Skip to contentSkip to conversation

OMAP OPP layer for 2.6.37

PATCH New Kevin Hilman 7 participants 30 messages Sep 15, 2010 – Sep 17, 2010

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

cover OMAP OPP layer for 2.6.37

This series introduces a layer to manage Operating Performance Points (OPPs) for OMAP SoCs.

Special thanks to Nishanth for doing the bulk of the work on this series and to Thara for doing significant review, testing and updates.

Kevin Hilman (1):
OMAP3: remove OPP interfaces from OMAP PM layer

Nishanth Menon (2):
OMAP: introduce OPP layer for device-specific OPPs
OMAP3: OPP: add OPP table data and initialization

Paul Walmsley (1):
OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

 Documentation/arm/OMAP/omap_pm                |   83 +++++
 arch/arm/mach-omap2/Makefile                  |    2 +
 arch/arm/mach-omap2/io.c                      |    3 +-
 arch/arm/mach-omap2/opp3xxx_data.c            |  126 +++++++
 arch/arm/mach-omap2/pm.c                      |    6 +
 arch/arm/mach-omap2/pm.h                      |    1 +
 arch/arm/plat-omap/Makefile                   |    6 +
 arch/arm/plat-omap/include/plat/omap-pm.h     |   31 +--
 arch/arm/plat-omap/include/plat/opp.h         |  145 ++++++++
 arch/arm/plat-omap/include/plat/opp_twl_tps.h |   21 ++
 arch/arm/plat-omap/omap-pm-noop.c             |   11 +-
 arch/arm/plat-omap/opp.c                      |  461 +++++++++++++++++++++++++
 arch/arm/plat-omap/opp_twl_tps.c              |   41 +++
 13 files changed, 904 insertions(+), 33 deletions(-)
 create mode 100644 arch/arm/mach-omap2/opp3xxx_data.c
 create mode 100644 arch/arm/plat-omap/include/plat/opp.h
 create mode 100644 arch/arm/plat-omap/include/plat/opp_twl_tps.h
 create mode 100644 arch/arm/plat-omap/opp.c
 create mode 100644 arch/arm/plat-omap/opp_twl_tps.c
Signature · 1 line
1.7.2.1
#1 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

From: Nishanth Menon <[email protected]>

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. These are called Operating Performance Points or OPPs. The actual definitions of OMAP Operating Points varies over silicon within the same family of devices. For a specific domain, you can have a set of {frequency, voltage} pairs. As the kernel boots and more information is available, a set of these are activated based on the precise nature of device the kernel boots up on. It is interesting to remember that each IP which belongs to a voltage domain may define their own set of OPPs on top of this.

This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

Note: OPP is a concept that can be used in all OMAPs, it is hence introduced under plat-omap

Contributions include:
Sanjeev Premi for the initial concept:

	http://patchwork.kernel.org/patch/50998/

Kevin Hilman for converting original design to device-based Kevin Hilman and Paul Walmsey for cleaning up many of the function abstractions, improvements and data structure handling Romit Dasgupta for using enums instead of opp pointers Thara Gopinath, Eduardo Valentin and Vishwanath BS for fixes and cleanups.

Discussions and comments from:
http://marc.info/?l=linux-omap&m=126033945313269&w=2
http://marc.info/?l=linux-omap&m=125482970102327&w=2
http://marc.info/?t=125809247500002&r=1&w=2
http://marc.info/?l=linux-omap&m=126025973426007&w=2
incorporated.

Cc Benoit Cousson <[email protected]>
Cc Madhusudhan Chikkature Rajashekar <[email protected]>
Cc Phil Carmody <[email protected]>
Cc Roberto Granados Dorado <[email protected]>
Cc Santosh Shilimkar <[email protected]>
Cc Sergio Alberto Aguirre Rodriguez <[email protected]>
Cc Tero Kristo <[email protected]>
Cc Eduardo Valentin <[email protected]>
Cc Paul Walmsley <[email protected]>
Cc Romit Dasgupta <[email protected]>
Cc Sanjeev Premi <[email protected]>
Cc Thara Gopinath <[email protected]>
Cc Vishwanath BS <[email protected]>
---
 Documentation/arm/OMAP/omap_pm        |   83 ++++++
 arch/arm/plat-omap/Makefile           |    5 +
 arch/arm/plat-omap/include/plat/opp.h |  145 +++++++++++
 arch/arm/plat-omap/opp.c              |  461 +++++++++++++++++++++++++++++++++
 4 files changed, 694 insertions(+), 0 deletions(-)
 create mode 100644 arch/arm/plat-omap/include/plat/opp.h
 create mode 100644 arch/arm/plat-omap/opp.c
Documentation/arm/OMAP/omap_pm +83 -0
index 5389440..6527cdf 100644
@@ -127,3 +127,86 @@ implementation needs:
 10. (*pdata->cpu_set_freq)(unsigned long f)
 
