[v2,1/2] service: add component useful work attribute
Checks
Commit Message
This commit adds a new attribute which allows the service to indicate
if the previous iteration of work was "useful". Useful work here implies
forward progress was made.
Exposing this information via an attribute to the application allows
tracking of CPU cycles as being useful or not-useful, and a CPU load
estimate can be deduced from that information.
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
---
v2:
- Add experimental tag to new function.
---
lib/librte_eal/common/rte_service.c | 19 +++++++++++++++++++
lib/librte_eal/include/rte_service.h | 5 +++++
.../include/rte_service_component.h | 13 +++++++++++++
lib/librte_eal/rte_eal_version.map | 3 +++
4 files changed, 40 insertions(+)
Comments
Hello Harry,
On Mon, Sep 14, 2020 at 4:36 PM Harry van Haaren
<harry.van.haaren@intel.com> wrote:
>
> This commit adds a new attribute which allows the service to indicate
> if the previous iteration of work was "useful". Useful work here implies
> forward progress was made.
>
> Exposing this information via an attribute to the application allows
> tracking of CPU cycles as being useful or not-useful, and a CPU load
> estimate can be deduced from that information.
>
> Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
This is missing a unit test.
Getting others' opinion on this addition would be good too.
Thanks.
> -----Original Message-----
> From: David Marchand <david.marchand@redhat.com>
> Sent: Monday, September 21, 2020 3:49 PM
> To: Van Haaren, Harry <harry.van.haaren@intel.com>
> Cc: dev <dev@dpdk.org>; Lukasz Wojciechowski
> <l.wojciechow@partner.samsung.com>; Honnappa Nagarahalli
> <Honnappa.Nagarahalli@arm.com>
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> attribute
>
> Hello Harry,
>
> On Mon, Sep 14, 2020 at 4:36 PM Harry van Haaren
> <harry.van.haaren@intel.com> wrote:
> >
> > This commit adds a new attribute which allows the service to indicate
> > if the previous iteration of work was "useful". Useful work here implies
> > forward progress was made.
> >
> > Exposing this information via an attribute to the application allows
> > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > estimate can be deduced from that information.
> >
> > Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
>
> This is missing a unit test.
> Getting others' opinion on this addition would be good too.
Thanks for review, sure will add a unit test.
+CC Jerin and Mattias; Eventdev SW PMD usage, I'd like to indicate if there has
been "useful work" done last iteration. Do you see any easy improvements? Concerns?
Regards, -Harry
<snip>
>
> Hello Harry,
>
> On Mon, Sep 14, 2020 at 4:36 PM Harry van Haaren
> <harry.van.haaren@intel.com> wrote:
> >
> > This commit adds a new attribute which allows the service to indicate
> > if the previous iteration of work was "useful". Useful work here
> > implies forward progress was made.
> >
> > Exposing this information via an attribute to the application allows
> > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > estimate can be deduced from that information.
> >
> > Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
>
> This is missing a unit test.
> Getting others' opinion on this addition would be good too.
Yes, adding unit tests would be good.
>
> Thanks.
>
>
> --
> David Marchand
On 2020-09-14 16:37, Harry van Haaren wrote:
> This commit adds a new attribute which allows the service to indicate
> if the previous iteration of work was "useful". Useful work here implies
> forward progress was made.
> Exposing this information via an attribute to the application allows
> tracking of CPU cycles as being useful or not-useful, and a CPU load
> estimate can be deduced from that information.
How would that tracking be implemented? rte_service.c already keeps
track of the amount of busy cycles per service. Would it be possible to
reuse that mechanism to achieve the same goal?
We did some prototyping on dynamic load balancing for the service core
framework, and then we extended the API is such a way that the service
callback would return a bool indicating if forward progress was made, if
I recall correctly. Sampling these counters allowed for tracking load on
both a per-lcore and per-service basis.
> Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
>
> ---
>
> v2:
> - Add experimental tag to new function.
