userspace OOM killing schemes that might
be in place for the workload.
Restore consistency by making all swapin stall accounting conditional on
the page actually being part of the workingset.
[1] commit 9c75b570ac89 ("mm/page_io.c: annotate refault stalls from swap_readpage")
[2] commit ddbc26d24535 ("mm/swap: implement workingset detection for anonymous LRU")
[3] commit 6e10a5dbcb2f ("mm/swap: don't SetPageWorkingset unconditionally during swapin")
Link: https://lkml.kernel.org/r/20220214214921.419687-1-hannes@cmpxchg.org Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reported-by: CGEL <cgel.zte@gmail.com> Acked-by: Minchan Kim <minchan@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Yu Zhao <yuzhao@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Huang Ying [Tue, 22 Mar 2022 21:46:27 +0000 (14:46 -0700)]
memory tiering: skip to scan fast memory
If the NUMA balancing isn't used to optimize the page placement among
sockets but only among memory types, the hot pages in the fast memory
node couldn't be migrated (promoted) to anywhere. So it's unnecessary
to scan the pages in the fast memory node via changing their PTE/PMD
mapping to be PROT_NONE. So that the page faults could be avoided too.
In the test, if only the memory tiering NUMA balancing mode is enabled,
the number of the NUMA balancing hint faults for the DRAM node is
reduced to almost 0 with the patch. While the benchmark score doesn't
change visibly.
Link: https://lkml.kernel.org/r/20220221084529.1052339-4-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Suggested-by: Dave Hansen <dave.hansen@linux.intel.com> Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com> Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Oscar Salvador <osalvador@suse.de> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Rik van Riel <riel@surriel.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Zi Yan <ziy@nvidia.com> Cc: Wei Xu <weixugc@google.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com> Cc: Feng Tang <feng.tang@intel.com> Cc: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Huang Ying [Tue, 22 Mar 2022 21:46:23 +0000 (14:46 -0700)]
NUMA balancing: optimize page placement for memory tiering system
With the advent of various new memory types, some machines will have
multiple types of memory, e.g. DRAM and PMEM (persistent memory). The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are usually
different.
In such system, because of the memory accessing pattern changing etc,
some pages in the slow memory may become hot globally. So in this
patch, the NUMA balancing mechanism is enhanced to optimize the page
placement among the different memory types according to hot/cold
dynamically.
In a typical memory tiering system, there are CPUs, fast memory and slow
memory in each physical NUMA node. The CPUs and the fast memory will be
put in one logical node (called fast memory node), while the slow memory
will be put in another (faked) logical node (called slow memory node).
That is, the fast memory is regarded as local while the slow memory is
regarded as remote. So it's possible for the recently accessed pages in
the slow memory node to be promoted to the fast memory node via the
existing NUMA balancing mechanism.
The original NUMA balancing mechanism will stop to migrate pages if the
free memory of the target node becomes below the high watermark. This
is a reasonable policy if there's only one memory type. But this makes
the original NUMA balancing mechanism almost do not work to optimize
page placement among different memory types. Details are as follows.
It's the common cases that the working-set size of the workload is
larger than the size of the fast memory nodes. Otherwise, it's
unnecessary to use the slow memory at all. So, there are almost always
no enough free pages in the fast memory nodes, so that the globally hot
pages in the slow memory node cannot be promoted to the fast memory
node. To solve the issue, we have 2 choices as follows,
a. Ignore the free pages watermark checking when promoting hot pages
from the slow memory node to the fast memory node. This will
create some memory pressure in the fast memory node, thus trigger
the memory reclaiming. So that, the cold pages in the fast memory
node will be demoted to the slow memory node.
b. Define a new watermark called wmark_promo which is higher than
wmark_high, and have kswapd reclaiming pages until free pages reach
such watermark. The scenario is as follows: when we want to promote
hot-pages from a slow memory to a fast memory, but fast memory's free
pages would go lower than high watermark with such promotion, we wake
up kswapd with wmark_promo watermark in order to demote cold pages and
free us up some space. So, next time we want to promote hot-pages we
might have a chance of doing so.
The choice "a" may create high memory pressure in the fast memory node.
If the memory pressure of the workload is high, the memory pressure
may become so high that the memory allocation latency of the workload
is influenced, e.g. the direct reclaiming may be triggered.
