This patch introduces DAMOS_MIGRATE_HOT action, which is similar to
DAMOS_MIGRATE_COLD, but proritizes hot pages.
It migrates pages inside the given region to the 'target_nid' NUMA node
in the sysfs.
Here is one of the example usage of this 'migrate_hot' action.
$ cd /sys/kernel/mm/damon/admin/kdamonds/<N>
$ cat contexts/<N>/schemes/<N>/action
migrate_hot
$ echo 0 > contexts/<N>/schemes/<N>/target_nid
$ echo commit > state
$ numactl -p 2 ./hot_cold 500M 600M &
$ numastat -c -p hot_cold
Per-node process memory usage (in MBs)
PID Node 0 Node 1 Node 2 Total
-------------- ------ ------ ------ -----
701 (hot_cold) 501 0 601 1101
Link: https://lkml.kernel.org/r/20240614030010.751-7-honggyu.kim@sk.com
Signed-off-by: Hyeongtak Ji <hyeongtak.ji@sk.com>
Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
Signed-off-by: SeongJae Park <sj@kernel.org>
Cc: Gregory Price <gregory.price@memverge.com>
Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Cc: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This patch introduces DAMOS_MIGRATE_COLD action, which is similar to
DAMOS_PAGEOUT, but migrate folios to the given 'target_nid' in the sysfs
instead of swapping them out.
The 'target_nid' sysfs knob informs the migration target node ID.
Here is one of the example usage of this 'migrate_cold' action.
$ cd /sys/kernel/mm/damon/admin/kdamonds/<N>
$ cat contexts/<N>/schemes/<N>/action
migrate_cold
$ echo 2 > contexts/<N>/schemes/<N>/target_nid
$ echo commit > state
$ numactl -p 0 ./hot_cold 500M 600M &
$ numastat -c -p hot_cold
Per-node process memory usage (in MBs)
PID Node 0 Node 1 Node 2 Total
-------------- ------ ------ ------ -----
701 (hot_cold) 501 0 601 1101
Since there are some common routines with pageout, many functions have
similar logics between pageout and migrate cold.
damon_pa_migrate_folio_list() is a minimized version of
shrink_folio_list().
Link: https://lkml.kernel.org/r/20240614030010.751-6-honggyu.kim@sk.com
Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
Signed-off-by: Hyeongtak Ji <hyeongtak.ji@sk.com>
Signed-off-by: SeongJae Park <sj@kernel.org>
Cc: Gregory Price <gregory.price@memverge.com>
Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Cc: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "DAMON based tiered memory management for CXL memory", v6.
Introduction
============
With the advent of CXL/PCIe attached DRAM, which will be called simply as
CXL memory in this cover letter, some systems are becoming more
heterogeneous having memory systems with different latency and bandwidth
characteristics. They are usually handled as different NUMA nodes in
separate memory tiers and CXL memory is used as slow tiers because of its
protocol overhead compared to local DRAM.
In this kind of systems, we need to be careful placing memory pages on
proper NUMA nodes based on the memory access frequency. Otherwise, some
frequently accessed pages might reside on slow tiers and it makes
performance degradation unexpectedly. Moreover, the memory access
patterns can be changed at runtime.
To handle this problem, we need a way to monitor the memory access
patterns and migrate pages based on their access temperature. The
DAMON(Data Access MONitor) framework and its DAMOS(DAMON-based Operation
Schemes) can be useful features for monitoring and migrating pages. DAMOS
provides multiple actions based on DAMON monitoring results and it can be
used for proactive reclaim, which means swapping cold pages out with
DAMOS_PAGEOUT action, but it doesn't support migration actions such as
demotion and promotion between tiered memory nodes.
This series supports two new DAMOS actions; DAMOS_MIGRATE_HOT for
promotion from slow tiers and DAMOS_MIGRATE_COLD for demotion from fast
tiers. This prevents hot pages from being stuck on slow tiers, which
makes performance degradation and cold pages can be proactively demoted to
slow tiers so that the system can increase the chance to allocate more hot
pages to fast tiers.
The DAMON provides various tuning knobs but we found that the proactive
demotion for cold pages is especially useful when the system is running
out of memory on its fast tier nodes.
Our evaluation result shows that it reduces the performance slowdown
compared to the default memory policy from 11% to 3~5% when the system
runs under high memory pressure on its fast tier DRAM nodes.
DAMON configuration
===================
The specific DAMON configuration doesn't have to be in the scope of this
patch series, but some rough idea is better to be shared to explain the
evaluation result.
The DAMON provides many knobs for fine tuning but its configuration file
is generated by HMSDK[3]. It includes gen_config.py script that generates
a json file with the full config of DAMON knobs and it creates multiple
kdamonds for each NUMA node when the DAMON is enabled so that it can run
hot/cold based migration for tiered memory.
Evaluation Workload
===================
The performance evaluation is done with redis[4], which is a widely used
in-memory database and the memory access patterns are generated via
YCSB[5]. We have measured two different workloads with zipfian and latest
distributions but their configs are slightly modified to make memory usage
higher and execution time longer for better evaluation.