>
> ---
> lib/librte_eal/common/rte_service.c | 19 +++++++++++++++++++
> lib/librte_eal/include/rte_service.h | 5 +++++
> .../include/rte_service_component.h | 13 +++++++++++++
> lib/librte_eal/rte_eal_version.map | 3 +++
> 4 files changed, 40 insertions(+)
>
> diff --git a/lib/librte_eal/common/rte_service.c b/lib/librte_eal/common/rte_service.c
> index 6a0e0ff65d..f9792a138b 100644
> --- a/lib/librte_eal/common/rte_service.c
> +++ b/lib/librte_eal/common/rte_service.c
> @@ -58,6 +58,7 @@ struct rte_service_spec_impl {
> uint32_t num_mapped_cores;
> uint64_t calls;
> uint64_t cycles_spent;
> + uint8_t useful_work_last_iter;
> } __rte_cache_aligned;
>
> /* the internal values of a service core */
> @@ -293,6 +294,21 @@ rte_service_component_unregister(uint32_t id)
> return 0;
> }
>
> +int32_t
> +rte_service_component_attr_set(uint32_t id, uint32_t attr, uint64_t value)
> +{
> + struct rte_service_spec_impl *s;
> + SERVICE_VALID_GET_OR_ERR_RET(id, s, -EINVAL);
> +
> + switch (attr) {
> + case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
> + s->useful_work_last_iter = value;
> + return 0;
> + default:
> + return -EINVAL;
> + };
> +}
> +
> int32_t
> rte_service_component_runstate_set(uint32_t id, uint32_t runstate)
> {
> @@ -778,6 +794,9 @@ rte_service_attr_get(uint32_t id, uint32_t attr_id, uint64_t *attr_value)
> return -EINVAL;
>
> switch (attr_id) {
> + case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
> + *attr_value = s->useful_work_last_iter;
> + return 0;
> case RTE_SERVICE_ATTR_CYCLES:
> *attr_value = s->cycles_spent;
> return 0;
> diff --git a/lib/librte_eal/include/rte_service.h b/lib/librte_eal/include/rte_service.h
> index e2d0a6dd32..e9836a1a68 100644
> --- a/lib/librte_eal/include/rte_service.h
> +++ b/lib/librte_eal/include/rte_service.h
> @@ -370,6 +370,11 @@ int32_t rte_service_dump(FILE *f, uint32_t id);
> */
> #define RTE_SERVICE_ATTR_CALL_COUNT 1
>
> +/**
> + * Returns if the last iteration of the service resulted in useful work done.
> + */
> +#define RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER 2
> +
> /**
> * Get an attribute from a service.
> *
> diff --git a/lib/librte_eal/include/rte_service_component.h b/lib/librte_eal/include/rte_service_component.h
> index 9e66ee7e29..534f41f531 100644
> --- a/lib/librte_eal/include/rte_service_component.h
> +++ b/lib/librte_eal/include/rte_service_component.h
> @@ -87,6 +87,19 @@ int32_t rte_service_component_register(const struct rte_service_spec *spec,
> */
> int32_t rte_service_component_unregister(uint32_t id);
>
> +/**
> + * Set an attribute for this service.
> + *
> + * Note this API is to be called by the service implementation, to make the
> + * statistic available via the usual attr_get() service APIs.
> + *
> + * @retval 0 Success
> + * @retval -EINVAL Invalid service id or attribute provided
> + */
> +__rte_experimental
> +int32_t rte_service_component_attr_set(uint32_t id, uint32_t attr,
> + uint64_t value);
> +
> /**
> * Private function to allow EAL to initialized default mappings.
> *
> diff --git a/lib/librte_eal/rte_eal_version.map b/lib/librte_eal/rte_eal_version.map
> index 0b18e2ef85..bb5e19ae9e 100644
> --- a/lib/librte_eal/rte_eal_version.map
> +++ b/lib/librte_eal/rte_eal_version.map
> @@ -397,6 +397,9 @@ EXPERIMENTAL {
> rte_mp_disable;
> rte_thread_register;
> rte_thread_unregister;
> +
> + # added in 20.11
> + rte_service_component_attr_set;
> };
>
> INTERNAL {
> -----Original Message-----
> From: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> Sent: Wednesday, September 23, 2020 12:40 PM
> To: Van Haaren, Harry <harry.van.haaren@intel.com>; dev@dpdk.org
> Cc: david.marchand@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> attribute
Hi Mattias,
Thanks for taking time to review & prompt reply!
> On 2020-09-14 16:37, Harry van Haaren wrote:
> > This commit adds a new attribute which allows the service to indicate
> > if the previous iteration of work was "useful". Useful work here implies
> > forward progress was made.
