Hi Volodymir,
On 20/01/2021 23:03, Volodymyr Babchuk wrote:
Julien Grall writes:
This is very interestingi too. Did you get any spikes with the
period
set to 100us? It would be fantastic if there were none.
3. Huge latency spike during domain creation. I conducted some
additional
On Thu, 2021-01-21 at 01:20 +, Volodymyr Babchuk wrote:
> Hi Dario,
>
Hi :-)
> Dario Faggioli writes:
> > Anyway, as I was saying, having a latency which is ~ 2x of your
> > period
> > is ok, and it should be expected (when you size the period). In
> > fact,
> > let's say that your budget is
Hi Andrew,
Andrew Cooper writes:
> On 12/01/2021 23:48, Volodymyr Babchuk wrote:
>> Hello community,
>>
>> Recently I was experimenting with context switching latencies in
>> Xen. I want to share my findings and discuss some issues I have found.
>>
>> My target was to ensure that no more 30us
Hi Dario,
Dario Faggioli writes:
> On Thu, 2021-01-14 at 15:33 -0800, Stefano Stabellini wrote:
>> On Tue, 12 Jan 2021, Volodymyr Babchuk wrote:
>> > 2. RTDS scheduler. With console disabled, things like "hexdump -v
>> >/dev/zero" didn't affected the latency so badly, but anyways,
>> >
On Thu, 21 Jan 2021, Volodymyr Babchuk wrote:
> > On Fri, 15 Jan 2021, Volodymyr Babchuk wrote:
> [...]
> >> >> [ 83.873162] rt_eval_tmu e6fc.tmu: Mean: 44 (1320 ns) stddev: 8
> >> >> (240 ns)
> >> >> [ 94.136632] rt_eval_tmu e6fc.tmu: Mean: 44 (1320 ns) stddev: 8
> >> >> (240 ns)
>
On Wed, 20 Jan 2021, Volodymyr Babchuk wrote:
> > On 15/01/2021 15:45, Volodymyr Babchuk wrote:
> >> Hi Julien,
> >> Julien Grall writes:
> >>
> >>> Hi Volodymyr, Stefano,
> >>>
> >>> On 14/01/2021 23:33, Stefano Stabellini wrote:
> + Bertrand, Andrew (see comment on alloc_heap_pages())
>
Stefano Stabellini writes:
> On Fri, 15 Jan 2021, Volodymyr Babchuk wrote:
[...]
>> >> [ 83.873162] rt_eval_tmu e6fc.tmu: Mean: 44 (1320 ns) stddev: 8
>> >> (240 ns)
>> >> [ 94.136632] rt_eval_tmu e6fc.tmu: Mean: 44 (1320 ns) stddev: 8
>> >> (240 ns)
>> >> [ 104.400098]
Hi Stefano,
Stefano Stabellini writes:
> On Fri, 15 Jan 2021, Julien Grall wrote:
>> On 15/01/2021 15:45, Volodymyr Babchuk wrote:
>> >
>> > Hi Julien,
>> >
>> > Julien Grall writes:
>> >
>> > > Hi Volodymyr, Stefano,
>> > >
>> > > On 14/01/2021 23:33, Stefano Stabellini wrote:
>> > > > +
Hi Julien,
Julien Grall writes:
> On 15/01/2021 15:45, Volodymyr Babchuk wrote:
>> Hi Julien,
>> Julien Grall writes:
>>
>>> Hi Volodymyr, Stefano,
>>>
>>> On 14/01/2021 23:33, Stefano Stabellini wrote:
+ Bertrand, Andrew (see comment on alloc_heap_pages())
>>>
>>> Long running
On Thu, 2021-01-14 at 15:33 -0800, Stefano Stabellini wrote:
> On Tue, 12 Jan 2021, Volodymyr Babchuk wrote:
> > 2. RTDS scheduler. With console disabled, things like "hexdump -v
> > /dev/zero" didn't affected the latency so badly, but anyways,
> > sometimes I got ~600us spikes. This is not
On 12/01/2021 23:48, Volodymyr Babchuk wrote:
> Hello community,
>
> Recently I was experimenting with context switching latencies in
> Xen. I want to share my findings and discuss some issues I have found.
>
> My target was to ensure that no more 30us passes between interrupt
> generation and
On 15/01/2021 23:41, Stefano Stabellini wrote:
> This is very interestingi too. Did you get any spikes with the
> period
> set to 100us? It would be fantastic if there were none.
>
>> 3. Huge latency spike during domain creation. I conducted some
>> additional tests,
Hi Stefano,
On 15/01/2021 23:17, Stefano Stabellini wrote:
On Fri, 15 Jan 2021, Volodymyr Babchuk wrote:
9us was in idle state. Interestingly enough, I got latency if 3us while
Dom0 was doing some CPU-intensive tasks. So, under load latency is lower
than in idle state. I didn't investigated
On Fri, 15 Jan 2021, Julien Grall wrote:
> On 15/01/2021 15:45, Volodymyr Babchuk wrote:
> >
> > Hi Julien,
> >
> > Julien Grall writes:
> >
> > > Hi Volodymyr, Stefano,
> > >
> > > On 14/01/2021 23:33, Stefano Stabellini wrote:
> > > > + Bertrand, Andrew (see comment on alloc_heap_pages())
>
On Fri, 15 Jan 2021, Volodymyr Babchuk wrote:
> >> ARMv8 platform. Namely Renesas Rcar H3 SoC on Salvator board.
> >>
> >> To accurately determine latency, I employed one of timer counter units
> >> (TMUs) available on the SoC. This is 32-bit timer with auto-reload,
> >> that can generate
On 15/01/2021 15:45, Volodymyr Babchuk wrote:
Hi Julien,
Julien Grall writes:
Hi Volodymyr, Stefano,
On 14/01/2021 23:33, Stefano Stabellini wrote:
+ Bertrand, Andrew (see comment on alloc_heap_pages())
Long running hypercalls are usually considered security issues.
In this case,
Hi Julien,
Julien Grall writes:
> Hi Volodymyr, Stefano,
>
> On 14/01/2021 23:33, Stefano Stabellini wrote:
>> + Bertrand, Andrew (see comment on alloc_heap_pages())
>
> Long running hypercalls are usually considered security issues.
>
> In this case, only the control domain can issue large
Hi Stefano,
Stefano Stabellini writes:
[...]
>>
>> ARMv8 platform. Namely Renesas Rcar H3 SoC on Salvator board.
>>
>> To accurately determine latency, I employed one of timer counter units
>> (TMUs) available on the SoC. This is 32-bit timer with auto-reload,
>> that can generate interrupt
Hi Volodymyr, Stefano,
On 14/01/2021 23:33, Stefano Stabellini wrote:
+ Bertrand, Andrew (see comment on alloc_heap_pages())
Long running hypercalls are usually considered security issues.
In this case, only the control domain can issue large memory allocation
(2GB at a time). Guest, would
+ Bertrand, Andrew (see comment on alloc_heap_pages())
Really great work, Volodymyr! Some comments below.
On Tue, 12 Jan 2021, Volodymyr Babchuk wrote:
> Hello community,
>
> Recently I was experimenting with context switching latencies in
> Xen. I want to share my findings and discuss some
Hello community,
Recently I was experimenting with context switching latencies in
Xen. I want to share my findings and discuss some issues I have found.
My target was to ensure that no more 30us passes between interrupt
generation and handling in a guest VM. Main goal was to determine if
Xen
21 matches
Mail list logo