Re: [Devel] [PATCH rh7] ploop: use GFP_NOIO in ploop_make_request

On 08/31/2015 06:59 AM, Konstantin Khorenko wrote: Maxim, please review. Maxim, Perhaps the following analysis by Alex Kompel will be helpful for you while reviewing this patch: I ran into the same issue. I think the problem arises when ploop code calls bio_alloc from ploop_make_request

[Devel] [PATCH RHEL7 COMMIT] ms/mm/vmscan: Do not wait for page writeback for GFP_NOFS allocations

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 32c7d3e46ac5734aa426767243a7e29657141ec3 Author: Michal Hocko Date: Mon Aug 31 18:57:12 2015 +0400

[Devel] [PATCH RHEL7 COMMIT] ve/sysctl: Introduce proc_doulongvec_minmax_virtual()

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 85474cc55aa11512b45b40bf382c620d78646992 Author: Andrey Ryabinin Date: Mon Aug 31 19:29:17 2015 +0400

[Devel] [PATCH RHEL7 COMMIT] ve/fs/aio: aio_nr & aio_max_nr variables virtualization

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit d5a0970d86642a4150439d8a599f2f359e75fbf4 Author: Andrey Ryabinin Date: Mon Aug 31 19:38:05 2015 +0400

[Devel] [PATCH RHEL7 COMMIT] pfcache/ext4: fix automatic csum calculation

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 17c90deb8c54a3feef19d557008fcb510bed8cd3 Author: Dmitry Monakhov Date: Mon Aug 31 20:03:06 2015 +0400

Re: [Devel] [RFC rh7 v6] ve/tty: vt -- Implement per VE support for console and terminals

On Mon, Aug 31, 2015 at 02:25:42PM +0300, Cyrill Gorcunov wrote: ... > Index: linux-pcs7.git/drivers/tty/pty.c > === > --- linux-pcs7.git.orig/drivers/tty/pty.c > +++ linux-pcs7.git/drivers/tty/pty.c > @@ -820,10 +820,524 @@ static

Re: [Devel] [RFC rh7 v6] ve/tty: vt -- Implement per VE support for console and terminals

On Mon, Aug 31, 2015 at 08:19:50PM +0300, Vladimir Davydov wrote: OK, to all prev comments, thank you! > > + tty = tty_init_dev(vttys_driver, idx); > > + if (IS_ERR(tty)) > > + goto err_install; > > So we have master:0 slave:1 here, right? yes > > > +

Re: [Devel] [RFC rh7 v6] ve/tty: vt -- Implement per VE support for console and terminals

On Mon, Aug 31, 2015 at 09:07:07PM +0300, Cyrill Gorcunov wrote: > On Mon, Aug 31, 2015 at 08:19:50PM +0300, Vladimir Davydov wrote: > > OK, to all prev comments, thank you! > > > > + tty = tty_init_dev(vttys_driver, idx); > > > + if (IS_ERR(tty)) > > > + goto

Re: [Devel] [RFC rh7 v6] ve/tty: vt -- Implement per VE support for console and terminals

On Mon, Aug 31, 2015 at 10:36:25PM +0300, Vladimir Davydov wrote: > > Yeah, you're right. But then we diverge from pty ref counting design, > which isn't good IMO. To fix that, let's increment slave's count twice > on vtty_install, once for file and once for master reference. Then we If we

[Devel] [PATCH rh7 3/5] ub: zap ub_tmpfs_respages

It is always 0 in both Vz7 and PCS6, so drop it. Signed-off-by: Vladimir Davydov --- include/bc/beancounter.h | 7 --- kernel/bc/beancounter.c | 1 - kernel/bc/vm_pages.c | 19 --- 3 files changed, 27 deletions(-) diff --git

[Devel] [PATCH rh7] ub: zap ub_dirty_pages

It is not used anywhere. Signed-off-by: Vladimir Davydov --- include/bc/io_acct.h | 7 --- 1 file changed, 7 deletions(-) diff --git a/include/bc/io_acct.h b/include/bc/io_acct.h index 5b51853e5b5f..fa7afb1cb605 100644 --- a/include/bc/io_acct.h +++

[Devel] [RFC rh7 v6] ve/tty: vt -- Implement per VE support for console and terminals

Previously in commit 8674c044330fad1458bd59b02f9037fb97e8b7af stubs for virtual terminals have been added, they support writes from kernel side which simply drops into the void. In the patch the code has been moved from kernel/ve/console.c to drivers/tty/pty.c to reuse a couple of pty helpers.

