On 01/24/2013 06:34 PM, Mike Galbraith wrote:
> On Thu, 2013-01-24 at 17:26 +0800, Michael Wang wrote:
>> On 01/24/2013 05:07 PM, Mike Galbraith wrote:
>>> On Thu, 2013-01-24 at 16:14 +0800, Michael Wang wrote:
>>>
Now it's time to work on v3 I think, let's see what we could get this time.
>>
On Thu, 2013-01-24 at 17:26 +0800, Michael Wang wrote:
> On 01/24/2013 05:07 PM, Mike Galbraith wrote:
> > On Thu, 2013-01-24 at 16:14 +0800, Michael Wang wrote:
> >
> >> Now it's time to work on v3 I think, let's see what we could get this time.
> >
> > Maybe v3 can try to not waste so much ram
On 01/24/2013 05:07 PM, Mike Galbraith wrote:
> On Thu, 2013-01-24 at 16:14 +0800, Michael Wang wrote:
>
>> Now it's time to work on v3 I think, let's see what we could get this time.
>
> Maybe v3 can try to not waste so much ram on affine map?
Yeah, that has been a question in my mind at very b
On Thu, 2013-01-24 at 16:14 +0800, Michael Wang wrote:
> Now it's time to work on v3 I think, let's see what we could get this time.
Maybe v3 can try to not waste so much ram on affine map?
Even better would be if it could just go away, along with relic of the
bad old days wake_affine(), and we
On 01/24/2013 03:47 PM, Mike Galbraith wrote:
> On Thu, 2013-01-24 at 15:15 +0800, Michael Wang wrote:
>> On 01/24/2013 02:51 PM, Mike Galbraith wrote:
>>> On Thu, 2013-01-24 at 14:01 +0800, Michael Wang wrote:
>>>
I've enabled WAKE flag on my box like you did, but still can't see
regres
On Thu, 2013-01-24 at 15:15 +0800, Michael Wang wrote:
> On 01/24/2013 02:51 PM, Mike Galbraith wrote:
> > On Thu, 2013-01-24 at 14:01 +0800, Michael Wang wrote:
> >
> >> I've enabled WAKE flag on my box like you did, but still can't see
> >> regression, and I've just tested on a power server wit
On 01/24/2013 02:51 PM, Mike Galbraith wrote:
> On Thu, 2013-01-24 at 14:01 +0800, Michael Wang wrote:
>
>> I've enabled WAKE flag on my box like you did, but still can't see
>> regression, and I've just tested on a power server with 64 cpu, also
>> failed to reproduce the issue (not compared with
On 01/24/2013 02:01 PM, Michael Wang wrote:
> On 01/23/2013 05:32 PM, Mike Galbraith wrote:
> [snip]
>> ---
>> include/linux/topology.h |6 ++---
>> kernel/sched/core.c | 41 ++---
>> kernel/sched/fair.c | 52
>> +--
On Thu, 2013-01-24 at 14:01 +0800, Michael Wang wrote:
> I've enabled WAKE flag on my box like you did, but still can't see
> regression, and I've just tested on a power server with 64 cpu, also
> failed to reproduce the issue (not compared with virgin yet, but can't
> see collapse).
I'm not surp
On 01/23/2013 05:32 PM, Mike Galbraith wrote:
[snip]
> ---
> include/linux/topology.h |6 ++---
> kernel/sched/core.c | 41 ++---
> kernel/sched/fair.c | 52
> +--
> 3 files changed, 70 insertions(+), 29
On Wed, 2013-01-23 at 17:26 +0800, Michael Wang wrote:
> On 01/23/2013 05:18 PM, Mike Galbraith wrote:
> > On Wed, 2013-01-23 at 17:00 +0800, Michael Wang wrote:
> >> On 01/23/2013 04:49 PM, Mike Galbraith wrote:
> >>> On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
> On 01/23/2013 0
On Wed, 2013-01-23 at 10:18 +0100, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 17:00 +0800, Michael Wang wrote:
> > On 01/23/2013 04:49 PM, Mike Galbraith wrote:
> > > On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
> > >> On 01/23/2013 04:20 PM, Mike Galbraith wrote:
> > >>> On Wed, 20
On 01/23/2013 05:18 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 17:00 +0800, Michael Wang wrote:
>> On 01/23/2013 04:49 PM, Mike Galbraith wrote:
>>> On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
On 01/23/2013 04:20 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 15:10 +
On Wed, 2013-01-23 at 17:00 +0800, Michael Wang wrote:
> On 01/23/2013 04:49 PM, Mike Galbraith wrote:
> > On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