>
> > Exposing this information via an attribute to the application allows
> > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > estimate can be deduced from that information.
>
>
> How would that tracking be implemented? rte_service.c already keeps
> track of the amount of busy cycles per service. Would it be possible to
> reuse that mechanism to achieve the same goal?
Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
cycles polling, and trying to move packets around its internal queues, but make
no forward progress. Measuring cycles spent in the service would not indicate
the correct "busyness" in that case.
In the suggested patchset, each service (e.g Eventdev SW PMD) can update
a statistic itself, pushing an attribute value into the service-cores layer.
This method allows each PMD to define "useful work" in its own way.
> We did some prototyping on dynamic load balancing for the service core
> framework, and then we extended the API is such a way that the service
> callback would return a bool indicating if forward progress was made, if
> I recall correctly. Sampling these counters allowed for tracking load on
> both a per-lcore and per-service basis.
The service callback return value can be stored/inspected on the service-core
itself, but how to show that to the application? It still requires an attribute API
like proposed below re-using "attr_get" API I think.
So really the only difference in the prototype you mention is how the
service itself communicates business to the service-cores infrastructure in EAL.
Perhaps re-purposing return-value is simpler, but it limits statistics from the
service to just business, and the API change requires all services to change.
Pros of adding an API as this patchset proposes is to push attribute values to
service-core in EAL is extensibility, and no API breakage.
Given that context, Ack / push-back to this suggested approach?
> > Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
> >
> > ---
> >
> > v2:
> > - Add experimental tag to new function.
> >
> > ---
> > lib/librte_eal/common/rte_service.c | 19 +++++++++++++++++++
> > lib/librte_eal/include/rte_service.h | 5 +++++
> > .../include/rte_service_component.h | 13 +++++++++++++
> > lib/librte_eal/rte_eal_version.map | 3 +++
> > 4 files changed, 40 insertions(+)
> >
> > diff --git a/lib/librte_eal/common/rte_service.c
> b/lib/librte_eal/common/rte_service.c
> > index 6a0e0ff65d..f9792a138b 100644
> > --- a/lib/librte_eal/common/rte_service.c
> > +++ b/lib/librte_eal/common/rte_service.c
> > @@ -58,6 +58,7 @@ struct rte_service_spec_impl {
> > uint32_t num_mapped_cores;
> > uint64_t calls;
> > uint64_t cycles_spent;
> > + uint8_t useful_work_last_iter;
> > } __rte_cache_aligned;
> >
> > /* the internal values of a service core */
> > @@ -293,6 +294,21 @@ rte_service_component_unregister(uint32_t id)
> > return 0;
> > }
> >
> > +int32_t
> > +rte_service_component_attr_set(uint32_t id, uint32_t attr, uint64_t value)
> > +{
> > + struct rte_service_spec_impl *s;
> > + SERVICE_VALID_GET_OR_ERR_RET(id, s, -EINVAL);
> > +
> > + switch (attr) {
> > + case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
> > + s->useful_work_last_iter = value;
> > + return 0;
> > + default:
> > + return -EINVAL;
> > + };
> > +}
> > +
> > int32_t
> > rte_service_component_runstate_set(uint32_t id, uint32_t runstate)
> > {
> > @@ -778,6 +794,9 @@ rte_service_attr_get(uint32_t id, uint32_t attr_id,
> uint64_t *attr_value)
> > return -EINVAL;
> >
> > switch (attr_id) {
> > + case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
> > + *attr_value = s->useful_work_last_iter;
> > + return 0;
> > case RTE_SERVICE_ATTR_CYCLES:
> > *attr_value = s->cycles_spent;
> > return 0;
> > diff --git a/lib/librte_eal/include/rte_service.h
> b/lib/librte_eal/include/rte_service.h
> > index e2d0a6dd32..e9836a1a68 100644
> > --- a/lib/librte_eal/include/rte_service.h
> > +++ b/lib/librte_eal/include/rte_service.h
> > @@ -370,6 +370,11 @@ int32_t rte_service_dump(FILE *f, uint32_t id);
> > */
> > #define RTE_SERVICE_ATTR_CALL_COUNT 1
> >
> > +/**
> > + * Returns if the last iteration of the service resulted in useful work done.
