[Devel] [NEW KERNEL] 3.10.0-229.7.2.vz7.9.13 (rhel7)

2015-11-23 Thread builder
Changelog: OpenVZ kernel rh7-3.10.0-229.7.2.vz7.9.13 * vzctl console fixed not to require extra "enter" * CT root should be able to override mlock limit as on the HN * vmscan enhanced: case when file lru is small in size * memory leak of cfq_set_request fixed * resurrect support for xt_connlimit

Re: [Devel] [PATCH rh7] net: avoid reference counter overflows on fib_rules in multicast forwarding

2015-11-23 Thread Stanislav Kinsburskiy
Andrey, please review 23.11.2015 17:08, Stanislav Kinsburskiy пишет: From: Hannes Frederic Sowa Bob Falken reported that after 4G packets, multicast forwarding stopped working. This was because of a rule reference counter overflow which freed the rule as soon as the overflow happend. This pat

[Devel] [RFC] NETNS_REFCNT_DEBUG define

2015-11-23 Thread Stanislav Kinsburskiy
Hello gentlemen, There was NETNS_REFCNT_DEBUG debug introduced in past to track down situations, when network namespace is being hold (not referenced; another ref count is used) by various objects, so, on net namespace release such situations with leaked network objects can be reported. We do

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiу
23 нояб. 2015 г. 17:15 пользователь Andrey Ryabinin написал: > > > > On 11/23/2015 07:04 PM, Stanislav Kinsburskiy wrote: > > > > > > 23.11.2015 16:53, Andrey Ryabinin пишет: > >> > >> On 11/23/2015 06:46 PM, Stanislav Kinsburskiу wrote: > >> > >>> +1 > >>> I'm curious, why network names

[Devel] [PATCH RHEL7 COMMIT] ve/memcg: enable kmem accounting for all cgroups

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit 35bf94617485b3842f8a45241673fb0f7e81189d Author: Vladimir Davydov Date: Mon Nov 23 20:19:26 2015 +0400 ve/memcg: e

[Devel] [PATCH rh7] net: avoid reference counter overflows on fib_rules in multicast forwarding

2015-11-23 Thread Stanislav Kinsburskiy
From: Hannes Frederic Sowa Bob Falken reported that after 4G packets, multicast forwarding stopped working. This was because of a rule reference counter overflow which freed the rule as soon as the overflow happend. This patch solves this by adding the FIB_LOOKUP_NOREF flag to fib_rules_lookup c

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 07:04 PM, Stanislav Kinsburskiy wrote: > > > 23.11.2015 16:53, Andrey Ryabinin пишет: >> >> On 11/23/2015 06:46 PM, Stanislav Kinsburskiу wrote: >> >>> +1 >>> I'm curious, why network namespace is not reported as leaked... >>> > 3) This object holds network namespace. And it a

[Devel] [PATCH rh7] ipmr, ip6mr: call ip6mr_free_table() on failure path

2015-11-23 Thread Stanislav Kinsburskiy
From: WANG Cong Signed-off-by: Cong Wang Signed-off-by: David S. Miller Signed-off-by: Stanislav Kinsburskiy --- net/ipv4/ipmr.c |2 +- net/ipv6/ip6mr.c |2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 4234b01..ce25a5c 1006

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
Please, ignore this patch. It's replaced by: "net: avoid reference counter overflows on fib_rules in multicast forwarding" 23.11.2015 13:47, Stanislav Kinsburskiy пишет: This patch fixed flase positive, reported by KASan. https://jira.sw.ru/browse/PSBM-41453 Signed-off-by: Stanislav Kinsburs

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
23.11.2015 16:53, Andrey Ryabinin пишет: On 11/23/2015 06:46 PM, Stanislav Kinsburskiу wrote: +1 I'm curious, why network namespace is not reported as leaked... 3) This object holds network namespace. And it also have to be leaked leaked after CT stop. Is it? Why do you think it leaked?

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 06:46 PM, Stanislav Kinsburskiу wrote: > +1 > I'm curious, why network namespace is not reported as leaked... > >> >>> 3) This object holds network namespace. And it also have to be leaked >>> leaked after CT stop. Is it? Why do you think it leaked? How the rule could hold the

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiу
23 нояб. 2015 г. 16:34 пользователь Andrey Ryabinin написал: > > > > On 11/23/2015 06:02 PM, Stanislav Kinsburskiy wrote: > > > > > > 23.11.2015 14:49, Andrey Ryabinin пишет: > >> On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: > >>> This patch fixed flase positive, reported by KASan.

