On Sun, Nov 13, 2016 at 12:53:28PM -0600, Christoph Lameter wrote:
> On Fri, 11 Nov 2016, Peter Zijlstra wrote:
>
> > On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> > > On Thu, 10 Nov 2016, Daniel Vacek wrote:
> > >
> > > > I believe Daniel's patches are the best thing we
On Sun, Nov 13, 2016 at 12:53:28PM -0600, Christoph Lameter wrote:
> On Fri, 11 Nov 2016, Peter Zijlstra wrote:
>
> > On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> > > On Thu, 10 Nov 2016, Daniel Vacek wrote:
> > >
> > > > I believe Daniel's patches are the best thing we
On Fri, 11 Nov 2016, Peter Zijlstra wrote:
> On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> > On Thu, 10 Nov 2016, Daniel Vacek wrote:
> >
> > > I believe Daniel's patches are the best thing we can do in current
> > > situation as the behavior now seems rather buggy and does
On Fri, 11 Nov 2016, Peter Zijlstra wrote:
> On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> > On Thu, 10 Nov 2016, Daniel Vacek wrote:
> >
> > > I believe Daniel's patches are the best thing we can do in current
> > > situation as the behavior now seems rather buggy and does
On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> On Thu, 10 Nov 2016, Daniel Vacek wrote:
>
> > I believe Daniel's patches are the best thing we can do in current
> > situation as the behavior now seems rather buggy and does not provide above
> > mentioned expectations set
On Fri, Nov 11, 2016 at 12:46:37PM -0600, Christoph Lameter wrote:
> On Thu, 10 Nov 2016, Daniel Vacek wrote:
>
> > I believe Daniel's patches are the best thing we can do in current
> > situation as the behavior now seems rather buggy and does not provide above
> > mentioned expectations set
On Thu, 10 Nov 2016, Daniel Vacek wrote:
> I believe Daniel's patches are the best thing we can do in current
> situation as the behavior now seems rather buggy and does not provide above
> mentioned expectations set when rt throttling was merged with default
> budget of 95% of CPU time. Nor if
On Thu, 10 Nov 2016, Daniel Vacek wrote:
> I believe Daniel's patches are the best thing we can do in current
> situation as the behavior now seems rather buggy and does not provide above
> mentioned expectations set when rt throttling was merged with default
> budget of 95% of CPU time. Nor if
On 11/08/2016 08:50 PM, Peter Zijlstra wrote:
>> The problem is that using RT_RUNTIME_SHARE a CPU will almost always
>> > borrow enough runtime to make a CPU intensive rt task to run forever...
>> > well not forever, but until the system crash because a kworker starved
>> > in this CPU. Kworkers
On 11/08/2016 08:50 PM, Peter Zijlstra wrote:
>> The problem is that using RT_RUNTIME_SHARE a CPU will almost always
>> > borrow enough runtime to make a CPU intensive rt task to run forever...
>> > well not forever, but until the system crash because a kworker starved
>> > in this CPU. Kworkers
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 21:06:50 +0100
> Daniel Bristot de Oliveira wrote:
>
> > The throttling allowed the kworker to run, but once the kworker went to
> > sleep, the RT tasks started to work again. In the previous behavior,
> > the
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 21:06:50 +0100
> Daniel Bristot de Oliveira wrote:
>
> > The throttling allowed the kworker to run, but once the kworker went to
> > sleep, the RT tasks started to work again. In the previous behavior,
> > the system would either go
On Tue, Nov 08, 2016 at 08:29:49PM +0100, Daniel Bristot de Oliveira wrote:
>
>
> On 11/08/2016 07:05 PM, Peter Zijlstra wrote:
> >> >
> >> > I know what we want to do, but there's some momentous problems that
> >> > need to be solved first.
> > Like what?
>
> The problem is that using
On Tue, Nov 08, 2016 at 08:29:49PM +0100, Daniel Bristot de Oliveira wrote:
>
>
> On 11/08/2016 07:05 PM, Peter Zijlstra wrote:
> >> >
> >> > I know what we want to do, but there's some momentous problems that
> >> > need to be solved first.
> > Like what?
>
> The problem is that using
On 11/08/2016 07:05 PM, Peter Zijlstra wrote:
>> >
>> > I know what we want to do, but there's some momentous problems that
>> > need to be solved first.