> > + */
> > +#define RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER 2
> > +
<snip>
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Van Haaren, Harry
> Sent: Wednesday, September 23, 2020 3:17 PM
> To: Mattias Rönnblom <mattias.ronnblom@ericsson.com>; dev@dpdk.org
> Cc: david.marchand@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> attribute
>
> > -----Original Message-----
> > From: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> > Sent: Wednesday, September 23, 2020 12:40 PM
> > To: Van Haaren, Harry <harry.van.haaren@intel.com>; dev@dpdk.org
> > Cc: david.marchand@dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> > attribute
>
> Hi Mattias,
>
> Thanks for taking time to review & prompt reply!
Ping Mattias - any input based on follow up discussion below?
> > On 2020-09-14 16:37, Harry van Haaren wrote:
> > > This commit adds a new attribute which allows the service to indicate
> > > if the previous iteration of work was "useful". Useful work here implies
> > > forward progress was made.
> >
> > > Exposing this information via an attribute to the application allows
> > > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > > estimate can be deduced from that information.
> >
> >
> > How would that tracking be implemented? rte_service.c already keeps
> > track of the amount of busy cycles per service. Would it be possible to
> > reuse that mechanism to achieve the same goal?
>
> Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
> cycles polling, and trying to move packets around its internal queues, but make
> no forward progress. Measuring cycles spent in the service would not indicate
> the correct "busyness" in that case.
>
> In the suggested patchset, each service (e.g Eventdev SW PMD) can update
> a statistic itself, pushing an attribute value into the service-cores layer.
> This method allows each PMD to define "useful work" in its own way.
>
> > We did some prototyping on dynamic load balancing for the service core
> > framework, and then we extended the API is such a way that the service
> > callback would return a bool indicating if forward progress was made, if
> > I recall correctly. Sampling these counters allowed for tracking load on
> > both a per-lcore and per-service basis.
>
> The service callback return value can be stored/inspected on the service-core
> itself, but how to show that to the application? It still requires an attribute API
> like proposed below re-using "attr_get" API I think.
>
> So really the only difference in the prototype you mention is how the
> service itself communicates business to the service-cores infrastructure in EAL.
>
> Perhaps re-purposing return-value is simpler, but it limits statistics from the
> service to just business, and the API change requires all services to change.
>
> Pros of adding an API as this patchset proposes is to push attribute values to
> service-core in EAL is extensibility, and no API breakage.
>
> Given that context, Ack / push-back to this suggested approach?
<snip patch contents>
On Mon, Oct 5, 2020 at 8:59 PM Van Haaren, Harry
<harry.van.haaren@intel.com> wrote:
> > -----Original Message-----
> > From: dev <dev-bounces@dpdk.org> On Behalf Of Van Haaren, Harry
> > Sent: Wednesday, September 23, 2020 3:17 PM
> > To: Mattias Rönnblom <mattias.ronnblom@ericsson.com>; dev@dpdk.org
> > Cc: david.marchand@dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> > attribute
> >
> > > -----Original Message-----
> > > From: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> > > Sent: Wednesday, September 23, 2020 12:40 PM
> > > To: Van Haaren, Harry <harry.van.haaren@intel.com>; dev@dpdk.org
> > > Cc: david.marchand@dpdk.org
> > > Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> > > attribute
> >
> > Hi Mattias,
> >
> > Thanks for taking time to review & prompt reply!
>
> Ping Mattias - any input based on follow up discussion below?
>
>
> > > On 2020-09-14 16:37, Harry van Haaren wrote:
> > > > This commit adds a new attribute which allows the service to indicate
> > > > if the previous iteration of work was "useful". Useful work here implies
> > > > forward progress was made.
> > >
> > > > Exposing this information via an attribute to the application allows
> > > > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > > > estimate can be deduced from that information.
> > >
> > >
> > > How would that tracking be implemented? rte_service.c already keeps
> > > track of the amount of busy cycles per service. Would it be possible to
> > > reuse that mechanism to achieve the same goal?
> >
> > Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
> > cycles polling, and trying to move packets around its internal queues, but make
> > no forward progress. Measuring cycles spent in the service would not indicate
> > the correct "busyness" in that case.
> >
> > In the suggested patchset, each service (e.g Eventdev SW PMD) can update
> > a statistic itself, pushing an attribute value into the service-cores layer.
> > This method allows each PMD to define "useful work" in its own way.