The choice "b" works much better at this aspect. If the memory
pressure of the workload is high, the hot pages promotion will stop
earlier because its allocation watermark is higher than that of the
normal memory allocation. So in this patch, choice "b" is implemented.
A new zone watermark (WMARK_PROMO) is added. Which is larger than the
high watermark and can be controlled via watermark_scale_factor.
In addition to the original page placement optimization among sockets,
the NUMA balancing mechanism is extended to be used to optimize page
placement according to hot/cold among different memory types. So the
sysctl user space interface (numa_balancing) is extended in a backward
compatible way as follow, so that the users can enable/disable these
functionality individually.
The sysctl is converted from a Boolean value to a bits field. The
definition of the flags is,
We have tested the patch with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent
Memory Model. The test results shows that the pmbench score can
improve up to 95.9%.
Thanks Andrew Morton to help fix the document format error.
Link: https://lkml.kernel.org/r/20220221084529.1052339-3-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com> Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Oscar Salvador <osalvador@suse.de> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Rik van Riel <riel@surriel.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Zi Yan <ziy@nvidia.com> Cc: Wei Xu <weixugc@google.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Feng Tang <feng.tang@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Huang Ying [Tue, 22 Mar 2022 21:46:20 +0000 (14:46 -0700)]
NUMA Balancing: add page promotion counter
Patch series "NUMA balancing: optimize memory placement for memory tiering system", v13
With the advent of various new memory types, some machines will have
multiple types of memory, e.g. DRAM and PMEM (persistent memory). The
memory subsystem of these machines can be called memory tiering system,
because the performance of the different types of memory are different.
After commit b62f840a5a6f ("device-dax: "Hotplug" persistent memory for
use like normal RAM"), the PMEM could be used as the cost-effective
volatile memory in separate NUMA nodes. In a typical memory tiering
system, there are CPUs, DRAM and PMEM in each physical NUMA node. The
CPUs and the DRAM will be put in one logical node, while the PMEM will
be put in another (faked) logical node.
To optimize the system overall performance, the hot pages should be
placed in DRAM node. To do that, we need to identify the hot pages in
the PMEM node and migrate them to DRAM node via NUMA migration.
In the original NUMA balancing, there are already a set of existing
mechanisms to identify the pages recently accessed by the CPUs in a node
and migrate the pages to the node. So we can reuse these mechanisms to
build the mechanisms to optimize the page placement in the memory
tiering system. This is implemented in this patchset.
At the other hand, the cold pages should be placed in PMEM node. So, we
also need to identify the cold pages in the DRAM node and migrate them
to PMEM node.
In commit a1ca6f335d1c ("mm/migrate: demote pages during reclaim"), a
mechanism to demote the cold DRAM pages to PMEM node under memory
pressure is implemented. Based on that, the cold DRAM pages can be
demoted to PMEM node proactively to free some memory space on DRAM node
to accommodate the promoted hot PMEM pages. This is implemented in this
patchset too.
We have tested the solution with the pmbench memory accessing benchmark
with the 80:20 read/write ratio and the Gauss access address
distribution on a 2 socket Intel server with Optane DC Persistent Memory
Model. The test results shows that the pmbench score can improve up to
95.9%.
This patch (of 3):
In a system with multiple memory types, e.g. DRAM and PMEM, the CPU
and DRAM in one socket will be put in one NUMA node as before, while
the PMEM will be put in another NUMA node as described in the
description of the commit b62f840a5a6f ("device-dax: "Hotplug"
persistent memory for use like normal RAM"). So, the NUMA balancing
mechanism will identify all PMEM accesses as remote access and try to
promote the PMEM pages to DRAM.
To distinguish the number of the inter-type promoted pages from that of
the inter-socket migrated pages. A new vmstat count is added. The
counter is per-node (count in the target node). So this can be used to
identify promotion imbalance among the NUMA nodes.