 11. (*pdata->cpu_get_freq)(void)
+
+OMAP OPP Layer
+==============
+OMAP SOCs have a standard set of tuples consisting of frequency and
+voltage pairs that the device will support per voltage domain. This
+is called Operating Performance Point or OPP. The actual definitions
+of OMAP OPP varies over silicon within the same family of devices.
+For a specific domain, you can have a set of {frequency, voltage}
+pairs. As the kernel boots and more information is available, a set
+of these are activated based on the precise nature of device the kernel
+boots up on. It is interesting to remember that each IP which belongs
+to a voltage domain may define their own set of OPPs on top of this.
+
+OPP layer of its own depends on silicon specific implementation and
+board specific data to finalize on the final set of OPPs available
+in a system
+
+Initial list initialization:
+---------------------------
+The normal initialization sequence is for boardxyz_init_irq to call
+omap2_init_common_hw (for omap2+) and which in turn does the default
+setup required.
+
+Silicon specific initialization: First the OPP layer needs to be told
+to initialize the tables for OMAP3, there are two options here:
+a) If the desire is to use the default tables defined for that silicon,
+the board file does not need to call any initialization function, the
+defaults are setup as part of initialization flow when
+omap2_init_common_hw is called.
+
+b) board files would like to customize the default definition. In this
+case, board file needs to call explicitly prior to table operations.
+the sequence is:
+boardxyz_init_irq()
+{
+	... do things ..
+	omap3_pm_init_opp_table()
+	.. customizations and other things ..
+	omap2_init_common_hw()
+}
+1. omap3_pm_init_opp_table - this in turn calls opp_init_list for all
+OPP types. This is the generic silicon operating points, however, the
+system may have additional features or customizations required. This
+flexibility is provided by the following apis:
+
+Query functions:
+----------------
+Search for OPPs for various cases:
+2. opp_find_freq_exact - exact search function
+3. opp_find_freq_floor - round_up search function
+4. opp_find_freq_ceil - round_down search function
+
+OPP modifier functions:
+----------------------
+This allows opp layer users to add customized OPPs or change the table
+for any need they may have
+5. opp_add - add a new OPP - NOTE: use struct omap_opp_def and define
+the custom OPP with OMAP_OPP_DEF for usage.
+6. opp_enable - enable a disabled OPP
+7. opp_disable - disable an enabled OPP
+
+OPP Data retrieval functions:
+----------------------------
+The following sets of functions are useful for drivers to retrieve
+data stored in opp layer for various functions.
+8. opp_get_voltage - retrieve voltage for an opp
+9. opp_get_freq - get the frequency for an opp
+10. opp_get_opp_count - get number of opps enabled for a domain
+
+Cpufreq table generation:
+------------------------
+11. opp_init_cpufreq_table - this translates the OPP layer's internal
+OPP arrangement into a table understood and operated upon by the
+cpufreq layer.
+
+Data Structures:
+---------------
+struct omap_opp * is a handle structure whose internals are known only
+to the OPP layer and is meant to hide the complexity away from users of
+opp layer.
+
+struct omap_opp_def * is the definitions that users can interface with
+opp layer and is meant to define one OPP.
arch/arm/plat-omap/Makefile +5 -0
index 9405831..c718a0a 100644
@@ -12,6 +12,11 @@ obj-  :=
 # OCPI interconnect support for 1710, 1610 and 5912
 obj-$(CONFIG_ARCH_OMAP16XX) += ocpi.o
 
+# OPP support in (OMAP3+ only at the moment)
+ifdef CONFIG_PM
+obj-$(CONFIG_ARCH_OMAP3) += opp.o
+endif
+
 # omap_device support (OMAP2+ only at the moment)
 obj-$(CONFIG_ARCH_OMAP2) += omap_device.o
 obj-$(CONFIG_ARCH_OMAP3) += omap_device.o
arch/arm/plat-omap/include/plat/opp.h new file +145 -0