> Like what?
The problem is that using RT_RUNTIME_SHARE a CPU will almost always
borrow enough runtime to make a CPU intensive rt task to
On 11/08/2016 07:05 PM, Peter Zijlstra wrote:
>> >
>> > I know what we want to do, but there's some momentous problems that
>> > need to be solved first.
> Like what?
The problem is that using RT_RUNTIME_SHARE a CPU will almost always
borrow enough runtime to make a CPU intensive rt task to
On Tue, Nov 08, 2016 at 12:17:10PM -0500, Steven Rostedt wrote:
> On Tue, 8 Nov 2016 17:51:33 +0100
> Peter Zijlstra wrote:
>
> > You really should already know this.
>
> I know what we want to do, but there's some momentous problems that
> need to be solved first.
Like
On Tue, Nov 08, 2016 at 12:17:10PM -0500, Steven Rostedt wrote:
> On Tue, 8 Nov 2016 17:51:33 +0100
> Peter Zijlstra wrote:
>
> > You really should already know this.
>
> I know what we want to do, but there's some momentous problems that
> need to be solved first.
Like what?
> Until then, we
On Tue, 8 Nov 2016 17:51:33 +0100
Peter Zijlstra wrote:
> You really should already know this.
I know what we want to do, but there's some momentous problems that
need to be solved first. Until then, we may be forced to continue with
hacks.
>
> As stands the current rt
On Tue, 8 Nov 2016 17:51:33 +0100
Peter Zijlstra wrote:
> You really should already know this.
I know what we want to do, but there's some momentous problems that
need to be solved first. Until then, we may be forced to continue with
hacks.
>
> As stands the current rt cgroup code (and all
On Tue, Nov 08, 2016 at 09:07:40AM -0500, Steven Rostedt wrote:
> On Tue, 8 Nov 2016 12:59:58 +0100
> Peter Zijlstra wrote:
>
> > No, none of this stands a chance of being accepted.
> >
> > This is making bad code worse.
>
> Peter,
>
> Instead of a flat out rejection,
On Tue, Nov 08, 2016 at 09:07:40AM -0500, Steven Rostedt wrote:
> On Tue, 8 Nov 2016 12:59:58 +0100
> Peter Zijlstra wrote:
>
> > No, none of this stands a chance of being accepted.
> >
> > This is making bad code worse.
>
> Peter,
>
> Instead of a flat out rejection, can you please provide
On Tue, 8 Nov 2016 12:59:58 +0100
Peter Zijlstra wrote:
> No, none of this stands a chance of being accepted.
>
> This is making bad code worse.
Peter,
Instead of a flat out rejection, can you please provide some
constructive criticism to let those that are working on
On Tue, 8 Nov 2016 12:59:58 +0100
Peter Zijlstra wrote:
> No, none of this stands a chance of being accepted.
>
> This is making bad code worse.
Peter,
Instead of a flat out rejection, can you please provide some
constructive criticism to let those that are working on this know what
would be
No, none of this stands a chance of being accepted.
This is making bad code worse.
No, none of this stands a chance of being accepted.
This is making bad code worse.
Hi Daniel,
On 07/11/16 14:51, Daniel Bristot de Oliveira wrote:
> On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
[...]
> > -) only issue might be that, if a non-RT task wakes up after the
> > unthrottle, it will have to wait, but worst-case it will have a chance
> > in the next throttling
Hi Daniel,
On 07/11/16 14:51, Daniel Bristot de Oliveira wrote:
> On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
[...]
> > -) only issue might be that, if a non-RT task wakes up after the
> > unthrottle, it will have to wait, but worst-case it will have a chance
> > in the next throttling
Hi all,
since GRUB reclaiming has been mentioned, I am going to add some
comments on it :)
On Mon, 7 Nov 2016 14:51:37 +0100
Daniel Bristot de Oliveira wrote:
[...]
> The sum of allocated runtime for all DL tasks will not to be greater
> than RT throttling enforcement
Hi all,
since GRUB reclaiming has been mentioned, I am going to add some
comments on it :)
On Mon, 7 Nov 2016 14:51:37 +0100
Daniel Bristot de Oliveira wrote:
[...]