> >> On 01/23/2013 04:20 PM, Mike Galbraith wrote:
> >>> On Wed, 2013-01-23 at 15:10 +0800, Michael Wang wrote:
> On 01/23/2013 0
On 01/23/2013 04:49 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
>> On 01/23/2013 04:20 PM, Mike Galbraith wrote:
>>> On Wed, 2013-01-23 at 15:10 +0800, Michael Wang wrote:
On 01/23/2013 02:28 PM, Mike Galbraith wrote:
>>>
> Abbreviated test run:
>>>
On Wed, 2013-01-23 at 16:30 +0800, Michael Wang wrote:
> On 01/23/2013 04:20 PM, Mike Galbraith wrote:
> > On Wed, 2013-01-23 at 15:10 +0800, Michael Wang wrote:
> >> On 01/23/2013 02:28 PM, Mike Galbraith wrote:
> >
> >>> Abbreviated test run:
> >>> Tasksjobs/min jti jobs/min/task re
On 01/23/2013 04:20 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 15:10 +0800, Michael Wang wrote:
>> On 01/23/2013 02:28 PM, Mike Galbraith wrote:
>
>>> Abbreviated test run:
>>> Tasksjobs/min jti jobs/min/task real cpu
>>> 640 158044.01 81 246.9438 24.54
On Wed, 2013-01-23 at 15:10 +0800, Michael Wang wrote:
> On 01/23/2013 02:28 PM, Mike Galbraith wrote:
> > Abbreviated test run:
> > Tasksjobs/min jti jobs/min/task real cpu
> > 640 158044.01 81 246.9438 24.54577.66 Wed Jan 23
> > 07:14:33 2013
> > 1280
On 01/23/2013 02:28 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 13:09 +0800, Michael Wang wrote:
>> On 01/23/2013 12:31 PM, Mike Galbraith wrote:
>
>>> Another thing that wants fixing: root can set flags for _existing_
>>> domains any way he likes,
>>
>> Can he? on running time changing the
On Wed, 2013-01-23 at 13:09 +0800, Michael Wang wrote:
> On 01/23/2013 12:31 PM, Mike Galbraith wrote:
> > Another thing that wants fixing: root can set flags for _existing_
> > domains any way he likes,
>
> Can he? on running time changing the domain flags? I do remember I used to
> send out so
On 01/23/2013 12:31 PM, Mike Galbraith wrote:
> On Wed, 2013-01-23 at 10:44 +0800, Michael Wang wrote:
>> On 01/22/2013 10:41 PM, Mike Galbraith wrote:
>>> On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
>>>
What about this patch? May be the wrong map is the killer on balance
path
On Wed, 2013-01-23 at 11:01 +0800, Michael Wang wrote:
> On 01/22/2013 07:34 PM, Mike Galbraith wrote:
> I suppose it's A, so my logical is:
> >> 1. find idle cpu in prev domain.
> >> 2. if failed and affine, find idle cpu in current domain.
> >
> > Hm. If cpu and prev_cpu are cache affine, you
On Wed, 2013-01-23 at 10:44 +0800, Michael Wang wrote:
> On 01/22/2013 10:41 PM, Mike Galbraith wrote:
> > On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
> >
> >> What about this patch? May be the wrong map is the killer on balance
> >> path, should we check it? ;-)
> >
> > [1.23224
On 01/22/2013 07:34 PM, Mike Galbraith wrote:
> On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
>> On 01/22/2013 04:03 PM, Mike Galbraith wrote:
>> [snip]
>>> ...
>
> That was with your change backed out, and the q/d below applied.
So that change will help to solve the is
On 01/22/2013 10:41 PM, Mike Galbraith wrote:
> On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
>
>> What about this patch? May be the wrong map is the killer on balance
>> path, should we check it? ;-)
>
> [1.232249] Brought up 40 CPUs
> [1.236003] smpboot: Total of 40 processors
On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
> What about this patch? May be the wrong map is the killer on balance
> path, should we check it? ;-)
[1.232249] Brought up 40 CPUs
[1.236003] smpboot: Total of 40 processors activated (180873.90 BogoMIPS)
[1.244744] CPU0 attach
On Tue, 2013-01-22 at 16:56 +0800, Michael Wang wrote:
> On 01/22/2013 04:03 PM, Mike Galbraith wrote:
> [snip]
> > ...
> >>>
> >>> That was with your change backed out, and the q/d below applied.
> >>
> >> So that change will help to solve the issue? good to know :)
> >>
> >> But it will invoke
On 01/22/2013 04:03 PM, Mike Galbraith wrote:
[snip]
> ...