Link: https://lkml.kernel.org/r/20220301085329.3210428-1-ying.huang@intel.com Link: https://lkml.kernel.org/r/20220221084529.1052339-1-ying.huang@intel.com Link: https://lkml.kernel.org/r/20220221084529.1052339-2-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Tested-by: Baolin Wang <baolin.wang@linux.alibaba.com> Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: Michal Hocko <mhocko@suse.com> Cc: Rik van Riel <riel@surriel.com> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Zi Yan <ziy@nvidia.com> Cc: Wei Xu <weixugc@google.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com> Cc: Feng Tang <feng.tang@intel.com> Cc: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hari Bathini [Tue, 22 Mar 2022 21:46:17 +0000 (14:46 -0700)]
powerpc/fadump: opt out from freeing pages on cma activation failure
With commit 1033151e9b41 ("powerpc/fadump: Reservationless firmware
assisted dump"), Linux kernel's Contiguous Memory Allocator (CMA) based
reservation was introduced in fadump. That change was aimed at using CMA
to let applications utilize the memory reserved for fadump while blocking
it from being used for kernel pages. The assumption was, even if CMA
activation fails for whatever reason, the memory still remains reserved to
avoid it from being used for kernel pages. But commit 023addc76f0a
("mm/cma: expose all pages to the buddy if activation of an area fails")
breaks this assumption as it started exposing all pages to buddy allocator
on CMA activation failure. It led to warning messages like below while
running crash-utility on vmcore of a kernel having above two commits:
Hari Bathini [Tue, 22 Mar 2022 21:46:14 +0000 (14:46 -0700)]
mm/cma: provide option to opt out from exposing pages on activation failure
Patch series "powerpc/fadump: handle CMA activation failure appropriately", v3.
Commit 023addc76f0a ("mm/cma: expose all pages to the buddy if
activation of an area fails") started exposing all pages to buddy
allocator on CMA activation failure. But there can be CMA users that
want to handle the reserved memory differently on CMA allocation
failure.
Provide an option to opt out from exposing pages to buddy for such
cases.
Link: https://lkml.kernel.org/r/20220117075246.36072-1-hbathini@linux.ibm.com Link: https://lkml.kernel.org/r/20220117075246.36072-2-hbathini@linux.ibm.com Signed-off-by: Hari Bathini <hbathini@linux.ibm.com> Reviewed-by: David Hildenbrand <david@redhat.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Mahesh Salgaonkar <mahesh@linux.ibm.com> Cc: Sourabh Jain <sourabhjain@linux.ibm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hugh Dickins [Tue, 22 Mar 2022 21:46:11 +0000 (14:46 -0700)]
mm/thp: refix __split_huge_pmd_locked() for migration PMD
Migration entries do not contribute to a page's reference count: move
__split_huge_pmd_locked()'s page_ref_add() into pmd_migration's else
block (along with the page_count() check - a page is quite likely to
have reference count frozen to 0 when a migration entry is found).
This will fix a very rare anonymous memory leak, after a
split_huge_pmd() raced with an anon split_huge_page() or an anon THP
migrate_pages(): since the wrongly raised refcount stopped the page
(perhaps small, perhaps huge, depending on when the race hit) from ever
being freed.
At first I thought there were worse risks, from prematurely unfreezing a
frozen page: but now think that would only affect page cache pages,
which do not come this way (except for anonymous pages in swap cache,
perhaps).
Link: https://lkml.kernel.org/r/84792468-f512-e48f-378c-e34c3641e97@google.com Fixes: c43d124cc97e ("mm/thp: fix __split_huge_pmd_locked() for migration PMD") Signed-off-by: Hugh Dickins <hughd@google.com> Reviewed-by: Yang Shi <shy828301@gmail.com> Cc: Ralph Campbell <rcampbell@nvidia.com> Cc: Zi Yan <ziy@nvidia.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
andrew.yang [Tue, 22 Mar 2022 21:46:08 +0000 (14:46 -0700)]
mm/migrate: fix race between lock page and clear PG_Isolated
When memory is tight, system may start to compact memory for large
continuous memory demands. If one process tries to lock a memory page
that is being locked and isolated for compaction, it may wait a long time
or even forever. This is because compaction will perform non-atomic
PG_Isolated clear while holding page lock, this may overwrite PG_waiters
set by the process that can't obtain the page lock and add itself to the
waiting queue to wait for the lock to be unlocked.
Huang Ying [Tue, 22 Mar 2022 21:46:05 +0000 (14:46 -0700)]
mm,migrate: fix establishing demotion target
In commit ad338e16d3db ("mm: migrate: support multiple target nodes
demotion"), after the first demotion target node is found, we will
continue to check the next candidate obtained via find_next_best_node().