> >
> > > We did some prototyping on dynamic load balancing for the service core
> > > framework, and then we extended the API is such a way that the service
> > > callback would return a bool indicating if forward progress was made, if
> > > I recall correctly. Sampling these counters allowed for tracking load on
> > > both a per-lcore and per-service basis.
> >
> > The service callback return value can be stored/inspected on the service-core
> > itself, but how to show that to the application? It still requires an attribute API
> > like proposed below re-using "attr_get" API I think.
> >
> > So really the only difference in the prototype you mention is how the
> > service itself communicates business to the service-cores infrastructure in EAL.
> >
> > Perhaps re-purposing return-value is simpler, but it limits statistics from the
> > service to just business, and the API change requires all services to change.
> >
> > Pros of adding an API as this patchset proposes is to push attribute values to
> > service-core in EAL is extensibility, and no API breakage.
> >
> > Given that context, Ack / push-back to this suggested approach?
I need a conclusion.
Is this required for 20.11?
> -----Original Message-----
> From: David Marchand <david.marchand@redhat.com>
> Sent: Friday, October 16, 2020 1:20 PM
> To: Van Haaren, Harry <harry.van.haaren@intel.com>; Mattias Rönnblom
> <mattias.ronnblom@ericsson.com>
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> attribute
>
> On Mon, Oct 5, 2020 at 8:59 PM Van Haaren, Harry
> <harry.van.haaren@intel.com> wrote:
> > > -----Original Message-----
> > > From: dev <dev-bounces@dpdk.org> On Behalf Of Van Haaren, Harry
> > > Sent: Wednesday, September 23, 2020 3:17 PM
> > > To: Mattias Rönnblom <mattias.ronnblom@ericsson.com>; dev@dpdk.org
> > > Cc: david.marchand@dpdk.org
> > > Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> > > attribute
> > >
> > > > -----Original Message-----
> > > > From: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> > > > Sent: Wednesday, September 23, 2020 12:40 PM
> > > > To: Van Haaren, Harry <harry.van.haaren@intel.com>; dev@dpdk.org
> > > > Cc: david.marchand@dpdk.org
> > > > Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful
> work
> > > > attribute
> > >
> > > Hi Mattias,
> > >
> > > Thanks for taking time to review & prompt reply!
> >
> > Ping Mattias - any input based on follow up discussion below?
> >
> >
> > > > On 2020-09-14 16:37, Harry van Haaren wrote:
> > > > > This commit adds a new attribute which allows the service to indicate
> > > > > if the previous iteration of work was "useful". Useful work here implies
> > > > > forward progress was made.
> > > >
> > > > > Exposing this information via an attribute to the application allows
> > > > > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > > > > estimate can be deduced from that information.
> > > >
> > > >
> > > > How would that tracking be implemented? rte_service.c already keeps
> > > > track of the amount of busy cycles per service. Would it be possible to
> > > > reuse that mechanism to achieve the same goal?
> > >
> > > Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
> > > cycles polling, and trying to move packets around its internal queues, but
> make
> > > no forward progress. Measuring cycles spent in the service would not indicate
> > > the correct "busyness" in that case.
> > >
> > > In the suggested patchset, each service (e.g Eventdev SW PMD) can update
> > > a statistic itself, pushing an attribute value into the service-cores layer.
> > > This method allows each PMD to define "useful work" in its own way.
> > >
> > > > We did some prototyping on dynamic load balancing for the service core
> > > > framework, and then we extended the API is such a way that the service
> > > > callback would return a bool indicating if forward progress was made, if
> > > > I recall correctly. Sampling these counters allowed for tracking load on
> > > > both a per-lcore and per-service basis.
> > >
> > > The service callback return value can be stored/inspected on the service-core
> > > itself, but how to show that to the application? It still requires an attribute API
> > > like proposed below re-using "attr_get" API I think.
> > >
> > > So really the only difference in the prototype you mention is how the
> > > service itself communicates business to the service-cores infrastructure in
> EAL.
> > >
> > > Perhaps re-purposing return-value is simpler, but it limits statistics from the
> > > service to just business, and the API change requires all services to change.
> > >
> > > Pros of adding an API as this patchset proposes is to push attribute values to
> > > service-core in EAL is extensibility, and no API breakage.
> > >
> > > Given that context, Ack / push-back to this suggested approach?
>
> I need a conclusion.
> Is this required for 20.11?
Given timeline - lets leave this until 21.02 release.
