The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit 90c3b81039abfa45f1037f86739a183374086912
Author: Pavel Tikhomirov
Date: Wed Oct 13 14:13:18 2021 +0300
cgroup: remove excess rcu_r
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit 0aa8d5b66160fa389809e3561b4b05bf62364d4c
Author: Valeriy Vdovin
Date: Wed Oct 13 14:13:30 2021 +0300
ve/cgroup: Add ve_owner field
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit 7f3e1ef61fc2a6dc3514dd9b64f595cc11388165
Author: Pavel Tikhomirov
Date: Wed Oct 13 14:13:32 2021 +0300
ve/cgroup: temporary ignore
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit fe5a9b228d84b824d82b7700cbbfd0322812ccbf
Author: Kirill Tkhai
Date: Wed Oct 13 14:13:32 2021 +0300
ve: use rcu_dereference for ve_
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit 898abc31593362995e322d2ae021ca9abfc353be
Author: Pavel Tikhomirov
Date: Wed Oct 13 14:13:31 2021 +0300
cgroup: split cgroup_get_ve
The commit is pushed to "branch-rh9-5.14.vz9.1.x-ovz" and will appear at
https://src.openvz.org/scm/ovz/vzkernel.git
after rh9-5.14.0-4.vz9.10.1
-->
commit 5697a105caf70bf9468025ca7192afebdbe80fbd
Author: Valeriy Vdovin
Date: Wed Oct 13 14:13:29 2021 +0300
ve/cgroup: Skip non-virtualiz
From: Andrey Ryabinin
Add 'lat' field to the mm_page_alloc tracepoint. It shows allocation
latency in microseconds (0.01 second).
Signed-off-by: Andrey Ryabinin
Reviewed-by: Cyrill Gorcunov
mm/page_alloc: fix latency in tracepoint.
Since transition from jiffies to sched_cloc
From: Andrey Ryabinin
Add to '/proc//vz_latency' column with maximal latency task have seen
in the last 2 minutes.
E.g.:
cat /proc/1/vz_latency
TypeTotal_latCalls Max (2min)
allocatomic:0 2940
Andrey Ryabinin (6):
ve/page_alloc, kstat: account allocation latencies per-task and
per-thread
/proc//vz_latency: Show maximal allocation latency in the last
2min.
/proc//vz_latency: Add scheduling stats
/proc/vz/latency: distinguish atomic allocations in irq from in task
atomi
From: Stanislav Kinsburskiy
Rebasing and splitting netfilters sybsystem
(port 66-diff-ve-net-netfilter-combined).
Part 1.
https://jira.sw.ru/browse/PSBM-18322
Signed-off-by: Kirill Tkhai
khorenko@: rebase to kernel-3.10.0-229.7.2.el7:
* hunk for include/net/netfilter/xt_log.h has been dropp
From: Vasily Averin
Weave Kubernetes plugin requires tuning of
/proc/sys/net/ipv4/neigh/weave/base_reachable_time in particular,
so let's export neighbour sysctls as well.
https://jira.sw.ru/browse/PSBM-92107
Signed-off-by: Konstantin Khorenko
(cherry picked from vz7 commit 8499e3458f18 ("ne
From: Pavel Tikhomirov
We have a situation when node worked for ~8 days and came to The Point
where jiffies > fence_wdog_jiffies64 (what ever updated
/sys/kernel/watchdog_timer stopped doing these) but node lived after
The Point for 17 more hours, these means that nobody called
fence_wdog_check_t
From: Andrey Ryabinin
Add to /proc/vz/latency 'alocirq' allocation type which shows allocation
latencies
done in irq contexts. 'alocatomic' now shows atomic allocations in task
contexts.
Also add 'Per-CPU alloc irq' which shows per-cpu 'alocirq' numbers.
Example of new output:
Version: 2.6
L
From: Andrey Ryabinin
Vstorage wants per-process allocation latencies:
- total accumulated latency (total time spent inside the kernel allocator)
- total alloc attempts (so that average latency can be calculated)
This adds /proc//vz_latency file which outputs the numbers:
Type
From: Konstantin Khorenko
Rebasing and splitting netfilters sybsystem
(port 66-diff-ve-net-netfilter-combined).
Part 1.
https://jira.sw.ru/browse/PSBM-18322
Signed-off-by: Kirill Tkhai
(cherry picked from vz7 commit c34a99c00f9d ("ve/netfilter: Implement
pernet net->ct.max / virtualize "nf_co
From: Andrey Ryabinin
Historically the "Lat" column in /proc/vz/latency showed max latency
in 5 seconds. Chnage it to max latency in the last 2 minutes, the same
as in the /proc//vz_latency
Signed-off-by: Andrey Ryabinin
(cherry-picked from vz7 commit 0c5707cfcc84 ("/proc/vz/latency: Show max
From: Konstantin Khorenko
Recent versions of containerd (as a part of k3s-1.19.5)
started to apply strict rules when parsing the contents of
'devices.list' files located in the devices cgroup.