The idea of evaluation using these migrate_{hot,cold} actions covers
system-wide memory management rather than partitioning hot/cold pages of a
single workload. The default memory allocation policy creates pages to
the fast tier DRAM node first, then allocates newly created pages to the
slow tier CXL node when the DRAM node has insufficient free space. Once
the page allocation is done then those pages never move between NUMA
nodes. It's not true when using numa balancing, but it is not the scope
of this DAMON based tiered memory management support.
If the working set of redis can be fit fully into the DRAM node, then the
redis will access the fast DRAM only. Since the performance of DRAM only
is faster than partially accessing CXL memory in slow tiers, this
environment is not useful to evaluate this patch series.
To make pages of redis be distributed across fast DRAM node and slow CXL
node to evaluate our migrate_{hot,cold} actions, we pre-allocate some cold
memory externally using mmap and memset before launching redis-server. We
assumed that there are enough amount of cold memory in datacenters as
TMO[6] and TPP[7] papers mentioned.
The evaluation sequence is as follows.
1. Turn on DAMON with DAMOS_MIGRATE_COLD action for DRAM node and
DAMOS_MIGRATE_HOT action for CXL node. It demotes cold pages on DRAM
node and promotes hot pages on CXL node in a regular interval.
2. Allocate a huge block of cold memory by calling mmap and memset at
the fast tier DRAM node, then make the process sleep to make the fast
tier has insufficient space for redis-server.
3. Launch redis-server and load prebaked snapshot image, dump.rdb. The
redis-server consumes 52GB of anon pages and 33GB of file pages, but
due to the cold memory allocated at 2, it fails allocating the entire
memory of redis-server on the fast tier DRAM node so it partially
allocates the remaining on the slow tier CXL node. The ratio of
DRAM:CXL depends on the size of the pre-allocated cold memory.
4. Run YCSB to make zipfian or latest distribution of memory accesses to
redis-server, then measure its execution time when it's completed.
5. Repeat 4 over 50 times to measure the average execution time for each
run.
6. Increase the cold memory size then repeat goes to 2.
For each test at 4 took about a minute so repeating it 50 times almost
took about 1 hour for each test with a specific cold memory from 440GB to
500GB in 10GB increments for each evaluation. So it took about more than
10 hours for both zipfian and latest workloads to get the entire
evaluation results. Repeating the same test set multiple times doesn't
show much difference so I think it might be enough to make the result
reliable.
Evaluation Results
==================
All the result values are normalized to DRAM-only execution time because
the workload cannot be faster than DRAM-only unless the workload hits the
peak bandwidth but our redis test doesn't go beyond the bandwidth limit.
So the DRAM-only execution time is the ideal result without affected by
the gap between DRAM and CXL performance difference. The NUMA node
environment is as follows.
node0 - local DRAM, 512GB with a CPU socket (fast tier)
node1 - disabled
node2 - CXL DRAM, 96GB, no CPU attached (slow tier)
The following is the result of generating zipfian distribution to
redis-server and the numbers are averaged by 50 times of execution.
1. YCSB zipfian distribution read only workload
memory pressure with cold memory on node0 with 512GB of local DRAM.
====================+================================================+=========
| cold memory occupied by mmap and memset |
| 0G 440G 450G 460G 470G 480G 490G 500G |
====================+================================================+=========
Execution time normalized to DRAM-only values | GEOMEAN
--------------------+------------------------------------------------+---------
DRAM-only | 1.00 - - - - - - - | 1.00
CXL-only | 1.19 - - - - - - - | 1.19
default | - 1.00 1.05 1.08 1.12 1.14 1.18 1.18 | 1.11
DAMON tiered | - 1.03 1.03 1.03 1.03 1.03 1.07 *1.05 | 1.04
DAMON lazy | - 1.04 1.03 1.04 1.05 1.06 1.06 *1.06 | 1.05
====================+================================================+=========
CXL usage of redis-server in GB | AVERAGE
--------------------+------------------------------------------------+---------
DRAM-only | 0.0 - - - - - - - | 0.0
CXL-only | 51.4 - - - - - - - | 51.4
default | - 0.6 10.6 20.5 30.5 40.5 47.6 50.4 | 28.7
DAMON tiered | - 0.6 0.5 0.4 0.7 0.8 7.1 5.6 | 2.2
DAMON lazy | - 0.5 3.0 4.5 5.4 6.4 9.4 9.1 | 5.5
====================+================================================+=========
Each test result is based on the execution environment as follows.
DRAM-only: redis-server uses only local DRAM memory.
CXL-only: redis-server uses only CXL memory.
default: default memory policy(MPOL_DEFAULT).
numa balancing disabled.
DAMON tiered: DAMON enabled with DAMOS_MIGRATE_COLD for DRAM
nodes and DAMOS_MIGRATE_HOT for CXL nodes.
DAMON lazy: same as DAMON tiered, but turn on DAMON just
before making memory access request via YCSB.
The above result shows the "default" execution time goes up as the size of
cold memory is increased from 440G to 500G because the more cold memory
used, the more CXL memory is used for the target redis workload and this
makes the execution time increase.