[Devel] [PATCH RHEL7 COMMIT] ms/netfilter: Restore xt_connlimit version 0

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit b470b1416c121def090d410f9c2fb2c804f6401c Author: Kirill Tkhai Date: Mon Nov 23 19:38:50 2015 +0400 ms/netfilter: R

[Devel] [PATCH RHEL7 COMMIT] ms/cfq-iosched: fix async oom queue handling

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit 93edfd2e140d8525c2090ce6e526f303b6490e1c Author: Tejun Heo Date: Mon Nov 23 19:30:37 2015 +0400 ms/cfq-iosched: fi

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 06:02 PM, Stanislav Kinsburskiy wrote: > > > 23.11.2015 14:49, Andrey Ryabinin пишет: >> On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: >>> This patch fixed flase positive, reported by KASan. >>> >> s/flase/false >> s/KASan/kmemleak >> >>> https://jira.sw.ru/browse/PSBM-4145

[Devel] [PATCH RHEL7 COMMIT] ms/cfq-iosched: simplify control flow in cfq_get_queue()

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit dcd21871f74e0faa10982e8de34661671c126556 Author: Tejun Heo Date: Mon Nov 23 19:30:36 2015 +0400 ms/cfq-iosched: si

[Devel] [PATCH RHEL7 COMMIT] ms/cfq-iosched: fix oom cfq_queue ref leak in cfq_set_request()

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit ee8789d3904c8028a72612038cfeaef293205fde Author: Tejun Heo Date: Mon Nov 23 19:30:37 2015 +0400 ms/cfq-iosched: fi

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 06:09 PM, Stanislav Kinsburskiy wrote: > > > 23.11.2015 14:49, Andrey Ryabinin пишет: >> On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: >>> This patch fixed flase positive, reported by KASan. >>> >> s/flase/false >> s/KASan/kmemleak >> >>> https://jira.sw.ru/browse/PSBM-4145

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
23.11.2015 16:09, Stanislav Kinsburskiy пишет: 23.11.2015 14:49, Andrey Ryabinin пишет: On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: This patch fixed flase positive, reported by KASan. s/flase/false s/KASan/kmemleak https://jira.sw.ru/browse/PSBM-41453 Signed-off-by: Stanislav

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
23.11.2015 14:49, Andrey Ryabinin пишет: On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: This patch fixed flase positive, reported by KASan. s/flase/false s/KASan/kmemleak https://jira.sw.ru/browse/PSBM-41453 Signed-off-by: Stanislav Kinsburskiy --- net/core/fib_rules.c |4 +++

[Devel] [PATCH RHEL7 COMMIT] ms/fsnotify: fix check in inotify fdinfo printing

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit efd7ba8c24a3ab0335e70ff9b76310338854a893 Author: Cyrill Gorcunov Date: Mon Nov 23 19:04:29 2015 +0400 ms/fsnotify:

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
23.11.2015 14:49, Andrey Ryabinin пишет: On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: This patch fixed flase positive, reported by KASan. s/flase/false s/KASan/kmemleak https://jira.sw.ru/browse/PSBM-41453 Signed-off-by: Stanislav Kinsburskiy --- net/core/fib_rules.c |4 +++

[Devel] [PATCH rh7] memcg: enable kmem accounting for all cgroups

2015-11-23 Thread Vladimir Davydov
Currently, kmem accounting is only enabled for a cgroup upon write to memory.kmem.limit_in_bytes. This is fine in case of containers, because we've already taught vzctl to do it, but we also want to enable it for system.slice, otherwise dcache used by the host will be scanned too aggressively in ca

[Devel] [PATCH RHEL7 COMMIT] ms/vmscan: do not force-scan file lru if its absolute size is small

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit 8a0590a689ba4e773503449f5833e2daaf4486c0 Author: Vladimir Davydov Date: Mon Nov 23 18:40:12 2015 +0400 ms/vmscan:

[Devel] [PATCH RHEL7 COMMIT] ve/mm: allow container's root to ignore mlock limit

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit 478ad724bb6e70890689904bf5e15d6391812ed2 Author: Andrey Ryabinin Date: Mon Nov 23 18:32:52 2015 +0400 ve/mm: allow

