Re: [PATCH] pm_qos: Add system bus performance parameter

[mark gross]

On Fri, Aug 27, 2010 at 07:31:46AM -0700, Kevin Hilman wrote:
> Saravana Kannan <skannan@xxxxxxxxxxxxxx> writes:
> 
> > Some drivers/devices might need some minimum system bus performance to
> > provide acceptable service. Provide a PM QoS parameter to send these requests
> > to.
> >
> > The new parameter is named "system bus performance" since it is generic enough
> > for the unit of the request to be frequency, bandwidth or something else that
> > might be appropriate. It's up to each implementation of the QoS provider to
> > define what the unit of the request would be.
> >
> > Signed-off-by: Saravana Kannan <skannan@xxxxxxxxxxxxxx>
> 
> With this current design, only one system-wide bus would be managed.
> What if a platform has more than one independently scalable bus?
> 
> I think the only scalable way to handle this kind of thing is to have
> per-device QoS constraints that can then be combined/aggregated by parent
> devices/busses.


I'm just realizing that Saravana may not be talking about a power
management problem.  I think she is talking about QoS, not constraining
the throttling of some platform bus or device. (i.e. not pm_qos)

I think its important to keep requirements clear. Are your ideas around
doing proper throttling constraints or are they about QoS proper?

I see a need for both but, as a QoS facility will likely call into
pm_qos to help grantee a qos it will be fundamentally different form
pm_qos.  

PM_QoS is a best effort API to constrain power state throttling.

So, what are you thinking about?  QoS or constraining power state
throttling?

--mark

> At LPC this year, I've proposed per-device QoS constraints[1] as a topic
> for the PM mini-conf.  I hope some folks from the MSM camp can be there
> for these discussions.
> 
> Kevin
> 
> [1] http://www.linuxplumbersconf.org/2010/ocw/proposals/819
> 
> > ---
> >  kernel/pm_qos_params.c |    9 +++++++++
> >  1 files changed, 9 insertions(+), 0 deletions(-)
> >
> > diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c
> > index 996a4de..1a44a67 100644
> > --- a/kernel/pm_qos_params.c
> > +++ b/kernel/pm_qos_params.c
> > @@ -93,12 +93,21 @@ static struct pm_qos_object network_throughput_pm_qos = {
> >  	.type = PM_QOS_MAX,
> >  };
> >  
> > +static BLOCKING_NOTIFIER_HEAD(system_bus_performance_notifier);
> > +static struct pm_qos_object system_bus_performance_pm_qos = {
> > +	.requests = PLIST_HEAD_INIT(system_bus_performance_pm_qos.requests, pm_qos_lock),
> > +	.notifiers = &system_bus_performance_notifier,
> > +	.name = "system_bus_performance",
> > +	.default_value = 0,
> > +	.type = PM_QOS_MAX,
> > +};
> >  
> >  static struct pm_qos_object *pm_qos_array[] = {
> >  	&null_pm_qos,
> >  	&cpu_dma_pm_qos,
> >  	&network_lat_pm_qos,
> >  	&network_throughput_pm_qos
> > +	&system_bus_performance_pm_qos
> >  };
> >  
> >  static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf,
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/