However, "DAMON tiered" and other DAMON results show less slowdown because
the DAMOS_MIGRATE_COLD action at DRAM node proactively demotes
pre-allocated cold memory to CXL node and this free space at DRAM
increases more chance to allocate hot or warm pages of redis-server to
fast DRAM node. Moreover, DAMOS_MIGRATE_HOT action at CXL node also
promotes hot pages of redis-server to DRAM node actively.
As a result, it makes more memory of redis-server stay in DRAM node
compared to "default" memory policy and this makes the performance
improvement.
Please note that the result numbers of "DAMON tiered" and "DAMON lazy" at
500G are marked with * stars, which means their test results are replaced
with reproduced tests that didn't have OOM issue.
That was needed because sometimes the test processes get OOM when DRAM has
insufficient space. The DAMOS_MIGRATE_HOT doesn't kick reclaim but just
gives up migration when there is not enough space at DRAM side. The
problem happens when there is competition between normal allocation and
migration and the migration is done before normal allocation, then the
completely unrelated normal allocation can trigger reclaim, which incurs
OOM.
Because of this issue, I have also tested more cases with
"demotion_enabled" flag enabled to make such reclaim doesn't trigger OOM,
but just demote reclaimed pages. The following test results show more
tests with "kswapd" marked.
2. YCSB zipfian distribution read only workload (with demotion_enabled true)
memory pressure with cold memory on node0 with 512GB of local DRAM.
====================+================================================+=========
| cold memory occupied by mmap and memset |
| 0G 440G 450G 460G 470G 480G 490G 500G |
====================+================================================+=========
Execution time normalized to DRAM-only values | GEOMEAN
--------------------+------------------------------------------------+---------
DAMON tiered | - 1.03 1.03 1.03 1.03 1.03 1.07 1.05 | 1.04
DAMON lazy | - 1.04 1.03 1.04 1.05 1.06 1.06 1.06 | 1.05
DAMON tiered kswapd | - 1.03 1.03 1.03 1.03 1.02 1.02 1.03 | 1.03
DAMON lazy kswapd | - 1.04 1.04 1.04 1.03 1.05 1.04 1.05 | 1.04
====================+================================================+=========
CXL usage of redis-server in GB | AVERAGE
--------------------+------------------------------------------------+---------
DAMON tiered | - 0.6 0.5 0.4 0.7 0.8 7.1 5.6 | 2.2
DAMON lazy | - 0.5 3.0 4.5 5.4 6.4 9.4 9.1 | 5.5
DAMON tiered kswapd | - 0.0 0.0 0.4 0.5 0.1 0.8 1.0 | 0.4
DAMON lazy kswapd | - 4.2 4.6 5.3 1.7 6.8 8.1 5.8 | 5.2
====================+================================================+=========
Each test result is based on the exeuction environment as follows.
DAMON tiered: same as before
DAMON lazy: same as before
DAMON tiered kswapd: same as DAMON tiered, but turn on
/sys/kernel/mm/numa/demotion_enabled to make
kswapd or direct reclaim does demotion.
DAMON lazy kswapd: same as DAMON lazy, but turn on
/sys/kernel/mm/numa/demotion_enabled to make
kswapd or direct reclaim does demotion.
The "DAMON tiered kswapd" and "DAMON lazy kswapd" didn't trigger OOM at
all unlike other tests because kswapd and direct reclaim from DRAM node
can demote reclaimed pages to CXL node independently from DAMON actions
and their results are slightly better than without having
"demotion_enabled".
In summary, the evaluation results show that DAMON memory management with
DAMOS_MIGRATE_{HOT,COLD} actions reduces the performance slowdown compared
to the "default" memory policy from 11% to 3~5% when the system runs with
high memory pressure on its fast tier DRAM nodes.
Having these DAMOS_MIGRATE_HOT and DAMOS_MIGRATE_COLD actions can make
tiered memory systems run more efficiently under high memory pressures.
This patch (of 7):
The alloc_demote_folio can be used out of vmscan.c so it'd be better to
remove static keyword from it.
Link: https://lkml.kernel.org/r/20240614030010.751-1-honggyu.kim@sk.com
Link: https://lkml.kernel.org/r/20240614030010.751-2-honggyu.kim@sk.com
Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Signed-off-by: SeongJae Park <sj@kernel.org>
Cc: Gregory Price <gregory.price@memverge.com>
Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Cc: Hyeongtak Ji <hyeongtak.ji@sk.com>
Cc: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Unlike other cgroup subsystems, the hugetlb cgroup does not provide a
static array of cftype that explicitly displays the properties, handling
functions, etc., of each file. Instead, it dynamically creates the
attribute of cftypes based on the hstate during the startup procedure.
This reduces the readability of the code.
To fix this issue, introduce two templates of cftypes, and rebuild the
attributes according to the hstate to make it ready to be added to cgroup
framework.