145 lines not on this page · raw message

arch/arm/plat-omap/opp.c new file +461 -0

461 lines not on this page · raw message

Signature · 1 line
1.7.2.1
#2 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 262 lines
Quoted 262 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of Kevin Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 1/4] OMAP: introduce OPP layer for device-specific OPPs

From: Nishanth Menon <[email protected]>

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. These are called Operating Performance Points or OPPs. The actual definitions of OMAP Operating Points varies over silicon within the same family of devices. For a specific domain, you can have a set of {frequency, voltage} pairs. As the kernel boots and more information is available, a set of these are activated based on the precise nature of device the kernel boots up on. It is interesting to remember that each IP which belongs to a voltage domain may define their own set of OPPs on top of this.

This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

Note: OPP is a concept that can be used in all OMAPs, it is hence introduced under plat-omap

Contributions include:
Sanjeev Premi for the initial concept:

	http://patchwork.kernel.org/patch/50998/

Kevin Hilman for converting original design to device-based Kevin Hilman and Paul Walmsey for cleaning up many of the function abstractions, improvements and data structure handling Romit Dasgupta for using enums instead of opp pointers Thara Gopinath, Eduardo Valentin and Vishwanath BS for fixes and cleanups.

Discussions and comments from:
http://marc.info/?l=linux-omap&m=126033945313269&w=2
http://marc.info/?l=linux-omap&m=125482970102327&w=2
http://marc.info/?t=125809247500002&r=1&w=2
http://marc.info/?l=linux-omap&m=126025973426007&w=2
incorporated.

Cc Benoit Cousson <[email protected]>
Cc Madhusudhan Chikkature Rajashekar <[email protected]>
Cc Phil Carmody <[email protected]>
Cc Roberto Granados Dorado <[email protected]>
Cc Santosh Shilimkar <[email protected]>
Cc Sergio Alberto Aguirre Rodriguez <[email protected]>
Cc Tero Kristo <[email protected]>
Cc Eduardo Valentin <[email protected]>
Cc Paul Walmsley <[email protected]>
Cc Romit Dasgupta <[email protected]>
Cc Sanjeev Premi <[email protected]>
Cc Thara Gopinath <[email protected]>
Cc Vishwanath BS <[email protected]>

---

Documentation/arm/OMAP/omap_pm        |   83 ++++++
arch/arm/plat-omap/Makefile           |    5 +
arch/arm/plat-omap/include/plat/opp.h |  145 +++++++++++
arch/arm/plat-omap/opp.c              |  461 +++++++++++++++++++++++++++++++++
4 files changed, 694 insertions(+), 0 deletions(-)

create mode 100644 arch/arm/plat-omap/include/plat/opp.h create mode 100644 arch/arm/plat-omap/opp.c

Documentation/arm/OMAP/omap_pm +83 -0
index 5389440..6527cdf 100644
@@ -127,3 +127,86 @@ implementation needs:
 10. (*pdata->cpu_set_freq)(unsigned long f)
 
