Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 6042e4a9 authored by Linux Build Service Account's avatar Linux Build Service Account Committed by Gerrit - the friendly Code Review server
Browse files

Merge "ARM: dts: msm: Add QoS latency values for eMMC & SD on 8937"

parents fb2506c8 e6a93993

arch/arm/boot/dts/qcom/msm8937.dtsi

+14 −0
Original line number Diff line number Diff line
@@ -1477,6 +1477,14 @@ 


		qcom,devfreq,freq-table = <52000000 200000000>;



		qcom,pm-qos-irq-type = "affine_irq";

		qcom,pm-qos-irq-latency = <2 200>;



		qcom,pm-qos-cpu-groups = <0x0f 0xf0>;

		qcom,pm-qos-cmdq-latency-us = <2 200>, <2 200>;



		qcom,pm-qos-legacy-latency-us = <2 200>, <2 200>;



		qcom,msm-bus,name = "sdhc1";

		qcom,msm-bus,num-cases = <9>;

		qcom,msm-bus,num-paths = <1>;
@@ -1509,6 +1517,12 @@ 


		qcom,bus-width = <4>;



		qcom,pm-qos-irq-type = "affine_irq";

		qcom,pm-qos-irq-latency = <2 200>;



		qcom,pm-qos-cpu-groups = <0x0f 0xf0>;

		qcom,pm-qos-legacy-latency-us = <2 200>, <2 200>;



		qcom,msm-bus,name = "sdhc2";

		qcom,msm-bus,num-cases = <8>;

		qcom,msm-bus,num-paths = <1>;

drivers/mmc/host/sdhci-msm.c

+16 −10
Original line number Diff line number Diff line
@@ -43,6 +43,7 @@ 
#include "sdhci-msm-ice.h"

#include "cmdq_hci.h"



#define QOS_REMOVE_DELAY_MS	10

#define CORE_POWER		0x0

#define CORE_SW_RST		(1 << 7)
@@ -3226,7 +3227,8 @@ void sdhci_msm_reset_workaround(struct sdhci_host *host, u32 enable) 
static void sdhci_msm_pm_qos_irq_unvote_work(struct work_struct *work)

{

	struct sdhci_msm_pm_qos_irq *pm_qos_irq =

		container_of(work, struct sdhci_msm_pm_qos_irq, unvote_work);

		container_of(work, struct sdhci_msm_pm_qos_irq,

			     unvote_work.work);



	if (atomic_read(&pm_qos_irq->counter))

		return;
@@ -3252,7 +3254,7 @@ void sdhci_msm_pm_qos_irq_vote(struct sdhci_host *host) 
		&& counter > 1)

		return;



	cancel_work_sync(&msm_host->pm_qos_irq.unvote_work);

	cancel_delayed_work_sync(&msm_host->pm_qos_irq.unvote_work);

	msm_host->pm_qos_irq.latency = latency->latency[host->power_policy];

	pm_qos_update_request(&msm_host->pm_qos_irq.req,

				msm_host->pm_qos_irq.latency);
@@ -3278,7 +3280,8 @@ void sdhci_msm_pm_qos_irq_unvote(struct sdhci_host *host, bool async) 
		return;



	if (async) {

		schedule_work(&msm_host->pm_qos_irq.unvote_work);

		schedule_delayed_work(&msm_host->pm_qos_irq.unvote_work,

				      msecs_to_jiffies(QOS_REMOVE_DELAY_MS));

		return;

	}
@@ -3333,7 +3336,7 @@ sdhci_msm_pm_qos_irq_enable_store(struct device *dev, 


	msm_host->pm_qos_irq.enabled = enable;

	if (!enable) {

		cancel_work_sync(&msm_host->pm_qos_irq.unvote_work);

		cancel_delayed_work_sync(&msm_host->pm_qos_irq.unvote_work);

		atomic_set(&msm_host->pm_qos_irq.counter, 0);

		msm_host->pm_qos_irq.latency = PM_QOS_DEFAULT_VALUE;

		pm_qos_update_request(&msm_host->pm_qos_irq.req,
@@ -3379,7 +3382,7 @@ void sdhci_msm_pm_qos_irq_init(struct sdhci_host *host) 
		cpumask_copy(&msm_host->pm_qos_irq.req.cpus_affine,

			cpumask_of(msm_host->pdata->pm_qos_data.irq_cpu));