Link: https://lkml.kernel.org/r/20240612092409.2027592-3-xiujianfeng@huawei.com
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: kernel test robot <oliver.sang@intel.com>
From: Xiu Jianfeng <xiujianfeng@huawei.com>
Subject: mm/hugetlb_cgroup: register lockdep key for cftype
Date: Tue, 18 Jun 2024 07:19:22 +0000
When CONFIG_DEBUG_LOCK_ALLOC is enabled, the following commands can
trigger a bug,
mount -t cgroup2 none /sys/fs/cgroup
cd /sys/fs/cgroup
echo "+hugetlb" > cgroup.subtree_control
The log is as below:
BUG: key ffff8880046d88d8 has not been registered!
------------[ cut here ]------------
DEBUG_LOCKS_WARN_ON(1)
WARNING: CPU: 3 PID: 226 at kernel/locking/lockdep.c:4945 lockdep_init_map_type+0x185/0x220
Modules linked in:
CPU: 3 PID: 226 Comm: bash Not tainted 6.10.0-rc4-next-20240617-g76db4c64526c #544
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
RIP: 0010:lockdep_init_map_type+0x185/0x220
Code: 00 85 c0 0f 84 6c ff ff ff 8b 3d 6a d1 85 01 85 ff 0f 85 5e ff ff ff 48 c7 c6 21 99 4a 82 48 c7 c7 60 29 49 82 e8 3b 2e f5
RSP: 0018:ffffc9000083fc30 EFLAGS: 00000282
RAX: 0000000000000000 RBX: ffffffff828dd820 RCX: 0000000000000027
RDX: ffff88803cd9cac8 RSI: 0000000000000001 RDI: ffff88803cd9cac0
RBP: ffff88800674fbb0 R08: ffffffff828ce248 R09: 00000000ffffefff
R10: ffffffff8285e260 R11: ffffffff828b8eb8 R12: ffff8880046d88d8
R13: 0000000000000000 R14: 0000000000000000 R15: ffff8880067281c0
FS: 00007f68601ea740(0000) GS:ffff88803cd80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005614f3ebc740 CR3: 000000000773a000 CR4: 00000000000006f0
Call Trace:
<TASK>
? __warn+0x77/0xd0
? lockdep_init_map_type+0x185/0x220
? report_bug+0x189/0x1a0
? handle_bug+0x3c/0x70
? exc_invalid_op+0x18/0x70
? asm_exc_invalid_op+0x1a/0x20
? lockdep_init_map_type+0x185/0x220
__kernfs_create_file+0x79/0x100
cgroup_addrm_files+0x163/0x380
? find_held_lock+0x2b/0x80
? find_held_lock+0x2b/0x80
? find_held_lock+0x2b/0x80
css_populate_dir+0x73/0x180
cgroup_apply_control_enable+0x12f/0x3a0
cgroup_subtree_control_write+0x30b/0x440
kernfs_fop_write_iter+0x13a/0x1f0
vfs_write+0x341/0x450
ksys_write+0x64/0xe0
do_syscall_64+0x4b/0x110
entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7f68602d9833
Code: 8b 15 61 26 0e 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 64 8b 04 25 18 00 00 00 85 c0 75 14 b8 01 00 00 00 08
RSP: 002b:00007fff9bbdf8e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 0000000000000009 RCX: 00007f68602d9833
RDX: 0000000000000009 RSI: 00005614f3ebc740 RDI: 0000000000000001
RBP: 00005614f3ebc740 R08: 000000000000000a R09: 0000000000000008
R10: 00005614f3db6ba0 R11: 0000000000000246 R12: 0000000000000009
R13: 00007f68603bd6a0 R14: 0000000000000009 R15: 00007f68603b8880
For lockdep, there is a sanity check in lockdep_init_map_type(), the
lock-class key must either have been allocated statically or must
have been registered as a dynamic key. However the commit e18df2889ff9
("mm/hugetlb_cgroup: prepare cftypes based on template") has changed
the cftypes from static allocated objects to dynamic allocated objects,
so the cft->lockdep_key must be registered proactively.
[xiujianfeng@huawei.com: fix BUG()]
Link: https://lkml.kernel.org/r/20240619015527.2212698-1-xiujianfeng@huawei.com
Link: https://lkml.kernel.org/r/20240618071922.2127289-1-xiujianfeng@huawei.com
Link: https://lore.kernel.org/all/602186b3-5ce3-41b3-90a3-134792cc2a48@samsung.com/
Fixes: e18df2889ff9 ("mm/hugetlb_cgroup: prepare cftypes based on template")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202406181046.8d8b2492-oliver.sang@intel.com
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Tested-by: SeongJae Park <sj@kernel.org>
Closes: https://lore.kernel.org/20240618233608.400367-1-sj@kernel.org
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Today, we do not have any observability of per-page metadata and how much
it takes away from the machine capacity. Thus, we want to describe the
amount of memory that is going towards per-page metadata, which can vary
depending on build configuration, machine architecture, and system use.
This patch adds 2 fields to /proc/vmstat that can used as shown below:
Accounting per-page metadata allocated by boot-allocator:
/proc/vmstat:nr_memmap_boot * PAGE_SIZE
Accounting per-page metadata allocated by buddy-allocator:
/proc/vmstat:nr_memmap * PAGE_SIZE
Accounting total Perpage metadata allocated on the machine:
(/proc/vmstat:nr_memmap_boot +
/proc/vmstat:nr_memmap) * PAGE_SIZE
Utility for userspace:
Observability: Describe the amount of memory overhead that is going to
per-page metadata on the system at any given time since this overhead is
not currently observable.
