All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lina Iyer <lina.iyer@linaro.org>
To: ulf.hansson@linaro.org, khilman@kernel.org, rjw@rjwysocki.net,
	linux-pm@vger.kernel.org, linux-arm-kernel@lists.infradead.org
Cc: lorenzo.pieralisi@arm.com, Juri.Lelli@arm.com,
	linux-arm-msm@vger.kernel.org, sboyd@codeaurora.org,
	brendan.jackman@arm.com, sudeep.holla@arm.com,
	andy.gross@linaro.org, Lina Iyer <lina.iyer@linaro.org>
Subject: [PATCH V5 1/6] drivers: cpuidle: Read CPU's idle state from PM domain
Date: Fri,  3 Mar 2017 13:48:12 -0800	[thread overview]
Message-ID: <1488577697-127445-2-git-send-email-lina.iyer@linaro.org> (raw)
In-Reply-To: <1488577697-127445-1-git-send-email-lina.iyer@linaro.org>

Currently CPUs idle states and idle states of its parent are represented
in a flattened model by the cpu-dile-states property of the CPU node.
The CPUs idle states are followed by its cluster idle states. With the
introduction of CPU PM domains, the CPUs and domain idle states may be
represented hierarchically as part of the domain DT definition. This
would mean presenting idle state information in 2 places - CPU nodes for
the CPU and the cluster's with the PM domains.

Also, it makes sense to define domains around each individual CPU since
each of them is a power domain in its own right. The CPU idle states can
now be represented as its domain's idle state, defined by the
domain-idle-states property. This avoids presenting idle states in
multiple places in the DT.

Modify the DT-based cpuidle driver to check for the presence of a CPU's
domain and if present read the domain-idle-states of the PM domain and
if the CPU's domain is absent, revert to reading in the cpu-idle-states
property of the CPU DT node.

Suggested-by: Sudeep Holla <sudeep.holla@qrm.com>
Signed-off-by: Lina Iyer <lina.iyer@linaro.org>
---
 drivers/cpuidle/dt_idle_states.c | 38 +++++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/cpuidle/dt_idle_states.c b/drivers/cpuidle/dt_idle_states.c
index ffca4fc..4df7d45 100644
--- a/drivers/cpuidle/dt_idle_states.c
+++ b/drivers/cpuidle/dt_idle_states.c
@@ -98,6 +98,39 @@ static int init_state_node(struct cpuidle_state *idle_state,
 }
 
 /*
+ * Get the state node at @idx. State node may be defined as domain's idle state
+ * if the CPU has its own domain or defined as CPU's idle state if it doesn't
+ * have a domain provider.
+ */
+static struct device_node *get_state_node(struct device_node *cpu_node,
+				unsigned int idx)
+{
+	struct device_node *dn;
+	bool cpu_has_domain = false;
+	struct of_phandle_args args;
+	const char *property;
+	int err;
+
+	err = of_parse_phandle_with_args(cpu_node, "power-domains",
+					"#power-domain-cells", 0, &args);
+	if (!err) {
+		dn = args.np;
+		err = of_count_phandle_with_args(dn, "domain-idle-states",
+								NULL);
+		cpu_has_domain = (err > 0);
+	}
+
+	if (cpu_has_domain) {
+		property = "domain-idle-states";
+	} else {
+		property = "cpu-idle-states";
+		dn = cpu_node;
+	}
+
+	return of_parse_phandle(dn, property, idx);
+}
+
+/*
  * Check that the idle state is uniform across all CPUs in the CPUidle driver
  * cpumask
  */
@@ -118,8 +151,7 @@ static bool idle_state_valid(struct device_node *state_node, unsigned int idx,
 	for (cpu = cpumask_next(cpumask_first(cpumask), cpumask);
 	     cpu < nr_cpu_ids; cpu = cpumask_next(cpu, cpumask)) {
 		cpu_node = of_cpu_device_node_get(cpu);
-		curr_state_node = of_parse_phandle(cpu_node, "cpu-idle-states",
-						   idx);
+		curr_state_node = get_state_node(cpu_node, idx);
 		if (state_node != curr_state_node)
 			valid = false;
 
@@ -176,7 +208,7 @@ int dt_init_idle_driver(struct cpuidle_driver *drv,
 	cpu_node = of_cpu_device_node_get(cpumask_first(cpumask));
 
 	for (i = 0; ; i++) {
-		state_node = of_parse_phandle(cpu_node, "cpu-idle-states", i);
+		state_node = get_state_node(cpu_node, i);
 		if (!state_node)
 			break;
 
-- 
2.7.4

WARNING: multiple messages have this Message-ID (diff)
From: lina.iyer@linaro.org (Lina Iyer)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH V5 1/6] drivers: cpuidle: Read CPU's idle state from PM domain
Date: Fri,  3 Mar 2017 13:48:12 -0800	[thread overview]
Message-ID: <1488577697-127445-2-git-send-email-lina.iyer@linaro.org> (raw)
In-Reply-To: <1488577697-127445-1-git-send-email-lina.iyer@linaro.org>

Currently CPUs idle states and idle states of its parent are represented
in a flattened model by the cpu-dile-states property of the CPU node.
The CPUs idle states are followed by its cluster idle states. With the
introduction of CPU PM domains, the CPUs and domain idle states may be
represented hierarchically as part of the domain DT definition. This
would mean presenting idle state information in 2 places - CPU nodes for
the CPU and the cluster's with the PM domains.