This is to find all demotion target nodes with same NUMA distance. But
one side effect of find_next_best_node() is that the candidate node
returned will be set in "used" parameter, even if the candidate node isn't
passed in the following NUMA distance checking, the candidate node will
not be used as demotion target node for the following nodes. For example,
for system as follows,
when we establish demotion target node for node 0, in the first round node
2 is added to the demotion target node set. Then in the second round,
node 3 is checked and failed because distance(0, 3) > distance(0, 2). But
node 3 is set in "used" nodemask too. When we establish demotion target
node for node 1, there is no available node. This is wrong, node 3 should
be set as the demotion target of node 1.
To fix this, if the candidate node is failed to pass the distance
checking, it will be cleared in "used" nodemask. So that it can be used
for the following node.
The bug can be reproduced and fixed with this patch on a 2 socket server
machine with DRAM and PMEM.
Link: https://lkml.kernel.org/r/20220128055940.1792614-1-ying.huang@intel.com Fixes: ad338e16d3db ("mm: migrate: support multiple target nodes demotion") Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com> Cc: Baolin Wang <baolin.wang@linux.alibaba.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Zi Yan <ziy@nvidia.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Yang Shi <shy828301@gmail.com> Cc: zhongjiang-ali <zhongjiang-ali@linux.alibaba.com> Cc: Xunlei Pang <xlpang@linux.alibaba.com> Cc: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hugh Dickins [Tue, 22 Mar 2022 21:45:59 +0000 (14:45 -0700)]
mempolicy: mbind_range() set_policy() after vma_merge()
v2.6.34 commit d48f50e31fa4 ("mm: fix mbind vma merge problem") introduced
vma_merge() to mbind_range(); but unlike madvise, mlock and mprotect, it
put a "continue" to next vma where its precedents go to update flags on
current vma before advancing: that left vma with the wrong setting in the
infamous vma_merge() case 8.
v3.10 commit 344ed2c40a43 ("mm: merging memory blocks resets mempolicy")
tried to fix that in vma_adjust(), without fully understanding the issue.
v3.11 commit e7b0d2d2cc4c ("mm: mempolicy: fix mbind_range() &&
vma_adjust() interaction") reverted that, and went about the fix in the
right way, but chose to optimize out an unnecessary mpol_dup() with a
prior mpol_equal() test. But on tmpfs, that also pessimized out the vital
call to its ->set_policy(), leaving the new mbind unenforced.
The user visible effect was that the pages got allocated on the local
node (happened to be 0), after the mbind() caller had specifically
asked for them to be allocated on node 1. There was not any page
migration involved in the case reported: the pages simply got allocated
on the wrong node.
Just delete that optimization now (though it could be made conditional on
vma not having a set_policy). Also remove the "next" variable: it turned
out to be blameless, but also pointless.
Baolin Wang [Tue, 22 Mar 2022 21:45:56 +0000 (14:45 -0700)]
mm: compaction: cleanup the compaction trace events
As Steven suggested [1], we should access the pointers from the trace
event to avoid dereferencing them to the tracepoint function when the
tracepoint is disabled.
mm: vmscan: fix documentation for page_check_references()
Commit b6df587f15fc ("mm/vmscan: protect the workingset on anonymous
LRU") requires to look twice for both mapped anon/file pages are used
more than once to take the decission of reclaim or activation. Correct
the documentation accordingly.
Link: https://lkml.kernel.org/r/1646925640-21324-1-git-send-email-quic_charante@quicinc.com Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm: workingset: replace IRQ-off check with a lockdep assert.
Commit 9df6368ade033 ("mm: workingset: add vmstat counter for shadow
nodes") introduced an IRQ-off check to ensure that a lock is held which
also disabled interrupts. This does not work the same way on PREEMPT_RT
because none of the locks, that are held, disable interrupts.
Replace this check with a lockdep assert which ensures that the lock is
held.
Link: https://lkml.kernel.org/r/20220301122143.1521823-3-bigeasy@linutronix.de Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Tejun Heo <tj@kernel.org> Cc: Zefan Li <lizefan.x@bytedance.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Marcelo Tosatti [Tue, 22 Mar 2022 21:45:47 +0000 (14:45 -0700)]
mm: lru_cache_disable: replace work queue synchronization with synchronize_rcu
On systems that run FIFO:1 applications that busy loop, any SCHED_OTHER
task that attempts to execute on such a CPU (such as work threads) will
not be scheduled, which leads to system hangs.
