Skip to contentSkip to conversation

OMAP3 SRF: omap3 srf driver

PATCH In review Rajendra Nayak 2 participants 3 messages Oct 16, 2008 – Oct 17, 2008

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 3/4 OMAP3 SRF: omap3 srf driver

Adds init/change_level/validate_level calls for latency resources on OMAP3

---
 arch/arm/mach-omap2/resource34xx.c |  126 +++++++++++++++++++++++++++++++++++++
 1 files changed, 126 insertions(+)
arch/arm/mach-omap2/resource34xx.c new file +126 -0
Index: linux-omap-2.6/arch/arm/mach-omap2/resource34xx.c
===================================================================
@@ -0,0 +1,126 @@
+/*
+ * linux/arch/arm/mach-omap2/resource34xx.c
+ * OMAP3 resource init/change_level/validate_level functions
+ *
+ * Copyright (C) 2007-2008 Texas Instruments, Inc.
+ * Rajendra Nayak <[email protected]>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * THIS PACKAGE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
+ * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
+ * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
+ * History:
+ *
+ */
+
+#include <linux/pm_qos_params.h>
+#include <mach/powerdomain.h>
+#include <mach/clockdomain.h>
+#include "resource34xx.h"
+
+int set_pwrdm_state(struct powerdomain *pwrdm, u32 state);
+
+/**
+ * init_latency - Initializes the mpu/core latency resource.
+ * @resp: Latency resource to be initalized
+ *
+ * No return value.
+ */
+void init_latency(struct shared_resource *resp)
+{
+	resp->no_of_users = 0;
+	resp->curr_level = RES_DEFAULTLEVEL;
+	*((u8 *)resp->resource_data) = 0;
+	return;
+}
+
+/**
+ * set_latency - Adds/Updates and removes the CPU_DMA_LATENCY in *pm_qos_params.
+ * @resp: resource pointer
+ * @latency: target latency to be set
+ *
+ * Returns 0 on success, or error values as returned by
+ * pm_qos_update_requirement/pm_qos_add_requirement.
+ */
+int set_latency(struct shared_resource *resp, u32 latency)
+{
+	u8 *pm_qos_req_added;
+
+	if (resp->curr_level == latency)
+		return 0;
+	else
+		/* Update the resources current level */
+		resp->curr_level = latency;
+
+	pm_qos_req_added = resp->resource_data;
+	if (latency == RES_DEFAULTLEVEL)
+		/* No more users left, remove the pm_qos_req if present */
+		if (*pm_qos_req_added) {
+			pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY,
+							resp->name);
+			*pm_qos_req_added = 0;
+			return 0;
+		}
+
+	if (*pm_qos_req_added) {
+		return pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY,
+						resp->name, latency);
+	} else {
+		*pm_qos_req_added = 1;
+		return pm_qos_add_requirement(PM_QOS_CPU_DMA_LATENCY,
+						resp->name, latency);
+	}
+}
+
+/**
+ * init_pd_latency - Initializes the power domain latency resource.
+ * @resp: Power Domain Latency resource to be initialized.
+ *
+ * No return value.
+ */
+void init_pd_latency(struct shared_resource *resp)
+{
+	struct pd_latency_db *pd_lat_db;
+
+	resp->no_of_users = 0;
+	resp->curr_level = PD_LATENCY_OFF;
+	pd_lat_db = resp->resource_data;
+	/* Populate the power domain associated with the latency resource */
+	pd_lat_db->pd = pwrdm_lookup(pd_lat_db->pwrdm_name);
+	return;
+}
+
+/**
+ * set_pd_latency - Updates the curr_level of the power domain resource.
+ * @resp: Power domain latency resource.
+ * @latency: New latency value acceptable.
+ *
+ * This function maps the latency in microsecs to the acceptable
+ * Power domain state using the latency DB.
+ * It then programs the power domain to enter the target state.
+ * Always returns 0.
+ */
+int set_pd_latency(struct shared_resource *resp, u32 latency)
+{
+	u32 pd_lat_level, ind;
+	struct pd_latency_db *pd_lat_db;
+	struct powerdomain *pwrdm;
+
+	pd_lat_db = resp->resource_data;
+	pwrdm = pd_lat_db->pd;
+	pd_lat_level = PD_LATENCY_OFF;
+	/* using the latency db map to the appropriate PD state */
+	for (ind = 0; ind < PD_LATENCY_MAXLEVEL; ind++) {
+		if (pd_lat_db->latency[ind] < latency) {
+			pd_lat_level = ind;
+			break;
+		}
+	}
+
+	resp->curr_level = pd_lat_level;
+	set_pwrdm_state(pwrdm, pd_lat_level);
+	return 0;
+}
#1 lore raw

arch/arm/mach-omap2/resource34xx.c line 57

Quoted 11 lines

+ */

arch/arm/mach-omap2/resource34xx.c new file +10 -0
+int set_latency(struct shared_resource *resp, u32 latency)
+{
+	u8 *pm_qos_req_added;
+
+	if (resp->curr_level == latency)
+		return 0;
+	else
+		/* Update the resources current level */
+		resp->curr_level = latency;
+

[Romit] I think this should be done outside this function, in update_resource_level function.

