| Message ID | 1408107129-29516-2-git-send-email-tomeu.vizoso@collabora.com (mailing list archive) |
|---|---|
| State | RFC, archived |
| Headers | show |
Hi! > Also adds a class type PM_QOS_SUM that aggregates the values by summing them. > > It can be used by memory controllers to calculate the optimum clock frequency > based on the bandwidth needs of the different memory clients. > > Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> > --- > Documentation/power/pm_qos_interface.txt | 4 +++- > include/linux/pm_qos.h | 5 ++++- > kernel/power/qos.c | 27 ++++++++++++++++++++++++++- > 3 files changed, 33 insertions(+), 3 deletions(-) > > diff --git a/Documentation/power/pm_qos_interface.txt b/Documentation/power/pm_qos_interface.txt > index a5da5c7..57782e8 100644 > --- a/Documentation/power/pm_qos_interface.txt > +++ b/Documentation/power/pm_qos_interface.txt > @@ -5,7 +5,8 @@ performance expectations by drivers, subsystems and user space applications on > one of the parameters. > > Two different PM QoS frameworks are available: > -1. PM QoS classes for cpu_dma_latency, network_latency, network_throughput. > +1. PM QoS classes for cpu_dma_latency, network_latency, network_throughput, > +memory_bandwidth. > 2. the per-device PM QoS framework provides the API to manage the per-device latency > constraints and PM QoS flags. > > @@ -13,6 +14,7 @@ Each parameters have defined units: > * latency: usec > * timeout: usec > * throughput: kbs (kilo bit / sec) > + * memory bandwidth: kbs (kilo bit / sec) Would mega bits per second make sense here? I suppose some many-core systems would have memory bandwith in > 10 terabit/sec, overflowing u32. Plus, if driver in 3/3 is just an example, perhaps comment should explain that clearly? Otherwise looks good, Acked-by: Pavel Machek <pavel@ucw.cz> -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 16 August 2014 08:35, Pavel Machek <pavel@ucw.cz> wrote: >> >> @@ -13,6 +14,7 @@ Each parameters have defined units: >> * latency: usec >> * timeout: usec >> * throughput: kbs (kilo bit / sec) >> + * memory bandwidth: kbs (kilo bit / sec) > > Would mega bits per second make sense here? > > I suppose some many-core systems would have memory bandwith > in > 10 terabit/sec, overflowing u32. Yeah, I was a bit concerned about excessive rounding increasing the chances of going to a higher frequency step than strictly necessary, but after thinking a bit more about it, I think it won't be a problem because it would be a bad idea anyway to only take into account explicit bandwidth requirements when calculating the optimal memory bus frequency. So I agree that a coarser unit such as mbps should be more appropriate here. > Plus, if driver in 3/3 is just an example, perhaps comment > should explain that clearly? Will do. > Otherwise looks good, > > Acked-by: Pavel Machek <pavel@ucw.cz> Thanks, Tomeu > -- > To unsubscribe from this list: send the line "unsubscribe linux-pm" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/Documentation/power/pm_qos_interface.txt b/Documentation/power/pm_qos_interface.txt index a5da5c7..57782e8 100644 --- a/Documentation/power/pm_qos_interface.txt +++ b/Documentation/power/pm_qos_interface.txt @@ -5,7 +5,8 @@ performance expectations by drivers, subsystems and user space applications on one of the parameters. Two different PM QoS frameworks are available: -1. PM QoS classes for cpu_dma_latency, network_latency, network_throughput. +1. PM QoS classes for cpu_dma_latency, network_latency, network_throughput, +memory_bandwidth. 2. the per-device PM QoS framework provides the API to manage the per-device latency constraints and PM QoS flags. @@ -13,6 +14,7 @@ Each parameters have defined units: * latency: usec * timeout: usec * throughput: kbs (kilo bit / sec) + * memory bandwidth: kbs (kilo bit / sec) 1. PM QoS framework diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h index 9ab4bf7..636e828 100644 --- a/include/linux/pm_qos.h +++ b/include/linux/pm_qos.h @@ -15,6 +15,7 @@ enum { PM_QOS_CPU_DMA_LATENCY, PM_QOS_NETWORK_LATENCY, PM_QOS_NETWORK_THROUGHPUT, + PM_QOS_MEMORY_BANDWIDTH, /* insert new class ID */ PM_QOS_NUM_CLASSES, @@ -32,6 +33,7 @@ enum pm_qos_flags_status { #define PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) #define PM_QOS_NETWORK_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) #define PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE 0 +#define PM_QOS_MEMORY_BANDWIDTH_DEFAULT_VALUE 0 #define PM_QOS_RESUME_LATENCY_DEFAULT_VALUE 0 #define PM_QOS_LATENCY_TOLERANCE_DEFAULT_VALUE 0 #define PM_QOS_LATENCY_TOLERANCE_NO_CONSTRAINT (-1) @@ -69,7 +71,8 @@ struct dev_pm_qos_request { enum pm_qos_type { PM_QOS_UNITIALIZED, PM_QOS_MAX, /* return the largest value */ - PM_QOS_MIN /* return the smallest value */ + PM_QOS_MIN, /* return the smallest value */ + PM_QOS_SUM /* return the sum */ }; /* diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 884b770..5f4c006 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -105,11 +105,27 @@ static struct pm_qos_object network_throughput_pm_qos = { }; +static BLOCKING_NOTIFIER_HEAD(memory_bandwidth_notifier); +static struct pm_qos_constraints memory_bw_constraints = { + .list = PLIST_HEAD_INIT(memory_bw_constraints.list), + .target_value = PM_QOS_MEMORY_BANDWIDTH_DEFAULT_VALUE, + .default_value = PM_QOS_MEMORY_BANDWIDTH_DEFAULT_VALUE, + .no_constraint_value = PM_QOS_MEMORY_BANDWIDTH_DEFAULT_VALUE, + .type = PM_QOS_SUM, + .notifiers = &memory_bandwidth_notifier, +}; +static struct pm_qos_object memory_bandwidth_pm_qos = { + .constraints = &memory_bw_constraints, + .name = "memory_bandwidth", +}; + + static struct pm_qos_object *pm_qos_array[] = { &null_pm_qos, &cpu_dma_pm_qos, &network_lat_pm_qos, - &network_throughput_pm_qos + &network_throughput_pm_qos, + &memory_bandwidth_pm_qos, }; static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, @@ -130,6 +146,9 @@ static const struct file_operations pm_qos_power_fops = { /* unlocked internal variant */ static inline int pm_qos_get_value(struct pm_qos_constraints *c) { + struct plist_node *node; + int total_value = 0; + if (plist_head_empty(&c->list)) return c->no_constraint_value; @@ -140,6 +159,12 @@ static inline int pm_qos_get_value(struct pm_qos_constraints *c) case PM_QOS_MAX: return plist_last(&c->list)->prio; + case PM_QOS_SUM: + plist_for_each(node, &c->list) + total_value += node->prio; + + return total_value; + default: /* runtime check for not using enum */ BUG();
Also adds a class type PM_QOS_SUM that aggregates the values by summing them. It can be used by memory controllers to calculate the optimum clock frequency based on the bandwidth needs of the different memory clients. Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> --- Documentation/power/pm_qos_interface.txt | 4 +++- include/linux/pm_qos.h | 5 ++++- kernel/power/qos.c | 27 ++++++++++++++++++++++++++- 3 files changed, 33 insertions(+), 3 deletions(-)