	INIT_WORK(&msm_host->pm_qos_irq.unvote_work,

	INIT_DELAYED_WORK(&msm_host->pm_qos_irq.unvote_work,

		sdhci_msm_pm_qos_irq_unvote_work);

	/* For initialization phase, set the performance latency */

	irq_latency = &msm_host->pdata->pm_qos_data.irq_latency;
@@ -3473,7 +3476,8 @@ static ssize_t sdhci_msm_pm_qos_group_enable_store(struct device *dev, 
	msm_host->pm_qos_group_enable = enable;

	if (!enable) {

		for (i = 0; i < nr_groups; i++) {

			cancel_work_sync(&msm_host->pm_qos[i].unvote_work);

			cancel_delayed_work_sync(

				&msm_host->pm_qos[i].unvote_work);

			atomic_set(&msm_host->pm_qos[i].counter, 0);

			msm_host->pm_qos[i].latency = PM_QOS_DEFAULT_VALUE;

			pm_qos_update_request(&msm_host->pm_qos[i].req,
@@ -3522,7 +3526,7 @@ void sdhci_msm_pm_qos_cpu_vote(struct sdhci_host *host, 
		&& counter > 1)

		return;



	cancel_work_sync(&pm_qos_group->unvote_work);

	cancel_delayed_work_sync(&pm_qos_group->unvote_work);



	pm_qos_group->latency = latency->latency[host->power_policy];

	pm_qos_update_request(&pm_qos_group->req, pm_qos_group->latency);
@@ -3531,7 +3535,8 @@ void sdhci_msm_pm_qos_cpu_vote(struct sdhci_host *host, 
static void sdhci_msm_pm_qos_cpu_unvote_work(struct work_struct *work)

{

	struct sdhci_msm_pm_qos_group *group =

		container_of(work, struct sdhci_msm_pm_qos_group, unvote_work);

		container_of(work, struct sdhci_msm_pm_qos_group,

			     unvote_work.work);



	if (atomic_read(&group->counter))

		return;
@@ -3551,7 +3556,8 @@ bool sdhci_msm_pm_qos_cpu_unvote(struct sdhci_host *host, int cpu, bool async) 
		return false;



	if (async) {

		schedule_work(&msm_host->pm_qos[group].unvote_work);

		schedule_delayed_work(&msm_host->pm_qos[group].unvote_work,

				      msecs_to_jiffies(QOS_REMOVE_DELAY_MS));

		return true;

	}
@@ -3581,7 +3587,7 @@ void sdhci_msm_pm_qos_cpu_init(struct sdhci_host *host, 


	for (i = 0; i < nr_groups; i++) {

		group = &msm_host->pm_qos[i];

		INIT_WORK(&group->unvote_work,

		INIT_DELAYED_WORK(&group->unvote_work,

			sdhci_msm_pm_qos_cpu_unvote_work);

		atomic_set(&group->counter, 0);

		group->req.type = PM_QOS_REQ_AFFINE_CORES;

drivers/mmc/host/sdhci-msm.h

+2 −2
Original line number Diff line number Diff line
@@ -111,7 +111,7 @@ struct sdhci_msm_pm_qos_data { 
 */

struct sdhci_msm_pm_qos_group {

	struct pm_qos_request req;

	struct work_struct unvote_work;

	struct delayed_work unvote_work;

	atomic_t counter;

	s32 latency;

};
@@ -119,7 +119,7 @@ struct sdhci_msm_pm_qos_group { 
/* PM QoS HW IRQ voting */

struct sdhci_msm_pm_qos_irq {

	struct pm_qos_request req;

	struct work_struct unvote_work;

	struct delayed_work unvote_work;

	struct device_attribute enable_attr;

	struct device_attribute status_attr;

	atomic_t counter;