Commit 3b82afd8791892578 ("mm: disable LRU pagevec during the migration
temporarily") relies on queueing work items on all online CPUs to ensure
visibility of lru_disable_count.
To fix this, replace the usage of work items with synchronize_rcu,
which provides the same guarantees.
Readers of lru_disable_count are protected by either disabling
preemption or rcu_read_lock:
Since v5.1 kernel, synchronize_rcu() is guaranteed to wait on
preempt_disable() regions of code. So any CPU which sees
lru_disable_count = 0 will have exited the critical section when
synchronize_rcu() returns.
Link: https://lkml.kernel.org/r/Yin7hDxdt0s/x+fp@fuller.cnet Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Reviewed-by: Nicolas Saenz Julienne <nsaenzju@redhat.com> Acked-by: Minchan Kim <minchan@kernel.org> Cc: Matthew Wilcox <willy@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Juri Lelli <juri.lelli@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Since commit 7a950e9c2c53 ("mm/list_lru.c: fix list_lru_count_node() to
be race free"), we are tracking the total number of lru entries in a
list_lru_node in its nr_items field.
In the case of memcg_reparent_list_lru_node(), there is nothing to be
done if nr_items is 0. We don't even need to take the nlru->lock as no
new lru entry could be added by a racing list_lru_add() to the draining
src_idx memcg at this point.
On systems that serve a lot of containers, it is possible that there can
be thousands of list_lru's present due to the fact that each container
may mount its own container specific filesystems. As a typical
container uses only a few cpus, it is likely that only the list_lru_node
that contains those cpus will be utilized while the rests may be empty.
In other words, there can be a lot of list_lru_node with 0 nr_items.
By skipping a lock/unlock operation and loading a cacheline from
memcg_lrus, a sizeable number of cpu cycles can be saved. That can be
substantial if we are talking about thousands of list_lru_node's with 0
nr_items.
Link: https://lkml.kernel.org/r/20220309144000.1470138-1-longman@redhat.com Signed-off-by: Waiman Long <longman@redhat.com> Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev> Cc: Muchun Song <songmuchun@bytedance.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Shakeel Butt <shakeelb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hugh Dickins [Tue, 22 Mar 2022 21:45:41 +0000 (14:45 -0700)]
mm: __isolate_lru_page_prepare() in isolate_migratepages_block()
__isolate_lru_page_prepare() conflates two unrelated functions, with the
flags to one disjoint from the flags to the other; and hides some of the
important checks outside of isolate_migratepages_block(), where the
sequence is better to be visible. It comes from the days of lumpy
reclaim, before compaction, when the combination made more sense.
Move what's needed by mm/compaction.c isolate_migratepages_block() inline
there, and what's needed by mm/vmscan.c isolate_lru_pages() inline there.
Shorten "isolate_mode" to "mode", so the sequence of conditions is easier
to read. Declare a "mapping" variable, to save one call to page_mapping()
(but not another: calling again after page is locked is necessary).
Simplify isolate_lru_pages() with a "move_to" list pointer.
Link: https://lkml.kernel.org/r/879d62a8-91cc-d3c6-fb3b-69768236df68@google.com Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: David Rientjes <rientjes@google.com> Reviewed-by: Alex Shi <alexs@kernel.org> Cc: Alexander Duyck <alexander.duyck@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hugh Dickins [Tue, 22 Mar 2022 21:45:38 +0000 (14:45 -0700)]
mm/fs: delete PF_SWAPWRITE
PF_SWAPWRITE has been redundant since v3.2 commit 0a365446b381 ("mm:
vmscan: do not writeback filesystem pages in direct reclaim").
Coincidentally, NeilBrown's current patch "remove inode_congested()"
deletes may_write_to_inode(), which appeared to be the one function which
took notice of PF_SWAPWRITE. But if you study the old logic, and the
conditions under which may_write_to_inode() was called, you discover that
flag and function have been pointless for a decade.