> The sum of allocated runtime for all DL tasks will not to be greater
> than RT throttling enforcement runtime. The DL scheduler
On Mon, 7 Nov 2016 21:33:02 +0100
Daniel Bristot de Oliveira wrote:
> On 11/07/2016 09:16 PM, Steven Rostedt wrote:
> > I'm confused? Are you saying that RR tasks don't get throttled in the
> > current code? That sounds like a bug to me.
>
> If the RT_RUNTIME_SHARING is
On Mon, 7 Nov 2016 21:33:02 +0100
Daniel Bristot de Oliveira wrote:
> On 11/07/2016 09:16 PM, Steven Rostedt wrote:
> > I'm confused? Are you saying that RR tasks don't get throttled in the
> > current code? That sounds like a bug to me.
>
> If the RT_RUNTIME_SHARING is enabled, the CPU in
On 11/07/2016 09:16 PM, Steven Rostedt wrote:
> I'm confused? Are you saying that RR tasks don't get throttled in the
> current code? That sounds like a bug to me.
If the RT_RUNTIME_SHARING is enabled, the CPU in which the RR tasks are
running (and pinned) will borrow RT runtime from another
On 11/07/2016 09:16 PM, Steven Rostedt wrote:
> I'm confused? Are you saying that RR tasks don't get throttled in the
> current code? That sounds like a bug to me.
If the RT_RUNTIME_SHARING is enabled, the CPU in which the RR tasks are
running (and pinned) will borrow RT runtime from another
On Mon, 7 Nov 2016 21:06:50 +0100
Daniel Bristot de Oliveira wrote:
> The throttling allowed the kworker to run, but once the kworker went to
> sleep, the RT tasks started to work again. In the previous behavior,
> the system would either go idle, or the kworker would starve
On Mon, 7 Nov 2016 21:06:50 +0100
Daniel Bristot de Oliveira wrote:
> The throttling allowed the kworker to run, but once the kworker went to
> sleep, the RT tasks started to work again. In the previous behavior,
> the system would either go idle, or the kworker would starve because
> the
On 11/07/2016 09:00 PM, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 13:54:12 -0600 (CST)
> Christoph Lameter wrote:
>
>> On Mon, 7 Nov 2016, Steven Rostedt wrote:
>>
>>> On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
>>> Christoph Lameter wrote:
>>>
SCHED_RR
On 11/07/2016 09:00 PM, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 13:54:12 -0600 (CST)
> Christoph Lameter wrote:
>
>> On Mon, 7 Nov 2016, Steven Rostedt wrote:
>>
>>> On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
>>> Christoph Lameter wrote:
>>>
SCHED_RR tasks alternately running on on cpu
On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
Christoph Lameter wrote:
> SCHED_RR tasks alternately running on on cpu can cause endless deferral of
> kworker threads. With the global effect of the OS processing reserved
> it may be the case that the processor we are executing never
On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
Christoph Lameter wrote:
> SCHED_RR tasks alternately running on on cpu can cause endless deferral of
> kworker threads. With the global effect of the OS processing reserved
> it may be the case that the processor we are executing never gets any
> time.
On Mon, 7 Nov 2016 13:54:12 -0600 (CST)
Christoph Lameter wrote:
> On Mon, 7 Nov 2016, Steven Rostedt wrote:
>
> > On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
> > Christoph Lameter wrote:
> >
> > > SCHED_RR tasks alternately running on on cpu can cause endless
On Mon, 7 Nov 2016 13:54:12 -0600 (CST)
Christoph Lameter wrote:
> On Mon, 7 Nov 2016, Steven Rostedt wrote:
>
> > On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
> > Christoph Lameter wrote:
> >
> > > SCHED_RR tasks alternately running on on cpu can cause endless deferral of
> > > kworker threads.