Debugging: Tracking the changes or absolute value in struct pages can help
detect anomalies as they can be correlated with other metrics in the
machine (e.g., memtotal, number of huge pages, etc).
page_ext overheads: Some kernel features such as page_owner
page_table_check that use page_ext can be optionally enabled via kernel
parameters. Having the total per-page metadata information helps users
precisely measure impact. Furthermore, page-metadata metrics will reflect
the amount of struct pages reliquished (or overhead reduced) when
hugetlbfs pages are reserved which will vary depending on whether hugetlb
vmemmap optimization is enabled or not.
For background and results see:
lore.kernel.org/all/20240220214558.3377482-1-souravpanda@google.com
Link: https://lkml.kernel.org/r/20240605222751.1406125-1-souravpanda@google.com
Signed-off-by: Sourav Panda <souravpanda@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Chen Linxuan <chenlinxuan@uniontech.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Ivan Babrou <ivan@cloudflare.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mike Rapoport (IBM) <rppt@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Tomas Mudrunka <tomas.mudrunka@gmail.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Xu <weixugc@google.com>
Cc: Yang Yang <yang.yang29@zte.com.cn>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
thuge-gen.c defines SHM_HUGE_* macros that are provided by the uapi since
4.14. These macros get redefined when compiling with Android's bionic
because its sys/shm.h will import the uapi definitions.
However if linux/shm.h is included, with glibc, sys/shm.h will clash on
some struct definitions:
/usr/include/linux/shm.h:26:8: error: redefinition of ‘struct shmid_ds’
26 | struct shmid_ds {
| ^~~~~~~~
In file included from /usr/include/x86_64-linux-gnu/bits/shm.h:45,
from /usr/include/x86_64-linux-gnu/sys/shm.h:30:
/usr/include/x86_64-linux-gnu/bits/types/struct_shmid_ds.h:24:8: note: originally defined here
24 | struct shmid_ds
| ^~~~~~~~
For now, guard the SHM_HUGE_* defines with ifndef to prevent redefinition
warnings on Android bionic.
Link: https://lkml.kernel.org/r/20240605223637.1374969-3-edliaw@google.com
Signed-off-by: Edward Liaw <edliaw@google.com>
Reviewed-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Bill Wendling <morbo@google.com>
Cc: Carlos Llamas <cmllamas@google.com>
Cc: Justin Stitt <justinstitt@google.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Zswap does not support storing or loading large folios. Until proper
support is added, attempts to load large folios from zswap are a bug.
For example, if a swapin fault observes that contiguous PTEs are pointing
to contiguous swap entries and tries to swap them in as a large folio,
swap_read_folio() will pass in a large folio to zswap_load(), but
zswap_load() will only effectively load the first page in the folio. If
the first page is not in zswap, the folio will be read from disk, even
though other pages may be in zswap.
In both cases, this will lead to silent data corruption. Proper support
needs to be added before large folio swapins and zswap can work together.
Looking at callers of swap_read_folio(), it seems like they are either
allocated from __read_swap_cache_async() or do_swap_page() in the
SWP_SYNCHRONOUS_IO path. Both of which allocate order-0 folios, so
everything is fine for now.
However, there is ongoing work to add to support large folio swapins [1].
To make sure new development does not break zswap (or get broken by
zswap), add minimal handling of incorrect loads of large folios to zswap.
First, move the call folio_mark_uptodate() inside zswap_load().
If a large folio load is attempted, and zswap was ever enabled on the
system, return 'true' without calling folio_mark_uptodate(). This will
prevent the folio from being read from disk, and will emit an IO error
because the folio is not uptodate (e.g. do_swap_fault() will return
VM_FAULT_SIGBUS). It may not be reliable recovery in all cases, but it is
better than nothing.
This was tested by hacking the allocation in __read_swap_cache_async() to
use order 2 and __GFP_COMP.
In the future, to handle this correctly, the swapin code should:
(a) Fall back to order-0 swapins if zswap was ever used on the
machine, because compressed pages remain in zswap after it is
disabled.
(b) Add proper support to swapin large folios from zswap (fully or
partially).
Probably start with (a) then followup with (b).
[1]https://lore.kernel.org/linux-mm/20240304081348.197341-6-21cnbao@gmail.com/
Link: https://lkml.kernel.org/r/20240611024516.1375191-3-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Acked-by: Barry Song <baohua@kernel.org>
Cc: Barry Song <baohua@kernel.org>
Cc: Chengming Zhou <chengming.zhou@linux.dev>
Cc: Chris Li <chrisl@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Nhat Pham <nphamcs@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Add zswap_never_enabled() to skip the xarray lookup in zswap_load() if
zswap was never enabled on the system. It is implemented using static
branches for efficiency, as enabling zswap should be a rare event. This
could shave some cycles off zswap_load() when CONFIG_ZSWAP is used but
zswap is never enabled.
