On Thu, Apr 16, 2009 at 02:37:53PM -0400, Vivek Goyal wrote:
> On Wed, Apr 08, 2009 at 10:37:59PM +0200, Andrea Righi wrote:
>
> [..]
> > >
> > > - I can think of atleast one usage of uppper limit controller where we
> > > might have spare IO resources still we don't want to give it to a
> > >
On Fri, 17 Apr 2009 10:44:32 +0900 (JST) Ryo Tsuruta wrote:
> > > Hmm, how about iotrack-cgroup ?
> > >
> >
> > Well. blockio_cgroup has the same character count and is more specific.
>
> How about blkio_cgroup ?
Sounds good.
___
Containers mailing
On Fri, 17 Apr 2009 10:49:43 +0900
Takuya Yoshikawa wrote:
> Hi,
>
> I have a few question.
>- I have not yet fully understood how your controller are using
> bio_cgroup. If my view is wrong please tell me.
>
> o In my view, bio_cgroup's implementation strongly depends on
>page_cgo
KAMEZAWA Hiroyuki wrote:
> On Tue, 14 Apr 2009 22:21:12 +0200
> Andrea Righi wrote:
>
>> +Example:
>> +* Create an association between an io-throttle group and a bio-cgroup group
>> + with "bio" and "blockio" subsystems mounted in different mount points:
>> + # mount -t cgroup -o bio bio-cgroup
* Andrew Morton [2009-04-16 17:44:28]:
> > Hmm, how about iotrack-cgroup ?
> >
>
> Well. blockio_cgroup has the same character count and is more specific.
Sounds good to me.
--
Balbir
___
Containers mailing list
contain...@lists.linux-found
Hi,
I have a few question.
- I have not yet fully understood how your controller are using
bio_cgroup. If my view is wrong please tell me.
o In my view, bio_cgroup's implementation strongly depends on
page_cgoup's. Could you explain for what purpose does this
functionality itself sh
From: Andrew Morton
Subject: Re: [PATCH 3/9] bio-cgroup controller
Date: Thu, 16 Apr 2009 17:44:28 -0700
> On Fri, 17 Apr 2009 09:20:40 +0900 KAMEZAWA Hiroyuki
> wrote:
>
> > On Thu, 16 Apr 2009 15:29:37 -0700
> > Andrew Morton wrote:
> >
> > > On Tue, 14 Apr 2009 22:21:14 +0200
> > > Andrea
On Tue, 14 Apr 2009 22:21:12 +0200
Andrea Righi wrote:
> +Example:
> +* Create an association between an io-throttle group and a bio-cgroup group
> + with "bio" and "blockio" subsystems mounted in different mount points:
> + # mount -t cgroup -o bio bio-cgroup /mnt/bio-cgroup/
> + # cd /mnt/bi
On Fri, 17 Apr 2009 09:20:40 +0900 KAMEZAWA Hiroyuki
wrote:
> On Thu, 16 Apr 2009 15:29:37 -0700
> Andrew Morton wrote:
>
> > On Tue, 14 Apr 2009 22:21:14 +0200
> > Andrea Righi wrote:
> >
> > > Subject: [PATCH 3/9] bio-cgroup controller
> >
> > Sorry, but I have to register extreme distres
On Thu, 16 Apr 2009 15:29:37 -0700
Andrew Morton wrote:
> On Tue, 14 Apr 2009 22:21:14 +0200
> Andrea Righi wrote:
>
> > Subject: [PATCH 3/9] bio-cgroup controller
>
> Sorry, but I have to register extreme distress at the name of this.
> The term "bio" is well-established in the kernel and he
On Thu, 16 Apr 2009 12:42:36 +0200
Andrea Righi wrote:
> On Thu, Apr 16, 2009 at 08:58:14AM +0900, KAMEZAWA Hiroyuki wrote:
> > On Wed, 15 Apr 2009 15:23:57 +0200
> > Andrea Righi wrote:
> >
> > > On Wed, Apr 15, 2009 at 09:38:50PM +0900, Ryo Tsuruta wrote:
> > > > Hi Andrea and Kamezawa-san,
>
On Tue, 14 Apr 2009 22:21:14 +0200
Andrea Righi wrote:
> Subject: [PATCH 3/9] bio-cgroup controller
Sorry, but I have to register extreme distress at the name of this.