Also, it makes sense to define domains around each individual CPU since
each of them is a power domain in its own right. The CPU idle states can
now be represented as its domain's idle state, defined by the
domain-idle-states property. This avoids presenting idle states in
multiple places in the DT.

Modify the DT-based cpuidle driver to check for the presence of a CPU's
domain and if present read the domain-idle-states of the PM domain and
if the CPU's domain is absent, revert to reading in the cpu-idle-states
property of the CPU DT node.

Suggested-by: Sudeep Holla <sudeep.holla@qrm.com>
Signed-off-by: Lina Iyer <lina.iyer@linaro.org>
---
 drivers/cpuidle/dt_idle_states.c | 38 +++++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/cpuidle/dt_idle_states.c b/drivers/cpuidle/dt_idle_states.c
index ffca4fc..4df7d45 100644
--- a/drivers/cpuidle/dt_idle_states.c
+++ b/drivers/cpuidle/dt_idle_states.c
@@ -98,6 +98,39 @@ static int init_state_node(struct cpuidle_state *idle_state,
 }
 
 /*
+ * Get the state node at @idx. State node may be defined as domain's idle state
+ * if the CPU has its own domain or defined as CPU's idle state if it doesn't
+ * have a domain provider.
+ */
+static struct device_node *get_state_node(struct device_node *cpu_node,
+				unsigned int idx)
+{
+	struct device_node *dn;
+	bool cpu_has_domain = false;
+	struct of_phandle_args args;
+	const char *property;
+	int err;
+
+	err = of_parse_phandle_with_args(cpu_node, "power-domains",
+					"#power-domain-cells", 0, &args);
+	if (!err) {
+		dn = args.np;
+		err = of_count_phandle_with_args(dn, "domain-idle-states",
+								NULL);
+		cpu_has_domain = (err > 0);
+	}
+
+	if (cpu_has_domain) {
+		property = "domain-idle-states";
+	} else {
+		property = "cpu-idle-states";
+		dn = cpu_node;
+	}
+
+	return of_parse_phandle(dn, property, idx);
+}
+
+/*
  * Check that the idle state is uniform across all CPUs in the CPUidle driver
  * cpumask
  */
@@ -118,8 +151,7 @@ static bool idle_state_valid(struct device_node *state_node, unsigned int idx,
 	for (cpu = cpumask_next(cpumask_first(cpumask), cpumask);
 	     cpu < nr_cpu_ids; cpu = cpumask_next(cpu, cpumask)) {
 		cpu_node = of_cpu_device_node_get(cpu);
-		curr_state_node = of_parse_phandle(cpu_node, "cpu-idle-states",
-						   idx);
+		curr_state_node = get_state_node(cpu_node, idx);
 		if (state_node != curr_state_node)
 			valid = false;
 
@@ -176,7 +208,7 @@ int dt_init_idle_driver(struct cpuidle_driver *drv,
 	cpu_node = of_cpu_device_node_get(cpumask_first(cpumask));
 
 	for (i = 0; ; i++) {
-		state_node = of_parse_phandle(cpu_node, "cpu-idle-states", i);
+		state_node = get_state_node(cpu_node, i);
 		if (!state_node)
 			break;
 
-- 
2.7.4

  reply	other threads:[~2017-03-03 21:48 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-03 21:48 [PATCH V5 0/6] PSCI: Support hierarchical PM domains Lina Iyer
2017-03-03 21:48 ` Lina Iyer
2017-03-03 21:48 ` Lina Iyer [this message]
2017-03-03 21:48   ` [PATCH V5 1/6] drivers: cpuidle: Read CPU's idle state from PM domain Lina Iyer
2017-03-13 13:58   ` Brendan Jackman
2017-03-13 13:58     ` Brendan Jackman
2017-03-03 21:48 ` [PATCH V5 2/6] drivers: firmware: psci: Allow OS Initiated suspend mode Lina Iyer
2017-03-03 21:48   ` Lina Iyer
2017-03-03 21:48 ` [PATCH V5 3/6] drivers: firmware: psci: Support cluster idle states for OS-Initiated Lina Iyer
2017-03-03 21:48   ` Lina Iyer
2017-03-03 21:48 ` [PATCH V5 4/6] drivers: firmwware: psci: Support hierachical idle states Lina Iyer
2017-03-03 21:48   ` Lina Iyer
2017-03-03 21:48 ` [PATCH V5 5/6] dt/bindings: Update binding for hierarchical PSCI states Lina Iyer
2017-03-03 21:48   ` Lina Iyer
2017-03-03 21:48 ` [PATCH V5 6/6] ARM64: dts: Define CPU power domain for MSM8916 Lina Iyer
2017-03-03 21:48   ` Lina Iyer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1488577697-127445-2-git-send-email-lina.iyer@linaro.org \
    --to=lina.iyer@linaro.org \
    --cc=Juri.Lelli@arm.com \
    --cc=andy.gross@linaro.org \
    --cc=brendan.jackman@arm.com \
    --cc=khilman@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=rjw@rjwysocki.net \
    --cc=sboyd@codeaurora.org \
    --cc=sudeep.holla@arm.com \
    --cc=ulf.hansson@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.