However, the real motivation behind this patch is two-fold:
- Incoming large folio swapin work will need to fallback to order-0
folios if zswap was ever enabled, because any part of the folio could be
in zswap, until proper handling of large folios with zswap is added.
- A warning and recovery attempt will be added in a following change in
case the above was not done incorrectly. Zswap will fail the read if
the folio is large and it was ever enabled.
Expose zswap_never_enabled() in the header for the swapin work to use
it later.
[yosryahmed@google.com: expose zswap_never_enabled() in the header]
Link: https://lkml.kernel.org/r/Zmjf0Dr8s9xSW41X@google.com
Link: https://lkml.kernel.org/r/20240611024516.1375191-2-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Reviewed-by: Nhat Pham <nphamcs@gmail.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Chengming Zhou <chengming.zhou@linux.dev>
Cc: Chris Li <chrisl@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Current call flow looks like this:
start_kernel
mm_core_init
mem_init
mem_init_print_info
rest_init
kernel_init
kernel_init_freeable
page_alloc_init_late
deferred_init_memmap
If CONFIG_DEFERRED_STRUCT_PAGE_INIT, the time mem_init_print_info()
calls, pages are not totally initialized and freed to buddy.
This has one issue
* nr_free_pages() just contains partial free pages in the system,
which is not we expect.
Let's print the mem info after defer_init is done.
Also this would help changing totalram_pages accounting, since we plan
to move the accounting into __free_pages_core().
Link: https://lkml.kernel.org/r/20240611145223.16872-1-richard.weiyang@gmail.com
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Mike Rapoport (IBM) <rppt@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "Reclaim lazyfree THP without splitting", v8.
This series adds support for reclaiming PMD-mapped THP marked as lazyfree
without needing to first split the large folio via
split_huge_pmd_address().
When the user no longer requires the pages, they would use
madvise(MADV_FREE) to mark the pages as lazy free. Subsequently, they
typically would not re-write to that memory again.
During memory reclaim, if we detect that the large folio and its PMD are
both still marked as clean and there are no unexpected references(such as
GUP), so we can just discard the memory lazily, improving the efficiency
of memory reclamation in this case.
Performance Testing
===================
On an Intel i5 CPU, reclaiming 1GiB of lazyfree THPs using
mem_cgroup_force_empty() results in the following runtimes in seconds
(shorter is better):
--------------------------------------------
| Old | New | Change |
--------------------------------------------
| 0.683426 | 0.049197 | -92.80% |
--------------------------------------------
This patch (of 8):
Introduce the labels walk_done and walk_abort as exit points to eliminate
duplicated exit code in the pagewalk loop.
Link: https://lkml.kernel.org/r/20240614015138.31461-1-ioworker0@gmail.com
Link: https://lkml.kernel.org/r/20240614015138.31461-2-ioworker0@gmail.com
Signed-off-by: Lance Yang <ioworker0@gmail.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Barry Song <baohua@kernel.org>
Cc: Bang Li <libang.li@antgroup.com>
Cc: Fangrui Song <maskray@google.com>
Cc: Jeff Xie <xiehuan09@gmail.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: SeongJae Park <sj@kernel.org>
Cc: Yang Shi <shy828301@gmail.com>
Cc: Yin Fengwei <fengwei.yin@intel.com>
Cc: Zach O'Keefe <zokeefe@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Most of the work done in folio_start_writeback is reversed in
folio_end_writeback. For e.g. NR_WRITEBACK and NR_ZONE_WRITE_PENDING are
incremented in start_writeback and decremented in end_writeback. Calling
end_writeback immediately after start_writeback (separated by
folio_unlock) cancels the affect of most of the work done in start hence
can be removed.
There is some extra work done in folio_end_writeback, however it is
incorrect/not applicable to zswap:
- folio_end_writeback incorrectly increments NR_WRITTEN counter,
eventhough the pages aren't written to disk, hence this change
corrects this behaviour.
- folio_end_writeback calls folio_rotate_reclaimable, but that only
makes sense for async writeback pages, while for zswap pages are
synchronously reclaimed.
Link: https://lkml.kernel.org/r/20240612100109.1616626-1-usamaarif642@gmail.com
Link: https://lkml.kernel.org/r/20240610143037.812955-1-usamaarif642@gmail.com
Signed-off-by: Usama Arif <usamaarif642@gmail.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Yosry Ahmed <yosryahmed@google.com>
Reviewed-by: Chengming Zhou <chengming.zhou@linux.dev>
Suggested-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Nhat Pham <nphamcs@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
create_pagecache_thp_and_fd() in split_huge_page_test.c used the variable
dummy to perform mmap read.
However, this test was skipped even on XFS which has large folio support.
The issue was compiler (gcc 13.2.0) was optimizing out the dummy variable,
therefore, not creating huge page in the page cache.
Use asm volatile() trick to force the compiler not to optimize out the
loop where we read from the mmaped addr. This is similar to what is being
done in other tests (cow.c, etc)
As the variable is now used in the asm statement, remove the unused
attribute.