The term "bio" is well-established in the kernel and here we have a new
definition for the same term: "block I/O".
"bio" was a
On Tue, 14 Apr 2009 22:21:11 +0200
Andrea Righi wrote:
> Objective
> ~
> The objective of the io-throttle controller is to improve IO performance
> predictability of different cgroups that share the same block devices.
We should get an IO controller into Linux. Does anyone have a reason
Serge E. Hallyn wrote:
> Quoting Alexey Dobriyan (adobri...@gmail.com):
>> To save nsproxy, or to not save nsproxy?
>>
>> Don't think much, save it.
>>
>> I argue that nsproxy should be removed totally, if someone thinks otherwise.
>> ;-)
>
> You've got Oren starting to agree with you too. I p
Quoting Alexey Dobriyan (adobri...@gmail.com):
> To save nsproxy, or to not save nsproxy?
>
> Don't think much, save it.
>
> I argue that nsproxy should be removed totally, if someone thinks otherwise.
> ;-)
You've got Oren starting to agree with you too. I personally don't
much care in princi
On Thu, Apr 16, 2009 at 01:25:35PM +0800, Gui Jianfeng wrote:
> Vivek Goyal wrote:
> > +#ifdef CONFIG_IOSCHED_CFQ_HIER
> > +static void changed_cgroup(struct io_context *ioc, struct cfq_io_context
> > *cic)
> > +{
> > + struct cfq_queue *async_cfqq = cic_to_cfqq(cic, 0);
> > + struct cfq_queue
Chris Friesen wrote:
> Alexey Dobriyan wrote:
>> On Thu, Apr 16, 2009 at 12:42:17AM +0200, Greg Kurz wrote:
>>> On Wed, 2009-04-15 at 23:56 +0400, Alexey Dobriyan wrote:
>>
There are sockets and live netns as the most complex example. I'm not
prepared to describe it exactly, but people
On Wed, Apr 08, 2009 at 10:37:59PM +0200, Andrea Righi wrote:
[..]
> >
> > - I can think of atleast one usage of uppper limit controller where we
> > might have spare IO resources still we don't want to give it to a
> > cgroup because customer has not paid for that kind of service level. In
>
Alexey Dobriyan wrote:
> On Thu, Apr 16, 2009 at 12:42:17AM +0200, Greg Kurz wrote:
>> On Wed, 2009-04-15 at 23:56 +0400, Alexey Dobriyan wrote:
>
>>> There are sockets and live netns as the most complex example. I'm not
>>> prepared to describe it exactly, but people wishing to do C/R with
>>> "l
Quoting Dan Smith (da...@us.ibm.com):
> SH> At checkpoint, you only need the uts_sem right around the reading
> SH> (and length-checking) of uts_ns->name. You're keeping it around
> SH> writing into the checkpoint image (and cr_hbuf_get).
>
> Right, okay.
>
> SH> At restart, you shouldn't actual
Quoting Dan Smith (da...@us.ibm.com):
> This patch adds a spot in the task record to save the personality value
> of the task_struct.
>
> It also masks out READ_IMPLIES_EXEC before starting to avoid read-only
> maps being restored as read-exec. To avoid re-setting this flag too early,
> it also i
SH> At checkpoint, you only need the uts_sem right around the reading
SH> (and length-checking) of uts_ns->name. You're keeping it around
SH> writing into the checkpoint image (and cr_hbuf_get).
Right, okay.
SH> At restart, you shouldn't actually need uts_sem - the nsproxy is
SH> not yet availab
Quoting Dan Smith (da...@us.ibm.com):
> Fix the potential for breakage if our UTS changes during checkpoint or
> restart by grabbing uts_sem during these operations.
Thanks, but I think you're waving too big a stick here.
At checkpoint, you only need the uts_sem right around the reading
(and leng
Quoting Alexey Dobriyan (adobri...@gmail.com):
> > What Alexey wants, I believe, is for users to be able to not have
> > to worry about there being exploitable bugs in restart(2) which
> > unprivileged users can play with. And for the usual distro-kernel
> > reasons, saying use 'CONFIG_CHECKPOINT=
On Thu, Apr 16, 2009 at 12:42:17AM +0200, Greg Kurz wrote:
> On Wed, 2009-04-15 at 23:56 +0400, Alexey Dobriyan wrote:
> > There are sockets and live netns as the most complex example. I'm not
> > prepared to describe it exactly, but people wishing to do C/R with
> > "leaks" should be very careful
Fix the potential for breakage if our UTS changes during checkpoint or
restart by grabbing uts_sem during these operations.
