Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Pavel Emelianov [EMAIL PROTECTED] writes:
That's how OpenVZ sees the pid namespaces.
The main idea is that kernel keeps operating with tasks pid
as it did before, but each task obtains one more pid for each
pid type -
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
The set of functions process_session, task_session, process_group
and task_pgrp is confusing, as the names can be mixed with each other
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
3. Cleaner logic for namespace migration: with this approach
one need to save the virtual pid and let global one change;
with Suka's logic this is not clear how to migrate
Eric W. Biederman wrote:
Pavel Emelianov [EMAIL PROTECTED] writes:
Eric W. Biederman wrote:
Pavel Emelianov [EMAIL PROTECTED] writes:
That's how OpenVZ sees the pid namespaces.
The main idea is that kernel keeps operating with tasks pid
as it did before, but each task obtains one more
Daniel Lezcano wrote:
Pavel Emelianov wrote:
Eric W. Biederman wrote:
Pavel Emelianov [EMAIL PROTECTED] writes:
That's how OpenVZ sees the pid namespaces.
The main idea is that kernel keeps operating with tasks pid
as it did before, but each task obtains one more pid for each
pid
On Thu, May 24, 2007 at 12:26:16AM +0200, Guillaume Chazarain wrote:
As a sidenote, while in CFS-v13 a nice=0 tasks seems to get 10x more CPU
than a nice=10 one, with the group fairness patch, the ratio drops to
less than 2x (for tasks with the same UID).
gah ..silly me.
Can you repeat your
On Wed, May 23, 2007 at 08:32:52PM +0200, Ingo Molnar wrote:
Here's an attempt to extend CFS (v13) to be fair at a group level,
rather than just at task level. The patch is in a very premature state
(passes simple tests, smp load balance not supported yet) at this
point. I am sending it
Pavel Emelianov wrote:
Daniel Lezcano wrote:
Pavel Emelianov wrote:
Eric W. Biederman wrote:
Pavel Emelianov [EMAIL PROTECTED] writes:
That's how OpenVZ sees the pid namespaces.
The main idea is that kernel keeps operating with tasks pid
as it did before, but
* Srivatsa Vaddagiri [EMAIL PROTECTED] wrote:
Can you repeat your tests with this patch pls? With the patch applied,
I am now getting the same split between nice 0 and nice 10 task as
CFS-v13 provides (90:10 as reported by top )
5418 guest 20 0 2464 304 236 R 90 0.0
Vasily Averin wrote:
Jeff, Tejun,
Our RHEL5-based OpenVZ linux kernel reports about SATA-related issues:
VIA VT6420 SATA RAID Controller on MSI motherboard, x86_64 kernel based on
latest RHEL5 kernel,
On booting hardware initialized properly and all works fine some time, but then
it detects
On Fri, May 25, 2007 at 01:11:40PM +0200, Ingo Molnar wrote:
I was planning on reusing smpnice concepts here, with the difference
that we balance group weights across CPU in addition to total weight
of CPUs.
ok, that would be (much) simpler that any explicit vcpu approach. Do you
plan
On Fri, May 25, 2007 at 10:29:51AM +0200, Ingo Molnar wrote:
btw., what are you thoughts about SMP?
I was planning on reusing smpnice concepts here, with the difference
that we balance group weights across CPU in addition to total weight of
CPUs.
For ex, assuming weight of each task is 10
* Srivatsa Vaddagiri [EMAIL PROTECTED] wrote:
On Fri, May 25, 2007 at 10:29:51AM +0200, Ingo Molnar wrote:
btw., what are you thoughts about SMP?
I was planning on reusing smpnice concepts here, with the difference
that we balance group weights across CPU in addition to total weight
of
I guess you have to use strace64 or smth like this.
Christian Kaiser2 wrote:
Kir Kolyshkin [EMAIL PROTECTED] wrote on 05/24/2007 05:10:54 PM:
Can you check the value of __NR_fairsched_chwt in vzctl? It is defined
in include/vzsyscalls.h depending on your architecture. For ppc it
should be
* Srivatsa Vaddagiri [EMAIL PROTECTED] wrote:
On Fri, May 25, 2007 at 01:11:40PM +0200, Ingo Molnar wrote:
I was planning on reusing smpnice concepts here, with the difference
that we balance group weights across CPU in addition to total weight
of CPUs.
ok, that would be (much)
Christian,
sys_fairsched_chwt() can return ENOENT in the only case:
when fairsched node ID was not found in the list.
but VE0 node is always on the list.
So I guess there is something wrong with syscall args in your case.
Please apply the attached patch to the kernel and vzctl
recompile/reboot
On Fri, May 25, 2007 at 02:05:36PM +0200, Ingo Molnar wrote:
great. Btw., could you please keep the up to this point there should be
no behavioral change in CFS fundamental splitup of your patches -
sure ..basically the changes required in CFS core is the introduction of two
structures -
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
This includes #ifdefs in get/put_pid_ns and rewriting
the child_reaper() function to the more logical view.