Link: https://lkml.kernel.org/r/20240606203619.677276-1-kernel@pankajraghav.com
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Luis Chamberlain <mcgrof@kernel.org>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Pankaj Raghav <p.raghav@samsung.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "mm: introduce pmd|pte_needs_soft_dirty_wp helpers and
utilize them", v2.
This patchset introduces the pte_need_soft_dirty_wp and
pmd_need_soft_dirty_wp helpers to determine if write protection is
required for softdirty tracking. These helpers enhance code readability
and improve the overall appearance.
They are then utilized in gup, mprotect, swap, and other related
functions.
This patch (of 2):
This patch introduces the pte_needs_soft_dirty_wp and
pmd_needs_soft_dirty_wp helpers to determine if write protection is
required for softdirty tracking. This can enhance code readability and
improve its overall appearance. These new helpers are then utilized in
gup, huge_memory, and mprotect.
Link: https://lkml.kernel.org/r/20240607211358.4660-1-21cnbao@gmail.com
Link: https://lkml.kernel.org/r/20240607211358.4660-2-21cnbao@gmail.com
Signed-off-by: Barry Song <v-songbaohua@oppo.com>
Suggested-by: David Hildenbrand <david@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Chris Li <chrisl@kernel.org>
Cc: Kairui Song <kasong@tencent.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
On Ubuntu 23.04, on a clean git tree, the selftests/mm build fails due 10
or 20 missing items, all of which are found in fs.h, which is created via
"make headers". However, as per [1], the idea is to stop requiring "make
headers", and instead, take a snapshot of the files and check them in.
Here are a few of the build errors:
vm_util.c:34:21: error: variable has incomplete type 'struct pm_scan_arg'
struct pm_scan_arg arg;
...
vm_util.c:45:28: error: use of undeclared identifier 'PAGE_IS_WPALLOWED'
...
vm_util.c:55:21: error: variable has incomplete type 'struct page_region'
...
vm_util.c:105:20: error: use of undeclared identifier 'PAGE_IS_SOFT_DIRTY'
To fix this, add fs.h, taken from a snapshot of ./usr/include/linux/fs.h
after running "make headers".
[1] commit e076eaca59 ("selftests: break the dependency upon local
header files")
Link: https://lkml.kernel.org/r/20240618022422.804305-5-jhubbard@nvidia.com
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Andrei Vagin <avagin@google.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Jeff Xu <jeffxu@chromium.org>
Cc: Kees Cook <kees@kernel.org>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "cleanups, fixes, and progress towards avoiding "make
headers"", v3.
Eventually, once the build succeeds on a sufficiently old distro, the idea
is to delete $(KHDR_INCLUDES) from the selftests/mm build, and then after
that, from selftests/lib.mk and all of the other selftest builds.
For now, this series merely achieves a clean build of selftests/mm on a
not-so-old distro: Ubuntu 23.04. In other words, after this series is
applied, it is possible to delete $(KHDR_INCLUDES) from
selftests/mm/Makefile and the build will still succeed.
1. Add tools/uapi/asm/unistd_[32|x32|64].h files, which include
definitions of __NR_mseal, and include them (indirectly) from the files
that use __NR_mseal. The new files are copied from ./usr/include/asm,
which is how we have agreed to do this sort of thing, see [1].
2. Add fs.h, similarly created: it was copied directly from a snapshot
of ./usr/include/linux/fs.h after running "make headers".
3. Add a few selected prctl.h values that the ksm and mdwe tests require.
4. Factor out some common code from mseal_test.c and seal_elf.c, into
a new mseal_helpers.h file.
5. Remove local __NR_* definitions and checks.
[1] commit e076eaca59 ("selftests: break the dependency upon local
header files")
This patch (of 6):
The selftests/mm build isn't exactly "broken", according to the current
documentation, which still claims that one must run "make headers",
before building the kselftests. However, according to the new plan to
get rid of that requirement [1], they are future-broken: attempting to
build selftests/mm *without* first running "make headers" will fail due
to not finding __NR_mseal.
Therefore, include asm-generic/unistd.h, which has all of the system
call numbers that are needed, abstracted across the various CPU arches.
Some explanation in support of this "asm-generic" approach:
For most user space programs, the header file inclusion behaves as per
this microblaze example, which comes from David Hildenbrand (thanks!):
arch/microblaze/include/asm/unistd.h
-> #include <uapi/asm/unistd.h>
arch/microblaze/include/uapi/asm/unistd.h
-> #include <asm/unistd_32.h>
-> Generated during "make headers"
usr/include/asm/unistd_32.h is generated via
arch/microblaze/kernel/syscalls/Makefile with the syshdr command.
So we never end up including asm-generic/unistd.h directly on
microblaze... [2]
However, those programs are installed on a single computer that has a
single set of asm and kernel headers installed.
In contrast, the kselftests are quite special, because they must
provide a set of user space programs that:
a) Mostly avoid using the installed (distro) system header files.
b) Build (and run) on all supported CPU architectures
c) Occasionally use symbols that have so new that they have not
yet been included in the distro's header files.
Doing (a) creates a new problem: how to get a set of cross-platform
headers that works in all cases.