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
> Christoph Lameter wrote:
>
> > SCHED_RR tasks alternately running on on cpu can cause endless deferral of
> > kworker threads. With the global effect of the OS processing reserved
> > it may be
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 13:30:15 -0600 (CST)
> Christoph Lameter wrote:
>
> > SCHED_RR tasks alternately running on on cpu can cause endless deferral of
> > kworker threads. With the global effect of the OS processing reserved
> > it may be the case that
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 10:55:38 -0600 (CST)
> Christoph Lameter wrote:
>
> > On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
> >
> > > With these two options set, the user will guarantee some runtime
> > > for non-rt-tasks on all
On Mon, 7 Nov 2016, Steven Rostedt wrote:
> On Mon, 7 Nov 2016 10:55:38 -0600 (CST)
> Christoph Lameter wrote:
>
> > On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
> >
> > > With these two options set, the user will guarantee some runtime
> > > for non-rt-tasks on all CPUs, while keeping
On Mon, 7 Nov 2016 19:49:03 +0100
Daniel Bristot de Oliveira wrote:
> On 11/07/2016 07:32 PM, Steven Rostedt wrote:
> >> Excellent this would improve the situation with deadlocks as a result of
> >> > cgroup_locks not being released due to lack of workqueue processing.
> >
On Mon, 7 Nov 2016 19:49:03 +0100
Daniel Bristot de Oliveira wrote:
> On 11/07/2016 07:32 PM, Steven Rostedt wrote:
> >> Excellent this would improve the situation with deadlocks as a result of
> >> > cgroup_locks not being released due to lack of workqueue processing.
> > ?? What deadlocks
On 11/07/2016 07:32 PM, Steven Rostedt wrote:
>> Excellent this would improve the situation with deadlocks as a result of
>> > cgroup_locks not being released due to lack of workqueue processing.
> ?? What deadlocks do you see? I mean, can you show the situation that
> throttling RT tasks will
On 11/07/2016 07:32 PM, Steven Rostedt wrote:
>> Excellent this would improve the situation with deadlocks as a result of
>> > cgroup_locks not being released due to lack of workqueue processing.
> ?? What deadlocks do you see? I mean, can you show the situation that
> throttling RT tasks will
On Mon, 7 Nov 2016 13:30:46 -0500
Steven Rostedt wrote:
> On Mon, 7 Nov 2016 12:22:21 -0600
> Clark Williams wrote:
>
> > I'm still reviewing the patch, but I have to wonder why bother with making
> > it a scheduler feature?
> >
> > The SCHED_FIFO
On Mon, 7 Nov 2016 13:30:46 -0500
Steven Rostedt wrote:
> On Mon, 7 Nov 2016 12:22:21 -0600
> Clark Williams wrote:
>
> > I'm still reviewing the patch, but I have to wonder why bother with making
> > it a scheduler feature?
> >
> > The SCHED_FIFO definition allows a fifo thread to starve
On 11/07/2016 07:30 PM, Steven Rostedt wrote:
>> I'm still reviewing the patch, but I have to wonder why bother with making
>> it a scheduler feature?
>> >
>> > The SCHED_FIFO definition allows a fifo thread to starve others
>> > because a fifo task will run until it yields. Throttling was
On 11/07/2016 07:30 PM, Steven Rostedt wrote:
>> I'm still reviewing the patch, but I have to wonder why bother with making
>> it a scheduler feature?
>> >
>> > The SCHED_FIFO definition allows a fifo thread to starve others
>> > because a fifo task will run until it yields. Throttling was
On Mon, 7 Nov 2016 10:55:38 -0600 (CST)
Christoph Lameter wrote:
> On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
>
> > With these two options set, the user will guarantee some runtime
> > for non-rt-tasks on all CPUs, while keeping real-time tasks running
> > as much as
On Mon, 7 Nov 2016 10:55:38 -0600 (CST)
Christoph Lameter wrote:
> On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
>
> > With these two options set, the user will guarantee some runtime
> > for non-rt-tasks on all CPUs, while keeping real-time tasks running
> > as much as possible.
>
>
On Mon, 7 Nov 2016 12:22:21 -0600
Clark Williams wrote:
> I'm still reviewing the patch, but I have to wonder why bother with making it
> a scheduler feature?
>
> The SCHED_FIFO definition allows a fifo thread to starve others
> because a fifo task will run until it
On Mon, 7 Nov 2016 12:22:21 -0600
Clark Williams wrote:
> I'm still reviewing the patch, but I have to wonder why bother with making it
> a scheduler feature?