 11. (*pdata->cpu_get_freq)(void)
+
+OMAP OPP Layer
+==============
+OMAP SOCs have a standard set of tuples consisting of frequency and
+voltage pairs that the device will support per voltage domain. This
+is called Operating Performance Point or OPP. The actual definitions
+of OMAP OPP varies over silicon within the same family of devices.
+For a specific domain, you can have a set of {frequency, voltage}
+pairs. As the kernel boots and more information is available, a set
+of these are activated based on the precise nature of device the kernel
+boots up on. It is interesting to remember that each IP which belongs
+to a voltage domain may define their own set of OPPs on top of this.
+
+OPP layer of its own depends on silicon specific implementation and
+board specific data to finalize on the final set of OPPs available
+in a system
+
+Initial list initialization:
+---------------------------
+The normal initialization sequence is for boardxyz_init_irq to call
+omap2_init_common_hw (for omap2+) and which in turn does the default
+setup required.
+
+Silicon specific initialization: First the OPP layer needs to be told
+to initialize the tables for OMAP3, there are two options here:
+a) If the desire is to use the default tables defined for that silicon,
+the board file does not need to call any initialization function, the
+defaults are setup as part of initialization flow when
+omap2_init_common_hw is called.
+
+b) board files would like to customize the default definition. In this
+case, board file needs to call explicitly prior to table operations.
+the sequence is:
+boardxyz_init_irq()
+{
+	... do things ..
+	omap3_pm_init_opp_table()
+	.. customizations and other things ..
+	omap2_init_common_hw()
+}
+1. omap3_pm_init_opp_table - this in turn calls opp_init_list for all
+OPP types. This is the generic silicon operating points, however, the
+system may have additional features or customizations required. This
+flexibility is provided by the following apis:
+
+Query functions:
+----------------
+Search for OPPs for various cases:
+2. opp_find_freq_exact - exact search function
+3. opp_find_freq_floor - round_up search function
+4. opp_find_freq_ceil - round_down search function
+
+OPP modifier functions:
+----------------------
+This allows opp layer users to add customized OPPs or change the table
+for any need they may have
+5. opp_add - add a new OPP - NOTE: use struct omap_opp_def and define
+the custom OPP with OMAP_OPP_DEF for usage.
+6. opp_enable - enable a disabled OPP
+7. opp_disable - disable an enabled OPP
+
+OPP Data retrieval functions:
+----------------------------
+The following sets of functions are useful for drivers to retrieve
+data stored in opp layer for various functions.
+8. opp_get_voltage - retrieve voltage for an opp
+9. opp_get_freq - get the frequency for an opp
+10. opp_get_opp_count - get number of opps enabled for a domain
+
+Cpufreq table generation:
+------------------------
+11. opp_init_cpufreq_table - this translates the OPP layer's internal
+OPP arrangement into a table understood and operated upon by the
+cpufreq layer.
+
+Data Structures:
+---------------
+struct omap_opp * is a handle structure whose internals are known only
+to the OPP layer and is meant to hide the complexity away from users of
+opp layer.
+
+struct omap_opp_def * is the definitions that users can interface with
+opp layer and is meant to define one OPP.
arch/arm/plat-omap/Makefile +0 -0
index 9405831..c718a0a 100644
@@ -12,6 +12,11 @@ obj-  :=
 # OCPI interconnect support for 1710, 1610 and 5912

obj-$(CONFIG_ARCH_OMAP16XX) += ocpi.o

arch/arm/plat-omap/Makefile +5 -0
+# OPP support in (OMAP3+ only at the moment)
+ifdef CONFIG_PM
+obj-$(CONFIG_ARCH_OMAP3) += opp.o
+endif
+
# omap_device support (OMAP2+ only at the moment)
obj-$(CONFIG_ARCH_OMAP2) += omap_device.o
obj-$(CONFIG_ARCH_OMAP3) += omap_device.o
arch/arm/plat-omap/include/plat/opp.h new file +79 -0
new file mode 100644
index 0000000..997b56e
@@ -0,0 +1,145 @@
+/*
+ * OMAP OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *	Nishanth Menon
+ *	Romit Dasgupta <[email protected]>
+ * Copyright (C) 2009 Deep Root Systems, LLC.
+ *	Kevin Hilman
+ *
+ * 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.
+ */
+#ifndef __ASM_ARM_OMAP_OPP_H
+#define __ASM_ARM_OMAP_OPP_H
+
+#include <linux/err.h>
+#include <linux/cpufreq.h>
+
+#include <plat/common.h>
+
+/**
+ * struct omap_opp_def - OMAP OPP Definition
+ * @hwmod_name:	Name of the hwmod for this domain
+ * @freq:	Frequency in hertz corresponding to this OPP
+ * @u_volt:	Nominal voltage in microvolts corresponding to this OPP
+ * @enabled:	True/false - is this OPP enabled/disabled by default
+ *
+ * OMAP SOCs have a standard set of tuples consisting of frequency and voltage
+ * pairs that the device will support per voltage domain. This is called
+ * Operating Points or OPP. The actual definitions of OMAP Operating Points
+ * varies over silicon within the same family of devices. For a specific
+ * domain, you can have a set of {frequency, voltage} pairs and this is denoted
+ * by an array of omap_opp_def. As the kernel boots and more information is
+ * available, a set of these are activated based on the precise nature of
+ * device the kernel boots up on. It is interesting to remember that each IP
+ * which belongs to a voltage domain may define their own set of OPPs on top
+ * of this - but this is handled by the appropriate driver.
+ */
+struct omap_opp_def {
+	char *hwmod_name;
+
+	unsigned long freq;
+	unsigned long u_volt;
+
+	bool enabled;
+};
+
+/*
+ * Initialization wrapper used to define an OPP.
+ * To point at the end of a terminator of a list of OPPs,
+ * use OMAP_OPP_DEF(NULL, 0, 0, 0)
+ */
+#define OMAP_OPP_DEF(_hwmod_name, _enabled, _freq, _uv)	\
+{						\
+	.hwmod_name	= _hwmod_name,		\
+	.enabled	= _enabled,		\
+	.freq		= _freq,		\
+	.u_volt		= _uv,			\
+}
+
+struct omap_opp;
+
+#ifdef CONFIG_PM
+
+unsigned long opp_get_voltage(const struct omap_opp *opp);
+
+unsigned long opp_get_freq(const struct omap_opp *opp);
+
+int opp_get_opp_count(struct device *dev);
+
+struct omap_opp *opp_find_freq_exact(struct device *dev,
+				     unsigned long freq, bool enabled);
+
+struct omap_opp *opp_find_freq_floor(struct device *dev, unsigned long *freq);
+
+struct omap_opp *opp_find_freq_ceil(struct device *dev, unsigned long *freq);
+
+int opp_add(const struct omap_opp_def *opp_def);
Quoted 8 lines