>>>
>>> That was with your change backed out, and the q/d below applied.
>>
>> So that change will help to solve the issue? good to know :)
>>
>> But it will invoke wake_affine() with out any delay, the benefit
>> of the patch set will be
On Tue, 2013-01-22 at 11:43 +0800, Michael Wang wrote:
> On 01/21/2013 05:44 PM, Mike Galbraith wrote:
> > On Mon, 2013-01-21 at 17:22 +0800, Michael Wang wrote:
> >> On 01/21/2013 05:09 PM, Mike Galbraith wrote:
> >>> On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
> On 01/21/2013 0
On 01/21/2013 05:44 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 17:22 +0800, Michael Wang wrote:
>> On 01/21/2013 05:09 PM, Mike Galbraith wrote:
>>> On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
On 01/21/2013 03:09 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 07:42 +
On Mon, 2013-01-21 at 10:44 +0100, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 17:22 +0800, Michael Wang wrote:
> > On 01/21/2013 05:09 PM, Mike Galbraith wrote:
> > > On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
> > >> On 01/21/2013 03:09 PM, Mike Galbraith wrote:
> > >>> On Mon, 20
On Mon, 2013-01-21 at 17:22 +0800, Michael Wang wrote:
> On 01/21/2013 05:09 PM, Mike Galbraith wrote:
> > On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
> >> On 01/21/2013 03:09 PM, Mike Galbraith wrote:
> >>> On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote:
> On Mon, 2013-
On 01/21/2013 05:09 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
>> On 01/21/2013 03:09 PM, Mike Galbraith wrote:
>>> On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote:
On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
>>>
> May be we coul
On Mon, 2013-01-21 at 16:46 +0800, Michael Wang wrote:
> On 01/21/2013 04:26 PM, Mike Galbraith wrote:
> > On Mon, 2013-01-21 at 15:34 +0800, Michael Wang wrote:
> >> On 01/21/2013 02:42 PM, Mike Galbraith wrote:
> >>> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
> >>>
> That seems
On Mon, 2013-01-21 at 15:45 +0800, Michael Wang wrote:
> On 01/21/2013 03:09 PM, Mike Galbraith wrote:
> > On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote:
> >> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
> >
> >>> May be we could try change this back to the old way later, aft
On 01/21/2013 04:26 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 15:34 +0800, Michael Wang wrote:
>> On 01/21/2013 02:42 PM, Mike Galbraith wrote:
>>> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
>>>
That seems like the default one, could you please show me the numbers in
On Mon, 2013-01-21 at 15:34 +0800, Michael Wang wrote:
> On 01/21/2013 02:42 PM, Mike Galbraith wrote:
> > On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
> >
> >> That seems like the default one, could you please show me the numbers in
> >> your datapoint file?
> >
> > Yup, I do not touc
On 01/21/2013 03:09 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote:
>> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
>
>>> May be we could try change this back to the old way later, after the aim
>>> 7 test on my server.
>>
>> Yeah, something funny is
On 01/21/2013 02:42 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
>
>> That seems like the default one, could you please show me the numbers in
>> your datapoint file?
>
> Yup, I do not touch the workfile. Datapoints is what you see in the
> tabulated result.
On Mon, 2013-01-21 at 07:42 +0100, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
> > May be we could try change this back to the old way later, after the aim
> > 7 test on my server.
>
> Yeah, something funny is going on.
Never entering balance path kills the co
On Mon, 2013-01-21 at 13:07 +0800, Michael Wang wrote:
> That seems like the default one, could you please show me the numbers in
> your datapoint file?
Yup, I do not touch the workfile. Datapoints is what you see in the
tabulated result...
1
1
1
5
5
5
10
10
10
...
so it does three consecutive
On 01/21/2013 12:38 PM, Mike Galbraith wrote:
> On Mon, 2013-01-21 at 10:50 +0800, Michael Wang wrote:
>> On 01/20/2013 12:09 PM, Mike Galbraith wrote:
>>> On Thu, 2013-01-17 at 13:55 +0800, Michael Wang wrote:
Hi, Mike
I've send out the v2, which I suppose it will fix the below BU
On Mon, 2013-01-21 at 10:50 +0800, Michael Wang wrote:
> On 01/20/2013 12:09 PM, Mike Galbraith wrote:
> > On Thu, 2013-01-17 at 13:55 +0800, Michael Wang wrote:
> >> Hi, Mike
> >>
> >> I've send out the v2, which I suppose it will fix the below BUG and
> >> perform better, please do let me know
On 01/20/2013 12:09 PM, Mike Galbraith wrote:
> On Thu, 2013-01-17 at 13:55 +0800, Michael Wang wrote:
>> Hi, Mike
>>
>> I've send out the v2, which I suppose it will fix the below BUG and
>> perform better, please do let me know if it still cause issues on your
>> arm7 machine.