>
> The SCHED_FIFO definition allows a fifo thread to starve others
> because a fifo task will run until it yields. Throttling was
On Mon, 7 Nov 2016 19:03:08 +0100
Tommaso Cucinotta wrote:
> On 07/11/2016 14:51, Daniel Bristot de Oliveira wrote:
> > Hi Tommaso,
>
> Hi,
>
> I'm cc-ing Luca for GRUB et al., pls find a few further notes below...
Thanks Tommaso! I've seen the email on the
On Mon, 7 Nov 2016 19:03:08 +0100
Tommaso Cucinotta wrote:
> On 07/11/2016 14:51, Daniel Bristot de Oliveira wrote:
> > Hi Tommaso,
>
> Hi,
>
> I'm cc-ing Luca for GRUB et al., pls find a few further notes below...
Thanks Tommaso! I've seen the email on the linux-rt-users mailing list,
and
On Mon, 7 Nov 2016 09:17:55 +0100
Daniel Bristot de Oliveira wrote:
> The rt throttling mechanism prevents the starvation of non-real-time
> tasks by CPU intensive real-time tasks. In terms of percentage,
> the default behavior allows real-time tasks to run up to 95% of a
>
On Mon, 7 Nov 2016 09:17:55 +0100
Daniel Bristot de Oliveira wrote:
> The rt throttling mechanism prevents the starvation of non-real-time
> tasks by CPU intensive real-time tasks. In terms of percentage,
> the default behavior allows real-time tasks to run up to 95% of a
> given period,
On 07/11/2016 14:51, Daniel Bristot de Oliveira wrote:
Hi Tommaso,
Hi,
I'm cc-ing Luca for GRUB et al., pls find a few further notes below...
On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
as anticipated live to Daniel:
-) +1 for the general concept, we'd need something similar also for
On 07/11/2016 14:51, Daniel Bristot de Oliveira wrote:
Hi Tommaso,
Hi,
I'm cc-ing Luca for GRUB et al., pls find a few further notes below...
On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
as anticipated live to Daniel:
-) +1 for the general concept, we'd need something similar also for
On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
> With these two options set, the user will guarantee some runtime
> for non-rt-tasks on all CPUs, while keeping real-time tasks running
> as much as possible.
Excellent this would improve the situation with deadlocks as a result of
On Mon, 7 Nov 2016, Daniel Bristot de Oliveira wrote:
> With these two options set, the user will guarantee some runtime
> for non-rt-tasks on all CPUs, while keeping real-time tasks running
> as much as possible.
Excellent this would improve the situation with deadlocks as a result of
Hi Tommaso,
On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
> as anticipated live to Daniel:
> -) +1 for the general concept, we'd need something similar also for
> SCHED_DEADLINE
Resumed: the sum of the runtime of deadline tasks will not be greater
than the "to_ratio(global_rt_period(),
Hi Tommaso,
On 11/07/2016 11:31 AM, Tommaso Cucinotta wrote:
> as anticipated live to Daniel:
> -) +1 for the general concept, we'd need something similar also for
> SCHED_DEADLINE
Resumed: the sum of the runtime of deadline tasks will not be greater
than the "to_ratio(global_rt_period(),
as anticipated live to Daniel:
-) +1 for the general concept, we'd need something similar also for
SCHED_DEADLINE
-) only issue might be that, if a non-RT task wakes up after the unthrottle, it
will have to wait, but worst-case it will have a chance in the next throttling
window
-) an
as anticipated live to Daniel:
-) +1 for the general concept, we'd need something similar also for
SCHED_DEADLINE
-) only issue might be that, if a non-RT task wakes up after the unthrottle, it
will have to wait, but worst-case it will have a chance in the next throttling
window
-) an
The rt throttling mechanism prevents the starvation of non-real-time
tasks by CPU intensive real-time tasks. In terms of percentage,
the default behavior allows real-time tasks to run up to 95% of a
given period, leaving the other 5% of the period for non-real-time
tasks. In the absence of non-rt
The rt throttling mechanism prevents the starvation of non-real-time
tasks by CPU intensive real-time tasks. In terms of percentage,
the default behavior allows real-time tasks to run up to 95% of a
given period, leaving the other 5% of the period for non-real-time
tasks. In the absence of non-rt
72 matches
Mail list logo