Fortunately, asm-generic headers solve that one. Which is why we need
to use them here--at least, for particularly difficult headers such as
unistd.h.
The reason this hasn't really come up yet, is that until now, the
kselftests requirement (which I'm trying to eventually remove) was that
"make headers" must first be run. That allowed the selftests to get a
snapshot of sufficiently new header files that looked just like (and
conflict with) the installed system headers.
And as an aside, this is also an improvement over past practices of
simply open-coding in a single (not per-arch) definition of a new
symbol, directly into the selftest code.
[1] commit e076eaca59 ("selftests: break the dependency upon local
header files")
[2] https://lore.kernel.org/all/0b152bea-ccb6-403e-9c57-08ed5e828135@redhat.com/
Link: https://lkml.kernel.org/r/20240618022422.804305-1-jhubbard@nvidia.com
Link: https://lkml.kernel.org/r/20240618022422.804305-2-jhubbard@nvidia.com
Fixes: 4926c7a52d ("selftest mm/mseal memory sealing")
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Jeff Xu <jeffxu@chromium.org>
Cc: Andrei Vagin <avagin@google.com>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Kees Cook <kees@kernel.org>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Commit [1] introduced IO polling support duding swapin to reduce swap read
latency for block devices that can be polled. However later commit [2]
removed polling support. Commit [3] removed the remnants of polling
support from read_swap_cache_async() and __read_swap_cache_async().
However, it left behind some remnants in swap_read_folio(), the
'synchronous' argument.
swap_read_folio() reads the folio synchronously if synchronous=true or if
SWP_SYNCHRONOUS_IO is set in swap_info_struct. The only caller that
passes synchronous=true is in do_swap_page() in the SWP_SYNCHRONOUS_IO
case.
Hence, the argument is redundant, it is only set to true when the swap
read would have been synchronous anyway. Remove it.
[1] Commit 23955622ff ("swap: add block io poll in swapin path")
[2] Commit 9650b453a3 ("block: ignore RWF_HIPRI hint for sync dio")
[3] Commit b243dcbf2f ("swap: remove remnants of polling from read_swap_cache_async")
Link: https://lkml.kernel.org/r/20240607045515.1836558-1-yosryahmed@google.com
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
It looks rather weird that totalhigh_pages() returns an "unsigned long"
but nr_free_highpages() returns an "unsigned int".
Let's return an "unsigned long" from nr_free_highpages() to be consistent.
While at it, use a plain "0" instead of a "0UL" in the !CONFIG_HIGHMEM
totalhigh_pages() implementation, to make these look alike as well.
Link: https://lkml.kernel.org/r/20240607083711.62833-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Patch series "mm/highmem: don't track highmem pages manually".
Let's remove highmem special-casing from adjust_managed_page_count(), to
result in less confusion why memblock manually adjusts totalram_pages, and
__free_pages_core() only adjusts the zone's managed pages -- what about
the highmem pages that adjust_managed_page_count() updates?
Now, we only maintain totalram_pages and a zone's managed pages
independent of highmem support. We can derive the number of highmem pages
simply by looking at the relevant zone's managed pages. I don't think
there is any particular fast path that needs a maximum-efficient
totalhigh_pages() implementation.
Note that highmem memory is currently initialized using
free_highmem_page()->free_reserved_page(), not __free_pages_core(). In
the future we might want to also use __free_pages_core() to initialize
highmem memory, to make that less special, and consider moving
totalram_pages updates into __free_pages_core() [1], so we can just use
adjust_managed_page_count() in there as well.
Booting a simple kernel in QEMU reveals no highmem accounting change:
Before:
Memory: 3095448K/3145208K available (14802K kernel code, 2073K rwdata,
5000K rodata, 740K init, 556K bss, 49760K reserved, 0K cma-reserved,
2244488K highmem)
After:
Memory: 3095276K/3145208K available (14802K kernel code, 2073K rwdata,
5000K rodata, 740K init, 556K bss, 49932K reserved, 0K cma-reserved,
2244488K highmem)
[1] https://lkml.kernel.org/r/20240601133402.2675-1-richard.weiyang@gmail.com
This patch (of 2):
Can we get rid of the highmem ifdef in adjust_managed_page_count()?
Likely yes: we don't have that many totalhigh_pages() users, and they all
don't seem to be very performance critical.
So let's implement totalhigh_pages() like nr_free_highpages(), collecting
information from all zones. This is now similar to what we do in
si_meminfo_node() to collect the per-node highmem page count.
In the common case (single node, 3-4 zones), we really shouldn't care. We
could optimize a bit further (only walk ZONE_HIGHMEM and ZONE_MOVABLE if
required), but there doesn't seem a real need for that.
[david@redhat.com: fix build bot complaint]
Link: https://lkml.kernel.org/r/b57e5bc4-eb72-40e3-add4-57dfa6e03df6@redhat.com
Link: https://lkml.kernel.org/r/20240607083711.62833-1-david@redhat.com
Link: https://lkml.kernel.org/r/20240607083711.62833-2-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Wei Yang <richard.weiyang@gmail.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