>
> s/arm7/aim7
>
On Thu, 2013-01-17 at 13:55 +0800, Michael Wang wrote:
> Hi, Mike
>
> I've send out the v2, which I suppose it will fix the below BUG and
> perform better, please do let me know if it still cause issues on your
> arm7 machine.
s/arm7/aim7
Someone swiped half of CPUs/ram, so the box is now 2 10
Hi, Mike
I've send out the v2, which I suppose it will fix the below BUG and
perform better, please do let me know if it still cause issues on your
arm7 machine.
Regards,
Michael Wang
On 01/14/2013 05:21 PM, Mike Galbraith wrote:
> On Sat, 2013-01-12 at 11:19 +0100, Mike Galbraith wrote:
>
>>
On 01/15/2013 12:52 PM, Mike Galbraith wrote:
> On Tue, 2013-01-15 at 11:10 +0800, Michael Wang wrote:
>> hanks for the testing, could you please tell me which benchmark
>> generate these results?
>
> aim7, using the compute workfile, and a datapoints file containing
> $Tasks. multitask -nl -f w
On Tue, 2013-01-15 at 11:10 +0800, Michael Wang wrote:
> hanks for the testing, could you please tell me which benchmark
> generate these results?
aim7, using the compute workfile, and a datapoints file containing
$Tasks. multitask -nl -f will prompt for the datapoints file. You'll
have to bump
On 01/14/2013 05:21 PM, Mike Galbraith wrote:
> On Sat, 2013-01-12 at 11:19 +0100, Mike Galbraith wrote:
>
>> Hm, low end takes a big hit.
>
> Bah, that's perturbations and knobs.
>
> aim7 compute, three individual runs + average
>
> Stock scheduler knobs..
>
> 3.8-wang
On 01/12/2013 04:01 PM, Mike Galbraith wrote:
> On Wed, 2013-01-09 at 17:28 +0800, Michael Wang wrote:
>> On 12/27/2012 02:08 PM, Michael Wang wrote:
>>> This patch set is trying to simplify the select_task_rq_fair() with
>>> schedule balance map.
>>>
>>> After get rid of the complex code and reor
On 01/11/2013 06:13 PM, Nikunj A Dadhania wrote:
> Hi Michael,
>
> Michael Wang writes:
>> Prev:
>> +-+-+---+
>> | 7484 MB | 32 | 42463 |
>> Post:
>> | 7483 MB | 32 | 44185 | +0.18%
> That should be +4.05%
H
On Sat, 2013-01-12 at 11:19 +0100, Mike Galbraith wrote:
> Hm, low end takes a big hit.
Bah, that's perturbations and knobs.
aim7 compute, three individual runs + average
Stock scheduler knobs..
3.8-wangavg 3.8-virgin
avgvs
aim7 compute
Tasksjobs/min jti jobs/min/task real cpu
1 440.41 100 440.4070 13.76 3.65
5 1923.81 99 384.7619 15.75 26.17
10 4223.00 99 422.2997 14.35 41.66
20 7632.24 87 381.6121 15.88
On Wed, 2013-01-09 at 17:28 +0800, Michael Wang wrote:
> On 12/27/2012 02:08 PM, Michael Wang wrote:
> > This patch set is trying to simplify the select_task_rq_fair() with
> > schedule balance map.
> >
> > After get rid of the complex code and reorganize the logical, pgbench show
> > the improve
Hi Michael,
Michael Wang writes:
> Prev:
> +-+-+---+
> | 7484 MB | 32 | 42463 |
> Post:
> | 7483 MB | 32 | 44185 | +0.18%
That should be +4.05%
Regards
Nikunj
--
To unsubscribe from this list: send the
This patch set is trying to simplify the select_task_rq_fair() with
schedule balance map.
After get rid of the complex code and reorganize the logical, pgbench show
the improvement.
Prev:
| db_size | clients | tps |
+-+-+---+
On 12/27/2012 02:08 PM, Michael Wang wrote:
> This patch set is trying to simplify the select_task_rq_fair() with
> schedule balance map.
>
> After get rid of the complex code and reorganize the logical, pgbench show
> the improvement.
>
> Prev:
> | db_size | clients | tps |
57 matches
Mail list logo