Quoted 14 lines
arch/arm/mach-omap2/resource34xx.c new file +5 -0
+	pm_qos_req_added = resp->resource_data;
+	if (latency == RES_DEFAULTLEVEL)
+		/* No more users left, remove the pm_qos_req if present */
+		if (*pm_qos_req_added) {
+

pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY,

arch/arm/mach-omap2/resource34xx.c new file +7 -0
+							resp->name);
+			*pm_qos_req_added = 0;
+			return 0;
+		}
+
+	if (*pm_qos_req_added) {
+		return

pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY,

arch/arm/mach-omap2/resource34xx.c line 71

Quoted 2 lines
arch/arm/mach-omap2/resource34xx.c new file +2 -0
+						resp->name, latency);
+	} else {

[Romit] Shouldn't the following line execute only if pm_qos_add_requirement returns successfully?

Quoted 13 lines
arch/arm/mach-omap2/resource34xx.c new file +13 -0
+		*pm_qos_req_added = 1;
+		return pm_qos_add_requirement(PM_QOS_CPU_DMA_LATENCY,
+						resp->name, latency);
+	}
+}
+
+int set_pd_latency(struct shared_resource *resp, u32 latency)
+{
+	u32 pd_lat_level, ind;
+	struct pd_latency_db *pd_lat_db;
+	struct powerdomain *pwrdm;
+
+	pd_lat_db = resp->resource_data;

arch/arm/mach-omap2/resource34xx.c line 122

Quoted 10 lines
arch/arm/mach-omap2/resource34xx.c new file +10 -0
+	pwrdm = pd_lat_db->pd;
+	pd_lat_level = PD_LATENCY_OFF;
+	/* using the latency db map to the appropriate PD state */
+	for (ind = 0; ind < PD_LATENCY_MAXLEVEL; ind++) {
+		if (pd_lat_db->latency[ind] < latency) {
+			pd_lat_level = ind;
+			break;
+		}
+	}
+

[Romit] Again is this the right place for this? It should be done only if set_pwrdm_state returns successfully. So I think that should be checked as well.

Quoted 4 lines
arch/arm/mach-omap2/resource34xx.c new file +4 -0
+	resp->curr_level = pd_lat_level;
+	set_pwrdm_state(pwrdm, pd_lat_level);
+	return 0;
+}
#2 lore raw

One final point

Quoted 19 lines
Quoted 7 lines
arch/arm/mach-omap2/resource34xx.c new file +7 -0
+int set_pd_latency(struct shared_resource *resp, u32 latency)
+{
+	u32 pd_lat_level, ind;
+	struct pd_latency_db *pd_lat_db;
+	struct powerdomain *pwrdm;
+
+	pd_lat_db = resp->resource_data;

arch/arm/mach-omap2/resource34xx.c line 122

Quoted 10 lines
arch/arm/mach-omap2/resource34xx.c new file +10 -0
+	pwrdm = pd_lat_db->pd;
+	pd_lat_level = PD_LATENCY_OFF;
+	/* using the latency db map to the appropriate PD state */
+	for (ind = 0; ind < PD_LATENCY_MAXLEVEL; ind++) {
+		if (pd_lat_db->latency[ind] < latency) {
+			pd_lat_level = ind;
+			break;
+		}
+	}
+

[Romit] Again is this the right place for this? It should be done only if set_pwrdm_state returns successfully. So I think that should be checked as well.

Quoted 4 lines

arch/arm/mach-omap2/resource34xx.c line 126

Quoted 4 lines
arch/arm/mach-omap2/resource34xx.c new file +4 -0
+	resp->curr_level = pd_lat_level;
+	set_pwrdm_state(pwrdm, pd_lat_level);
+	return 0;
+}

[Romit] In retrospect, I think the code is not right. Someone sets a latency for say 30ms and what gets updated is the index in the curr_level. Next [s]he sets a latency value and we will be comparing index with time. Cant we have a function that would convert the latency (in time) to the right index?