This doesn't fit logically into any other patch
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
3. Cleaner logic for namespace migration: with this approach
one need to save the virtual pid and let global one change;
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
The set of functions process_session, task_session, process_group
and task_pgrp is
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
3. Cleaner logic for namespace migration: with this approach
one need to save the virtual pid and let global
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Pavel Emelianov [EMAIL PROTECTED] writes:
That's how OpenVZ sees the pid namespaces.
The main idea is that kernel keeps operating with tasks pid
as it did before, but
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Eric W. Biederman ([EMAIL PROTECTED]):
Serge E. Hallyn [EMAIL PROTECTED] writes:
3. Cleaner logic for namespace migration: with this approach
Christian Kaiser2 wrote:
Kirill Korotaev [EMAIL PROTECTED] wrote on 05/25/2007 02:41:09 PM:
Christian,
sys_fairsched_chwt() can return ENOENT in the only case:
when fairsched node ID was not found in the list.
but VE0 node is always on the list.
So I guess there is something wrong with
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
Serge E. Hallyn wrote:
Quoting Pavel Emelianov ([EMAIL PROTECTED]):
This includes #ifdefs in get/put_pid_ns and rewriting
the child_reaper() function to the more logical
Christian, please check the patch attached.
Kirill
Christian Kaiser2 wrote:
Kirill Korotaev [EMAIL PROTECTED] wrote on 05/25/2007 02:41:09 PM:
Christian,
sys_fairsched_chwt() can return ENOENT in the only case:
when fairsched node ID was not found in the list.
but VE0 node is always on the
Kirill Korotaev [EMAIL PROTECTED] wrote on 05/25/2007 04:04:35 PM:
Christian, please check the patch attached.
Kirill
--- ./include/asm-powerpc/systbl.h.ve2351 2007-05-24 11:48:17.
0 +0400
+++ ./include/asm-powerpc/systbl.h 2007-05-25 17:58:34.0 +0400
@@ -304,7 +304,7
Pavel Emelianov [EMAIL PROTECTED] writes:
*I* would like to know how you migrate a *part* of a virtual
server? What happens with pids, IPC ids, network connections?
There are many entities in VS that are not bound to task, but to
VS and if you migrate only half of them you're risking in
On Fri, May 25, 2007 at 05:05:16PM +0400, Kirill Korotaev wrote:
That way the scheduler would first pick a virtual CPU to schedule, and
then pick a user from that virtual CPU, and then a task from the user.
don't you mean the vice versa:
first use to scheduler, then VCPU (which is
Pavel Emelianov [EMAIL PROTECTED] writes:
Eric W. Biederman wrote:
Pavel Emelianov [EMAIL PROTECTED] writes:
That's true. Sending of signal from parent ns to children
is tricky question. It has many solutions, I wanted to
discuss which one is better:
With unix domain sockets and the
On Fri, 2007-05-25 at 16:27 +0400, Pavel Emelianov wrote:
refcount is not the only badness. As I said you have
race in testing superblock's namespace and setting it
and one more: do you use one struct proc_dir_entry
proc_root for all the superolocks?
Yep. Do you see some problems with
On Fri, 2007-05-25 at 10:39 +0400, Pavel Emelianov wrote:
It is, but if we make patch that OOPSes the kernel in 0.1%
of cases and we do know this - this MUST be fixed.
Oh, yeah, it certainly needs to be fixed. I just wanted to make sure
you weren't seeing some endemic problem with the
Srivatsa Vaddagiri wrote:
On Fri, May 25, 2007 at 05:05:16PM +0400, Kirill Korotaev wrote:
That way the scheduler would first pick a virtual CPU to schedule, and
then pick a user from that virtual CPU, and then a task from the user.
don't you mean the vice versa:
first use to scheduler, then
On Fri, May 25, 2007 at 08:18:56PM +0400, Kirill Korotaev wrote:
2 physical CPUs can't select the same VCPU at the same time.
i.e. VCPU can be running on 1 PCPU only at the moment.
and vice versa: PCPU can run only 1 VCPU at the given moment.
So serialization is done when we need to assign
Quoting [EMAIL PROTECTED] ([EMAIL PROTECTED]):
Subject: Enable signaling child reaper from parent ns.
From: Sukadev Bhattiprolu [EMAIL PROTECTED]
The reaper of a child namespace must receive signals from its parent pid
namespace but not receive any signals from its own namespace.
Dave Hansen [EMAIL PROTECTED] wrote:
| On Thu, 2007-05-24 at 13:24 +0400, Pavel Emelianov wrote:
| | +int is_global_init(struct task_struct *tsk)
| | +{
| | + return (task_active_pid_ns(tsk) == init_pid_ns tsk-pid == 1);
| |
| | This can OOPS if you pass arbitrary task to this
36 matches
Mail list logo