Link: https://lkml.kernel.org/r/75e80e7-742d-e3bd-531-614db8961e4@google.com Signed-off-by: Hugh Dickins <hughd@google.com> Cc: NeilBrown <neilb@suse.de> Cc: Jan Kara <jack@suse.de> Cc: "Darrick J. Wong" <djwong@kernel.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Matthew Wilcox <willy@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix following coccicheck warning:
tools/testing/selftests/vm/userfaultfd.c:556:23-24:
WARNING this kind of initialization is deprecated
`unsigned long page_nr = *(&page_nr)` has the same form of
uninitialized_var() macro. I remove the redundant assignement. It has
been tested with gcc (Debian 8.3.0-6) 8.3.0.
The patch which removed uninitialized_var() is:
https://lore.kernel.org/all/20121028102007.GA7547@gmail.com/ And there is
very few "/* GCC */" comments in the Linux kernel code now.
Nadav Amit [Tue, 22 Mar 2022 21:45:32 +0000 (14:45 -0700)]
userfaultfd: provide unmasked address on page-fault
Userfaultfd is supposed to provide the full address (i.e., unmasked) of
the faulting access back to userspace. However, that is not the case for
quite some time.
Even running "userfaultfd_demo" from the userfaultfd man page provides the
wrong output (and contradicts the man page). Notice that
"UFFD_EVENT_PAGEFAULT event" shows the masked address (7fc5e30b3000) and
not the first read address (0x7fc5e30b300f).
Address returned by mmap() = 0x7fc5e30b3000
fault_handler_thread():
poll() returns: nready = 1; POLLIN = 1; POLLERR = 0
UFFD_EVENT_PAGEFAULT event: flags = 0; address = 7fc5e30b3000
(uffdio_copy.copy returned 4096)
Read address 0x7fc5e30b300f in main(): A
Read address 0x7fc5e30b340f in main(): A
Read address 0x7fc5e30b380f in main(): A
Read address 0x7fc5e30b3c0f in main(): A
The exact address is useful for various reasons and specifically for
prefetching decisions. If it is known that the memory is populated by
certain objects whose size is not page-aligned, then based on the faulting
address, the uffd-monitor can decide whether to prefetch and prefault the
adjacent page.
This bug has been for quite some time in the kernel: since commit ce0aac205752 ("mm: use vmf->address instead of of vmf->virtual_address")
vmf->virtual_address"), which dates back to 2016. A concern has been
raised that existing userspace application might rely on the old/wrong
behavior in which the address is masked. Therefore, it was suggested to
provide the masked address unless the user explicitly asks for the exact
address.
Add a new userfaultfd feature UFFD_FEATURE_EXACT_ADDRESS to direct
userfaultfd to provide the exact address. Add a new "real_address" field
to vmf to hold the unmasked address. Provide the address to userspace
accordingly.
Initialize real_address in various code-paths to be consistent with
address, even when it is not used, to be on the safe side.
Link: https://lkml.kernel.org/r/20220218041003.3508-1-namit@vmware.com Signed-off-by: Nadav Amit <namit@vmware.com> Acked-by: Peter Xu <peterx@redhat.com> Reviewed-by: David Hildenbrand <david@redhat.com> Acked-by: Mike Rapoport <rppt@linux.ibm.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Howells [Tue, 22 Mar 2022 21:45:26 +0000 (14:45 -0700)]
mm/hugetlb.c: export PageHeadHuge()
Export PageHeadHuge() - it's used by folio_test_hugetlb() and thence by
such as folio_file_page() and folio_contains(). Matthew suggested I use
the first of those instead of doing the same calculation manually - but I
can't call it from a module.
Kirill suggested rearranging things to put it in a header, but that
introduces header dependencies because of where constants are defined.
[akpm@linux-foundation.org: s/EXPORT_SYMBOL/EXPORT_SYMBOL_GPL/, per Christoph]
Mike Kravetz [Tue, 22 Mar 2022 21:45:20 +0000 (14:45 -0700)]
hugetlb: clean up potential spectre issue warnings
Recently introduced code allows numa nodes to be specified on the kernel
command line for hugetlb allocations or CMA reservations. The node
values are user specified and used as indicies into arrays. This
generated the following smatch warnings:
ARCH_WANT_GENERAL_HUGETLB config has duplicate definitions on platforms
that subscribe it. Instead make it a generic config option which can be
selected on applicable platforms when required.