I think the above solution is adequate, but don't want to rush folks.
Thanks for following up, chat next release. -Harry
Hello Harry, Mattias,
On Fri, Oct 16, 2020 at 2:57 PM Van Haaren, Harry
<harry.van.haaren@intel.com> wrote:
> > > > > On 2020-09-14 16:37, Harry van Haaren wrote:
> > > > > > This commit adds a new attribute which allows the service to indicate
> > > > > > if the previous iteration of work was "useful". Useful work here implies
> > > > > > forward progress was made.
> > > > >
> > > > > > Exposing this information via an attribute to the application allows
> > > > > > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > > > > > estimate can be deduced from that information.
> > > > >
> > > > >
> > > > > How would that tracking be implemented? rte_service.c already keeps
> > > > > track of the amount of busy cycles per service. Would it be possible to
> > > > > reuse that mechanism to achieve the same goal?
> > > >
> > > > Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
> > > > cycles polling, and trying to move packets around its internal queues, but
> > make
> > > > no forward progress. Measuring cycles spent in the service would not indicate
> > > > the correct "busyness" in that case.
> > > >
> > > > In the suggested patchset, each service (e.g Eventdev SW PMD) can update
> > > > a statistic itself, pushing an attribute value into the service-cores layer.
> > > > This method allows each PMD to define "useful work" in its own way.
> > > >
> > > > > We did some prototyping on dynamic load balancing for the service core
> > > > > framework, and then we extended the API is such a way that the service
> > > > > callback would return a bool indicating if forward progress was made, if
> > > > > I recall correctly. Sampling these counters allowed for tracking load on
> > > > > both a per-lcore and per-service basis.
> > > >
> > > > The service callback return value can be stored/inspected on the service-core
> > > > itself, but how to show that to the application? It still requires an attribute API
> > > > like proposed below re-using "attr_get" API I think.
> > > >
> > > > So really the only difference in the prototype you mention is how the
> > > > service itself communicates business to the service-cores infrastructure in
> > EAL.
> > > >
> > > > Perhaps re-purposing return-value is simpler, but it limits statistics from the
> > > > service to just business, and the API change requires all services to change.
> > > >
> > > > Pros of adding an API as this patchset proposes is to push attribute values to
> > > > service-core in EAL is extensibility, and no API breakage.
> > > >
> > > > Given that context, Ack / push-back to this suggested approach?
> >
> > I need a conclusion.
> > Is this required for 20.11?
>
> Given timeline - lets leave this until 21.02 release.
> I think the above solution is adequate, but don't want to rush folks.
>
> Thanks for following up, chat next release. -Harry
>
Did some discussion happen?
21.02-rc1 is coming soon.
> -----Original Message-----
> From: David Marchand <david.marchand@redhat.com>
> Sent: Wednesday, January 13, 2021 2:43 PM
> To: Van Haaren, Harry <harry.van.haaren@intel.com>; mattias.ronnblom
> <mattias.ronnblom@ericsson.com>
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] service: add component useful work
> attribute
>
> Hello Harry, Mattias,
>
> On Fri, Oct 16, 2020 at 2:57 PM Van Haaren, Harry
> <harry.van.haaren@intel.com> wrote:
> > > > > > On 2020-09-14 16:37, Harry van Haaren wrote:
> > > > > > > This commit adds a new attribute which allows the service to indicate
> > > > > > > if the previous iteration of work was "useful". Useful work here implies
> > > > > > > forward progress was made.
> > > > > >
> > > > > > > Exposing this information via an attribute to the application allows
> > > > > > > tracking of CPU cycles as being useful or not-useful, and a CPU load
> > > > > > > estimate can be deduced from that information.
> > > > > >
> > > > > >
> > > > > > How would that tracking be implemented? rte_service.c already keeps
> > > > > > track of the amount of busy cycles per service. Would it be possible to
> > > > > > reuse that mechanism to achieve the same goal?
> > > > >
> > > > > Tracking "busy cycles" is not exactly the same - Eventdev SW PMD can spend
> > > > > cycles polling, and trying to move packets around its internal queues, but
> > > make
> > > > > no forward progress. Measuring cycles spent in the service would not indicate
> > > > > the correct "busyness" in that case.
> > > > >
> > > > > In the suggested patchset, each service (e.g Eventdev SW PMD) can update
> > > > > a statistic itself, pushing an attribute value into the service-cores layer.