[Devel] [PATCH rh7] writeback: revert ub dirty limit related stuff

This patch reverts ub dirty limit related hunks brought by the initial commit 2a8b5de95918. None of them actually works, so this patch introduces no functional changes. Dirty set control will be reimplemented in the scope of PSBM-33841. Signed-off-by: Vladimir Davydov ---

[Devel] [PATCH rh7 4/5] ub: rename private_data2 to iolimit

ub->private_data2 is only used for storing iolimit housekeeping struct, so call it appropriately. Signed-off-by: Vladimir Davydov --- include/bc/beancounter.h | 3 +-- kernel/bc/beancounter.c | 2 +- kernel/ve/vziolimit.c| 14 +++--- 3 files changed, 9

[Devel] [PATCH rh7 5/5] ub: fix typo: s/ub_dirty_radio/ub_dirty_ratio

Signed-off-by: Vladimir Davydov --- include/bc/io_acct.h| 2 +- kernel/bc/beancounter.c | 4 ++-- kernel/bc/io_acct.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/include/bc/io_acct.h b/include/bc/io_acct.h index

[Devel] [PATCH rh7 1/5] ub: zap unused socket accounting bits

It should have been done in the scope of c73bfca7594c ("bc: Rip old network buffers and sockets accounting"). Signed-off-by: Vladimir Davydov --- include/bc/beancounter.h | 24 kernel/bc/beancounter.c | 8 2 files changed, 32

[Devel] [PATCH rh7 2/5] ub: drop swapin/swapout stats

Swapin/swapout cannot be accounted by beancounters anymore, because memory management moved to memcg. Right now, these stats are not provided by memcg, so this patch simply drops them from /proc/vmstat inside container and from /proc/bc/CTID/vmaux on the host. If anybody requests these counters,

[Devel] [PATCH RHEL7 COMMIT] ms/cgroup: remove now unused css_depth()

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 8999360445307b87d687dac0b551d21e02386426 Author: Tejun Heo Date: Wed Jun 12 21:04:48 2013 -0700

[Devel] [PATCH RHEL7 COMMIT] ve/video/logo: show Odin's logo on boot

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 051055ab058cba98ca8e0fe93689162588cb556a Author: Andrey Ryabinin Date: Mon Aug 31 17:22:42 2015 +0400

Re: [Devel] [PATCH 00/17] oom killer enhancements

Kirill, please review. -- Best regards, Konstantin Khorenko, Virtuozzo Linux Kernel Team On 08/14/2015 08:03 PM, Vladimir Davydov wrote: - Patches 1-3 fix stalls on global (1, 2) and local (3) reclaim. https://jira.sw.ru/browse/PSBM-35155 - Patches 4-6 revert our code implementing per

[Devel] [PATCH RHEL7 COMMIT] config.OpenVZ: show Odin's logo on boot

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit eeddf77a216082687f589e408d3f560ced20ab96 Author: Konstantin Khorenko Date: Mon Aug 31 17:24:47 2015

[Devel] [PATCH RHEL7 COMMIT] ve/tty: vt -- Fix nil dereference due to race

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 0fa0a39ad2c644f55f447cef85e5a9a8f06e43b7 Author: Cyrill Gorcunov Date: Mon Aug 31 17:08:30 2015 +0400

[Devel] [PATCH RHEL7 COMMIT] ms/mm: memcontrol: reclaim at least once for __GFP_NORETRY

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit 8aee40dd982b476c208aee78e19cd756f2dac8a7 Author: Johannes Weiner Date: Mon Aug 31 17:09:47 2015 +0400

Re: [Devel] [PATCH rh7] memcg: count all oom kills

Kirill, please review. -- Best regards, Konstantin Khorenko, Virtuozzo Linux Kernel Team On 08/14/2015 12:48 PM, Vladimir Davydov wrote: We do not count processes killed because they share victim's mm. Fix it. Fixes: 66053f4201e41 ("memcg: count oom kills") Signed-off-by: Vladimir Davydov

[Devel] [PATCH RHEL7 COMMIT] ve/kmod: fix out-of-bounds access in call_modprobe()

The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.4 --> commit e2164f15d2f004ce076da3aa925b681bd8cde8d8 Author: Andrey Ryabinin Date: Mon Aug 31 17:15:30 2015 +0400

Re: [Devel] [PATCH rh7] ploop: use GFP_NOIO in ploop_make_request

Maxim, please review. Do we need the same in PCS6? -- Best regards, Konstantin Khorenko, Virtuozzo Linux Kernel Team On 08/17/2015 04:30 PM, Vladimir Davydov wrote: Currently, we use GFP_NOFS, which may result in a dead lock as follows: filemap_fault do_mpage_readpage submit_bio