Link: https://lkml.kernel.org/r/1643718465-4324-1-git-send-email-anshuman.khandual@arm.com Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Russell King <linux@armlinux.org.uk> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Muchun Song [Tue, 22 Mar 2022 21:45:09 +0000 (14:45 -0700)]
selftests: vm: add a hugetlb test case
Since the head vmemmap page frame associated with each HugeTLB page is
reused, we should hide the PG_head flag of tail struct page from the
user. Add a tese case to check whether it is work properly. The test
steps are as follows.
1) alloc 2MB hugeTLB
2) get each page frame
3) apply those APIs in each page frame
4) Those APIs work completely the same as before.
Reading the flags of a page by /proc/kpageflags is done in
stable_page_flags(), which has invoked PageHead(), PageTail(),
PageCompound() and compound_head().
If those APIs work properly, the head page must have 15 and 17 bits set.
And tail pages must have 16 and 17 bits set but 15 bit unset. Those
flags are checked in check_page_flags().
Link: https://lkml.kernel.org/r/20211101031651.75851-5-songmuchun@bytedance.com Signed-off-by: Muchun Song <songmuchun@bytedance.com> Reviewed-by: Barry Song <song.bao.hua@hisilicon.com> Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com> Cc: Chen Huang <chenhuang5@huawei.com> Cc: David Hildenbrand <david@redhat.com> Cc: Fam Zheng <fam.zheng@bytedance.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Qi Zheng <zhengqi.arch@bytedance.com> Cc: Xiongchun Duan <duanxiongchun@bytedance.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Muchun Song [Tue, 22 Mar 2022 21:45:06 +0000 (14:45 -0700)]
mm: sparsemem: use page table lock to protect kernel pmd operations
The init_mm.page_table_lock is used to protect kernel page tables, we
can use it to serialize splitting vmemmap PMD mappings instead of mmap
write lock, which can increase the concurrency of vmemmap_remap_free().
Actually, It increase the concurrency between allocations of HugeTLB
pages. But it is not the only benefit. There are a lot of users of
mmap read lock of init_mm. The mmap write lock is holding through
vmemmap_remap_free(), removing mmap write lock usage to make it does not
affect other users of mmap read lock. It is not making anything worse
and always a win to move.
Now the kernel page table walker does not hold the page_table_lock when
walking pmd entries. There may be consistency issue of a pmd entry,
because pmd entry might change from a huge pmd entry to a PTE page
table. There is only one user of kernel page table walker, namely
ptdump. The ptdump already considers the consistency, which use a local
variable to cache the value of pmd entry. But we also need to update
->action to ACTION_CONTINUE to make sure the walker does not walk every
pte entry again when concurrent thread has split the huge pmd.
Link: https://lkml.kernel.org/r/20211101031651.75851-4-songmuchun@bytedance.com Signed-off-by: Muchun Song <songmuchun@bytedance.com> Cc: Barry Song <song.bao.hua@hisilicon.com> Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com> Cc: Chen Huang <chenhuang5@huawei.com> Cc: David Hildenbrand <david@redhat.com> Cc: Fam Zheng <fam.zheng@bytedance.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Qi Zheng <zhengqi.arch@bytedance.com> Cc: Xiongchun Duan <duanxiongchun@bytedance.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Muchun Song [Tue, 22 Mar 2022 21:45:03 +0000 (14:45 -0700)]
mm: hugetlb: replace hugetlb_free_vmemmap_enabled with a static_key
The page_fixed_fake_head() is used throughout memory management and the
conditional check requires checking a global variable, although the
overhead of this check may be small, it increases when the memory cache
comes under pressure. Also, the global variable will not be modified
after system boot, so it is very appropriate to use static key machanism.
Link: https://lkml.kernel.org/r/20211101031651.75851-3-songmuchun@bytedance.com Signed-off-by: Muchun Song <songmuchun@bytedance.com> Reviewed-by: Barry Song <song.bao.hua@hisilicon.com> Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com> Cc: Chen Huang <chenhuang5@huawei.com> Cc: David Hildenbrand <david@redhat.com> Cc: Fam Zheng <fam.zheng@bytedance.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Qi Zheng <zhengqi.arch@bytedance.com> Cc: Xiongchun Duan <duanxiongchun@bytedance.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Muchun Song [Tue, 22 Mar 2022 21:45:00 +0000 (14:45 -0700)]
mm: hugetlb: free the 2nd vmemmap page associated with each HugeTLB page
Patch series "Free the 2nd vmemmap page associated with each HugeTLB
page", v7.