> > > > > This method allows each PMD to define "useful work" in its own way.
> > > > >
> > > > > > We did some prototyping on dynamic load balancing for the service core
> > > > > > framework, and then we extended the API is such a way that the service
> > > > > > callback would return a bool indicating if forward progress was made, if
> > > > > > I recall correctly. Sampling these counters allowed for tracking load on
> > > > > > both a per-lcore and per-service basis.
> > > > >
> > > > > The service callback return value can be stored/inspected on the service-core
> > > > > itself, but how to show that to the application? It still requires an attribute
> API
> > > > > like proposed below re-using "attr_get" API I think.
> > > > >
> > > > > So really the only difference in the prototype you mention is how the
> > > > > service itself communicates business to the service-cores infrastructure in
> > > EAL.
> > > > >
> > > > > Perhaps re-purposing return-value is simpler, but it limits statistics from the
> > > > > service to just business, and the API change requires all services to change.
> > > > >
> > > > > Pros of adding an API as this patchset proposes is to push attribute values to
> > > > > service-core in EAL is extensibility, and no API breakage.
> > > > >
> > > > > Given that context, Ack / push-back to this suggested approach?
> > >
> > > I need a conclusion.
> > > Is this required for 20.11?
> >
> > Given timeline - lets leave this until 21.02 release.
> > I think the above solution is adequate, but don't want to rush folks.
> >
> > Thanks for following up, chat next release. -Harry
> >
>
> Did some discussion happen?
> 21.02-rc1 is coming soon.
Thanks for flagging, I'm not aware of any continuation of the discussions.
Mattias, do you have any updates?
This v2 currently fails to apply on a version.map file conflict, but otherwise is still fine IMO.
@@ -58,6 +58,7 @@ struct rte_service_spec_impl {
uint32_t num_mapped_cores;
uint64_t calls;
uint64_t cycles_spent;
+ uint8_t useful_work_last_iter;
} __rte_cache_aligned;
/* the internal values of a service core */
@@ -293,6 +294,21 @@ rte_service_component_unregister(uint32_t id)
return 0;
}
+int32_t
+rte_service_component_attr_set(uint32_t id, uint32_t attr, uint64_t value)
+{
+ struct rte_service_spec_impl *s;
+ SERVICE_VALID_GET_OR_ERR_RET(id, s, -EINVAL);
+
+ switch (attr) {
+ case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
+ s->useful_work_last_iter = value;
+ return 0;
+ default:
+ return -EINVAL;
+ };
+}
+
int32_t
rte_service_component_runstate_set(uint32_t id, uint32_t runstate)
{
@@ -778,6 +794,9 @@ rte_service_attr_get(uint32_t id, uint32_t attr_id, uint64_t *attr_value)
return -EINVAL;
switch (attr_id) {
+ case RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER:
+ *attr_value = s->useful_work_last_iter;
+ return 0;
case RTE_SERVICE_ATTR_CYCLES:
*attr_value = s->cycles_spent;
return 0;
@@ -370,6 +370,11 @@ int32_t rte_service_dump(FILE *f, uint32_t id);
*/
#define RTE_SERVICE_ATTR_CALL_COUNT 1
+/**
+ * Returns if the last iteration of the service resulted in useful work done.
+ */
+#define RTE_SERVICE_ATTR_USEFUL_WORK_LAST_ITER 2
+
/**
* Get an attribute from a service.
*
@@ -87,6 +87,19 @@ int32_t rte_service_component_register(const struct rte_service_spec *spec,
*/
int32_t rte_service_component_unregister(uint32_t id);
+/**
+ * Set an attribute for this service.
+ *
+ * Note this API is to be called by the service implementation, to make the
+ * statistic available via the usual attr_get() service APIs.
+ *
+ * @retval 0 Success
+ * @retval -EINVAL Invalid service id or attribute provided
+ */
+__rte_experimental
+int32_t rte_service_component_attr_set(uint32_t id, uint32_t attr,
+ uint64_t value);
+
/**
* Private function to allow EAL to initialized default mappings.
*
@@ -397,6 +397,9 @@ EXPERIMENTAL {
rte_mp_disable;
rte_thread_register;
rte_thread_unregister;
+
+ # added in 20.11
+ rte_service_component_attr_set;
};
INTERNAL {