Cc: or...@cs.columbia.edu
Signed-off-by: Dan Smith
---
checkpoint/ckpt_task.c | 12
checkpoint/rstr_task.c | 17 +++--
2 files changed, 19 inse
On Wed, Apr 15, 2009 at 04:16:09PM -0500, Serge E. Hallyn wrote:
> Quoting Oren Laadan (or...@cs.columbia.edu):
> >
> >
> > Serge E. Hallyn wrote:
> > > Quoting Dave Hansen (d...@linux.vnet.ibm.com):
> > >> On Wed, 2009-04-15 at 23:21 +0400, Alexey Dobriyan wrote:
> > >>> Is sysctl to control CAP
Quoting Alexey Dobriyan (adobri...@gmail.com):
> FIXME: check VMA has same parameters.
> FIXME: abort if target kernel has vDSO disabled (?)
> FIXME: restore pages, vDSO is writable after all.
>
> Signed-off-by: Alexey Dobriyan
> ---
>
> arch/x86/vdso/vdso32-setup.c |6 ++
> include/linux/c
Quoting Alexey Dobriyan (adobri...@gmail.com):
> Add "start" argument, it will be used to map vDSO to exactly same place
> on restart(2).
>
> Signed-off-by: Alexey Dobriyan
IIRC the first (bprm) argument is actually completely unused. So in my
version of this patch (in mid-march) I dropped that
This patch adds a spot in the task record to save the personality value
of the task_struct.
It also masks out READ_IMPLIES_EXEC before starting to avoid read-only
maps being restored as read-exec. To avoid re-setting this flag too early,
it also introduces a "task_ctx" object to hold information
With this patch, dm-ioband can work with the bio cgroup.
Based on 2.6.30-rc1
Signed-off-by: Hirokazu Takahashi
Signed-off-by: Ryo Tsuruta
---
drivers/md/dm-ioband-type.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
Index: linux-2.6.30-rc1/drivers/md/dm-ioband-type.c
This patch contains several hooks that let the bio-cgroup framework to know
which bio-group is the owner of a page before starting I/O against the page.
Based on 2.6.30-rc1
Signed-off-by: Hirokazu Takahashi
Signed-off-by: Ryo Tsuruta
---
fs/buffer.c |2 ++
fs/direct-io.c |
This is the body of bio-cgroup.
Based on 2.6.30-rc1
Signed-off-by: Hirokazu Takahashi
Signed-off-by: Ryo Tsuruta
---
block/blk-ioc.c | 30 ++---
include/linux/biotrack.h | 83 ++
include/linux/cgroup_subsys.h |6 +
include/linux/iocontext.h |1
i
This patch makes the page_cgroup framework be able to be used even if
the compile option of the cgroup memory controller is off.
So bio-cgroup can use this framework without the memory controller.
Based on 2.6.30-rc1
Signed-off-by: Hirokazu Takahashi
Signed-off-by: Ryo Tsuruta
---
include/linu
Hi all,
This is a new release of bio-cgroup which provides an IO tracking
mechanism. The patches can be applied to the kernel 2.6.30-rc1 and you
can also download them from the following site.
http://people.valinux.co.jp/~ryov/bio-cgroup/
What's bio-cgroup all about?
Hi Andrea and Kamezawa-san,
> > > +#ifdef CONFIG_CGROUP_BIO
> > > +/*
> > > + * use lower 16 bits for flags and reserve the rest for the bio-cgroup id
> > > + */
> > > +#define BIO_CGROUP_ID_SHIFT (16)
> > > +#define BIO_CGROUP_ID_BITS (8 * sizeof(unsigned long) -
> > > BIO_CGROUP_ID_SHIFT)
On Thu, Apr 16, 2009 at 08:58:14AM +0900, KAMEZAWA Hiroyuki wrote:
> On Wed, 15 Apr 2009 15:23:57 +0200
> Andrea Righi wrote:
>
> > On Wed, Apr 15, 2009 at 09:38:50PM +0900, Ryo Tsuruta wrote:
> > > Hi Andrea and Kamezawa-san,
> > >
> > > > Ryo, it would be great if you can look at this and fix/
37 matches
Mail list logo