This series can minimize the overhead of struct page for 2MB HugeTLB
pages significantly. It further reduces the overhead of struct page by
12.5% for a 2MB HugeTLB compared to the previous approach, which means
2GB per 1TB HugeTLB. It is a nice gain. Comments and reviews are
welcome. Thanks.
The main implementation and details can refer to the commit log of patch
1. In this series, I have changed the following four helpers, the
following table shows the impact of the overhead of those helpers.
+------------------+-----------------------+
| APIs | head page | tail page |
+------------------+-----------+-----------+
| PageHead() | Y | N |
+------------------+-----------+-----------+
| PageTail() | Y | N |
+------------------+-----------+-----------+
| PageCompound() | N | N |
+------------------+-----------+-----------+
| compound_head() | Y | N |
+------------------+-----------+-----------+
Y: Overhead is increased.
N: Overhead is _NOT_ increased.
It shows that the overhead of those helpers on a tail page don't change
between "hugetlb_free_vmemmap=on" and "hugetlb_free_vmemmap=off". But the
overhead on a head page will be increased when "hugetlb_free_vmemmap=on"
(except PageCompound()). So I believe that Matthew Wilcox's folio series
will help with this.
The users of PageHead() and PageTail() are much less than compound_head()
and most users of PageTail() are VM_BUG_ON(), so I have done some tests
about the overhead of compound_head() on head pages.
I have tested the overhead of calling compound_head() on a head page,
which is 2.11ns (Measure the call time of 10 million times
compound_head(), and then average).
For a head page whose address is not aligned with PAGE_SIZE or a
non-compound page, the overhead of compound_head() is 2.54ns which is
increased by 20%. For a head page whose address is aligned with
PAGE_SIZE, the overhead of compound_head() is 2.97ns which is increased by
40%. Most pages are the former. I do not think the overhead is
significant since the overhead of compound_head() itself is low.
This patch (of 5):
This patch minimizes the overhead of struct page for 2MB HugeTLB pages
significantly. It further reduces the overhead of struct page by 12.5%
for a 2MB HugeTLB compared to the previous approach, which means 2GB per
1TB HugeTLB (2MB type).
After the feature of "Free sonme vmemmap pages of HugeTLB page" is
enabled, the mapping of the vmemmap addresses associated with a 2MB
HugeTLB page becomes the figure below.
As we can see, the 2nd vmemmap page frame (indexed by 1) is reused and
remaped. However, the 2nd vmemmap page frame is also can be freed to
the buddy allocator, then we can change the mapping from the figure
above to the figure below.
After we do this, all tail vmemmap pages (1-7) are mapped to the head
vmemmap page frame (0). In other words, there are more than one page
struct with PG_head associated with each HugeTLB page. We __know__ that
there is only one head page struct, the tail page structs with PG_head are
fake head page structs. We need an approach to distinguish between those
two different types of page structs so that compound_head(), PageHead()
and PageTail() can work properly if the parameter is the tail page struct
but with PG_head.
The following code snippet describes how to distinguish between real and
fake head page struct.
if (test_bit(PG_head, &page->flags)) {
unsigned long head = READ_ONCE(page[1].compound_head);
if (head & 1) {
if (head == (unsigned long)page + 1)
==> head page struct
else
==> tail page struct
} else
==> head page struct
}
We can safely access the field of the @page[1] with PG_head because the
@page is a compound page composed with at least two contiguous pages.
[songmuchun@bytedance.com: restore lost comment changes]
Link: https://lkml.kernel.org/r/20211101031651.75851-1-songmuchun@bytedance.com Link: https://lkml.kernel.org/r/20211101031651.75851-2-songmuchun@bytedance.com Signed-off-by: Muchun Song <songmuchun@bytedance.com> Reviewed-by: Barry Song <song.bao.hua@hisilicon.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: Oscar Salvador <osalvador@suse.de> Cc: Michal Hocko <mhocko@suse.com> Cc: David Hildenbrand <david@redhat.com> Cc: Chen Huang <chenhuang5@huawei.com> Cc: Bodeddula Balasubramaniam <bodeddub@amazon.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Xiongchun Duan <duanxiongchun@bytedance.com> Cc: Fam Zheng <fam.zheng@bytedance.com> Cc: Qi Zheng <zhengqi.arch@bytedance.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / kernel.git / log