On 13 February 2013 21:03, Paul Turner wrote:
> On Tue, Feb 12, 2013 at 5:19 AM, Vincent Guittot
> wrote:
>> need_active_balance uses env->src_cpu which is set only if there is more
>> than 1 task on the run queue.
>> We must set the src_cpu field unconditionnally
>> otherwise the test
On 13 February 2013 21:02, Damien Wyart wrote:
>> Bingo, that was the problem in my setup: as the patch was applied
>> through a script with others, I had missed the error message about the
>> conflict (I have also another conflict which can be safely ignored so
>> the new one did not catch my
On 13 February 2013 21:02, Damien Wyart damien.wy...@gmail.com wrote:
Bingo, that was the problem in my setup: as the patch was applied
through a script with others, I had missed the error message about the
conflict (I have also another conflict which can be safely ignored so
the new one did
On 13 February 2013 21:03, Paul Turner p...@google.com wrote:
On Tue, Feb 12, 2013 at 5:19 AM, Vincent Guittot
vincent.guit...@linaro.org wrote:
need_active_balance uses env-src_cpu which is set only if there is more
than 1 task on the run queue.
We must set the src_cpu field unconditionnally
On Tue, Feb 12, 2013 at 5:19 AM, Vincent Guittot
wrote:
> need_active_balance uses env->src_cpu which is set only if there is more
> than 1 task on the run queue.
> We must set the src_cpu field unconditionnally
> otherwise the test "env->src_cpu > env->dst_cpu" will always fail if there is
>
> Bingo, that was the problem in my setup: as the patch was applied
> through a script with others, I had missed the error message about the
> conflict (I have also another conflict which can be safely ignored so
> the new one did not catch my eye)... So the patch was only
> half-applied, and the
* Vincent Guittot [2013-02-13 18:49]:
> I have look into Frederic's tree but i didn't find any reason that
> could explain your problem. May be Frederic will have some ideas
> I have also tested his branch with and without my patch and both
> kernel are booting (on an ARM platform without using
On 13 February 2013 15:28, Vincent Guittot wrote:
> On 13 February 2013 15:08, Damien Wyart wrote:
>> * Vincent Guittot [2013-02-13 13:08]:
>>> Damien,
>>> Regarding your sched_domain config and especially the flags field, you
>>> should not be impacted by my patch because
>>> -
On 13 February 2013 15:08, Damien Wyart wrote:
> * Vincent Guittot [2013-02-13 13:08]:
>> Damien,
>> Regarding your sched_domain config and especially the flags field, you
>> should not be impacted by my patch because
>> - need_active_balance is the only new place that use env->src_cpu in
>> the
* Vincent Guittot [2013-02-13 13:08]:
> Damien,
> Regarding your sched_domain config and especially the flags field, you
> should not be impacted by my patch because
> - need_active_balance is the only new place that use env->src_cpu in
> the load_balance function
> - and your machine will never
Damien,
Regarding your sched_domain config and especially the flags field, you
should not be impacted by my patch because
- need_active_balance is the only new place that use env->src_cpu in
the load_balance function
- and your machine will never test the condition: "env->src_cpu >
env->dst_cpu"
Damien,
Regarding your sched_domain config and especially the flags field, you
should not be impacted by my patch because
- need_active_balance is the only new place that use env-src_cpu in
the load_balance function
- and your machine will never test the condition: env-src_cpu
env-dst_cpu in
* Vincent Guittot vincent.guit...@linaro.org [2013-02-13 13:08]:
Damien,
Regarding your sched_domain config and especially the flags field, you
should not be impacted by my patch because
- need_active_balance is the only new place that use env-src_cpu in
the load_balance function
- and your
On 13 February 2013 15:08, Damien Wyart damien.wy...@gmail.com wrote:
* Vincent Guittot vincent.guit...@linaro.org [2013-02-13 13:08]:
Damien,
Regarding your sched_domain config and especially the flags field, you
should not be impacted by my patch because
- need_active_balance is the only
On 13 February 2013 15:28, Vincent Guittot vincent.guit...@linaro.org wrote:
On 13 February 2013 15:08, Damien Wyart damien.wy...@gmail.com wrote:
* Vincent Guittot vincent.guit...@linaro.org [2013-02-13 13:08]:
Damien,
Regarding your sched_domain config and especially the flags field, you
* Vincent Guittot vincent.guit...@linaro.org [2013-02-13 18:49]:
I have look into Frederic's tree but i didn't find any reason that
could explain your problem. May be Frederic will have some ideas
I have also tested his branch with and without my patch and both
kernel are booting (on an ARM
Bingo, that was the problem in my setup: as the patch was applied
through a script with others, I had missed the error message about the
conflict (I have also another conflict which can be safely ignored so
the new one did not catch my eye)... So the patch was only
half-applied, and the final
On Tue, Feb 12, 2013 at 5:19 AM, Vincent Guittot
vincent.guit...@linaro.org wrote:
need_active_balance uses env-src_cpu which is set only if there is more
than 1 task on the run queue.
We must set the src_cpu field unconditionnally
otherwise the test env-src_cpu env-dst_cpu will always fail
Hi Damien,
Thanks for the test and the feedback.
Could you send me the sched_domain configuration of your machine with
the kernel that boots on your machine ?
It's available in /proc/sys/kernel/sched_domain/cpu*/
This should not have any impact on your machine but it looks like it have one.
Hi,
I tested this on top of 3.8-rc7 and this made the machine (x86_64, Core
i7 920) unable to boot (very early as nothing at all is displayed on
screen). Nothing in the kernel log (after booting with a working
kernel).
Double-checked by just backing out only this patch and this made the
machine
need_active_balance uses env->src_cpu which is set only if there is more
than 1 task on the run queue. We must set the src_cpu field unconditionnally
otherwise the test "env->src_cpu > env->dst_cpu" will always fail if there is
only 1 task on the run queue
Signed-off-by: Vincent Guittot
---
need_active_balance uses env-src_cpu which is set only if there is more
than 1 task on the run queue. We must set the src_cpu field unconditionnally
otherwise the test env-src_cpu env-dst_cpu will always fail if there is
only 1 task on the run queue
Signed-off-by: Vincent Guittot
Hi,
I tested this on top of 3.8-rc7 and this made the machine (x86_64, Core
i7 920) unable to boot (very early as nothing at all is displayed on
screen). Nothing in the kernel log (after booting with a working
kernel).
Double-checked by just backing out only this patch and this made the
machine
Hi Damien,
Thanks for the test and the feedback.
Could you send me the sched_domain configuration of your machine with
the kernel that boots on your machine ?
It's available in /proc/sys/kernel/sched_domain/cpu*/
This should not have any impact on your machine but it looks like it have one.
24 matches
Mail list logo