Namely, the access token is allowed to contain only those values [rwm],
that are described in
https://ww
From: Andrey Ryabinin
Add scheduling latencies to /proc//vz_latency.
They are the same as alloc latencies - total cumulative
latency, number of schedule events, and latency maximum
in the last 2 minutes.
The sysctl kernel.sched_schedstats must be enabled to
see these stats.
https://jira.sw.ru/b
From: Konstantin Khorenko
At the moment XFS is not marked as ready to be mounted inside non-init
userns (see xfs_fs_type), while we previously decided to allow using XFS
inside a CT:
b86b9049ba50 ("xfs: allow to mount xfs fs inside a Container")
https://jira.sw.ru/browse/PSBM-72401
so let's
From: Kirill Tkhai
Patchset description:
Port autoloading of netfilter modules functuonality
https://jira.sw.ru/browse/PSBM-28910
Signed-off-by: Kirill Tkhai
Kirill Tkhai (4):
kmod: Move check of VE permitions from __call_usermodehelper_exec() to
upper functions
kmod: Port autol
From: Stanislav Kinsburskiy
Patchset description:
Create conntrack structures only if they are really needed
Allocate conntracks only after there is a rule which uses them.
v2: Allow after there is a rule and never prohibit.
khorenko@: the idea behind all of this:
we want to provide the possi
From: Kirill Tkhai
Adds latency calculation for:
kstat_glob.swap_in
kstat_glob.page_in
kstat_glob.alloc_lat
And fail count in:
kstat_glob.alloc_fails
Also incorporates fixups patches:
kstat: Make kstat_glob::swap_in percpu - core part
ve/mm/kstat: Port diff-ve-kstat-disable-interrupt
From: Jan Dakinevich
Some unprivileged containers desire this sysctl. Apparently it is safe
for userns root, and there is no reason to restrict them.
https://jira.sw.ru/browse/PSBM-91032
Signed-off-by: Jan Dakinevich
Reviewed-by: Vasily Averin
(cherry picked from vz7 commit dccf5143b132 ("ne
From: Pavel Tikhomirov
We've seen that docker 20+ not only writes "a *:* rwm" to privileged
docker container device-cgroup (as pre-19 version did) but also checks
the content after write, and docker expects that all devices are allowed
for privileged docker container.
In our VZCT we obviously ca
From: Kirill Tkhai
Allow conntracks to be allocated in case of these
rules are inserted.
https://jira.sw.ru/browse/PSBM-51050
Signed-off-by: Kirill Tkhai
Reviewed-by: Andrei Vagin
vz8 rebase notes:
=
(cherry picked from vz7 commit 60931ce1ffcf ("net: Mark conntrack users
in n
From: Kirill Tkhai
Allow conntracks to be allocated in case of these
rules are inserted.
https://jira.sw.ru/browse/PSBM-51050
Signed-off-by: Kirill Tkhai
Reviewed-by: Andrei Vagin
+++
ve/net: Delete allow_conntrack_allocation() from nf_synproxy
Since nf_conntrack_alloc() is not called ther
From: Konstantin Khorenko
Rebasing and splitting netfilters sybsystem
(port 66-diff-ve-net-netfilter-combined).
Part 1.
https://jira.sw.ru/browse/PSBM-18322
* diff-ve-nf-make-nf_ct_expect_max-sysctl-virtual
Author: Pavel Emelyanov
Subject: [PATCH rh6] ve: Make nf_ct_expect_max "virtualized"
Dat
From: Andrey Zhadchenko
Existing cgroup_get_e_css() is not suited for cgroup-v1 and will always
return root cgroup css. Implement new cgroup_get_e_ve_css to return
ve css.
https://jira.sw.ru/browse/PSBM-131253
Signed-off-by: Andrey Zhadchenko
Reviewed-by: Kirill Tkhai
===
---
Andrey Zhadchenko (2):
kernel/cgroup: implement cgroup_get_e_ve_css
mm/backing-dev: associate writeback with correct blkcg
include/linux/cgroup.h |2 ++
kernel/cgroup/cgroup.c | 19 +++
mm/backing-dev.c | 22 --
3 files change
From: Andrey Zhadchenko
Use cgroup_get_e_ve_css to get correct blkcg_css for writeback instances.
https://jira.sw.ru/browse/PSBM-131253
Signed-off-by: Andrey Zhadchenko
Reviewed-by: Kirill Tkhai
v2:
khorenko@: introduce a wrapper for getting blkcg_css from memcg_css.
===
30 matches
Mail list logo