[Devel] [PATCH RHEL7 COMMIT] ve/vtty: n_tty -- Allow write on sole slave vtty peer

2015-11-23 Thread Konstantin Khorenko
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.9.12 --> commit 14ac9decf1ef2375fbb7fdc35c4a8bd402d2f841 Author: Cyrill Gorcunov Date: Mon Nov 23 18:16:07 2015 +0400 ve/vtty: n_t

Re: [Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 03:47 PM, Stanislav Kinsburskiy wrote: > This patch fixed flase positive, reported by KASan. > s/flase/false s/KASan/kmemleak > https://jira.sw.ru/browse/PSBM-41453 > > Signed-off-by: Stanislav Kinsburskiy > --- > net/core/fib_rules.c |4 > 1 file changed, 4 insertions(+

Re: [Devel] [PATCH rh7] vtty: n_tty -- Allow write on sole slave vtty peer

2015-11-23 Thread Cyrill Gorcunov
On Mon, Nov 23, 2015 at 04:04:22PM +0300, Vladimir Davydov wrote: > > +#ifdef CONFIG_VE > > + if (tty_hung_up_p(file) || > > + (tty->link && !tty->link->count && > > +!(test_bit(TTY_PINNED_BY_OTHER, &tty->link->flags { > > +#else > > if (tty_h

[Devel] [PATCH] module: do not taint kernel with TAINT_OOT_MODULE

2015-11-23 Thread Evgenii Shatokhin
https://jira.sw.ru/browse/PSBM-41455 When an out-of-tree kernel module is loaded, the kernel is tainted with TAINT_OOT_MODULE (see check_modinfo() in kernel/module.c). add_taint() will dump the stack trace in this case. If I understand it correctly, this will be seen as an error during the testin

Re: [Devel] [PATCH rh7] vtty: n_tty -- Allow write on sole slave vtty peer

2015-11-23 Thread Vladimir Davydov
On Wed, Nov 18, 2015 at 04:16:07PM +0300, Cyrill Gorcunov wrote: > In pcs6 the pty counting has been somewhat sophisticated so in > pcs7 we've simplified it to bring as less changes into vanilla > code as possible (introducing TTY_PINNED_BY_OTHER bit to order > closing sequence). The new accounting

[Devel] [PATCH rh7] fib_rules: mark default fib reles as BLACK

2015-11-23 Thread Stanislav Kinsburskiy
This patch fixed flase positive, reported by KASan. https://jira.sw.ru/browse/PSBM-41453 Signed-off-by: Stanislav Kinsburskiy --- net/core/fib_rules.c |4 1 file changed, 4 insertions(+) diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 0e91311..bc69cef 100644 --- a/net/c

Re: [Devel] [PATCH rh7] mm, ve: allow container's root to to ignore mlock limit

2015-11-23 Thread Vladimir Davydov
On Mon, Nov 23, 2015 at 12:51:49PM +0300, Andrey Ryabinin wrote: > Global root is allowed to exceed memlock limit, so this should be > allowed for container's root too. > capable() works only for global root, so use ve_capable() instead. > > Signed-off-by: Andrey Ryabinin Reviewed-by: Vladimir D

[Devel] [PATCH rh7] vmscan: do not force-scan file lru if its absolute size is small

2015-11-23 Thread Vladimir Davydov
We assume there is enough inactive page cache if the size of inactive file lru is greater than the size of active file lru, in which case we force-scan file lru ignoring anonymous pages. While this logic works fine when there are plenty of page cache pages, it fails if the size of file lru is small

Re: [Devel] [PATCH rh7] mm, ve: allow container's root to to ignore mlock limit

2015-11-23 Thread Andrey Ryabinin
On 11/23/2015 12:51 PM, Andrey Ryabinin wrote: > Global root is allowed to exceed memlock limit, so this should be > allowed for container's root too. > capable() works only for global root, so use ve_capable() instead. > https://jira.sw.ru/browse/PSBM-41405 > Signed-off-by: Andrey Ryabinin > -

[Devel] [PATCH rh7] mm, ve: allow container's root to to ignore mlock limit

2015-11-23 Thread Andrey Ryabinin
Global root is allowed to exceed memlock limit, so this should be allowed for container's root too. capable() works only for global root, so use ve_capable() instead. Signed-off-by: Andrey Ryabinin --- mm/mlock.c | 8 mm/mmap.c | 6 +++--- mm/mremap.c | 2 +- 3 files changed, 8 inser