arch/arm/plat-omap/include/plat/opp.h line 87

Quoted 8 lines
arch/arm/plat-omap/include/plat/opp.h new file +8 -0
+
+int opp_enable(struct omap_opp *opp);
+
+int opp_disable(struct omap_opp *opp);
+
+void opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table);
+#else

Hello Kevin,

IN case of CONFIG_PM not being defined the else part will cause a compilation break as the signature of these APIs defined in the else part do not match with the signature in the if part.

Quoted 50 lines
Quoted 50 lines
arch/arm/plat-omap/include/plat/opp.h new file +50 -0
+static inline unsigned long opp_get_voltage(const struct omap_opp *opp)
+{
+	return 0;
+}
+
+static inline unsigned long opp_get_freq(const struct omap_opp *opp)
+{
+	return 0;
+}
+
+static inline int opp_get_opp_count(struct omap_opp *oppl)
+{
+	return 0;
+}
+
+static inline struct omap_opp *opp_find_freq_exact(struct omap_opp *oppl,
+						   unsigned long freq,
+						   bool enabled)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline struct omap_opp *opp_find_freq_floor(struct omap_opp *oppl,
+						   unsigned long *freq)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline struct omap_opp *opp_find_freq_ceil(struct omap_opp *oppl,
+						  unsigned long *freq)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline struct omap_opp *opp_add(struct omap_opp *oppl,
+				       const struct omap_opp_def *opp_def)
+{
+	return ERR_PTR(-EINVAL);
+}
+
+static inline int opp_enable(struct omap_opp *opp)
+{
+	return 0;
+}
+
+static inline int opp_disable(struct omap_opp *opp)
+{
+	return 0;
+}
+
Quoted 8 lines

arch/arm/plat-omap/include/plat/opp.h line 145

Quoted 8 lines
arch/arm/plat-omap/include/plat/opp.h new file +8 -0
+static inline
+void opp_init_cpufreq_table(struct omap_opp *opps,
+			    struct cpufreq_frequency_table **table)
+{
+}
+
+#endif		/* CONFIG_PM */
+#endif		/* __ASM_ARM_OMAP_OPP_H */

<snip>

Regards
Thara

#3 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 3 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap- owner at vger.kernel.org] On Behalf Of Gopinath, Thara

[...]

Quoted 6 lines
Quoted 1 line

arch/arm/plat-

Quoted 1 line
arch/arm/plat- renamed from arch/arm/plat-omap/include/plat/opp.h +0 -0

omap/include/plat/opp.h

Quoted 4 lines
Quoted 4 lines

new file mode 100644
index 0000000..997b56e

arch/arm/plat-omap/include/plat/opp.h new file +0 -0

[..]

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
arch/arm/plat-omap/include/plat/opp.h new file +3 -0
+
+#ifdef CONFIG_PM
+

[..]

Quoted 15 lines
Quoted 1 line
Quoted 1 line

+struct omap_opp *opp_find_freq_ceil(struct device *dev, unsigned long

*freq);

Quoted 10 lines

arch/arm/plat-omap/include/plat/opp.h line 87

Quoted 10 lines
arch/arm/plat-omap/include/plat/opp.h new file +10 -0
+
+int opp_add(const struct omap_opp_def *opp_def);
+
+int opp_enable(struct omap_opp *opp);
+
+int opp_disable(struct omap_opp *opp);
+
+void opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table);
+#else

Hello Kevin,

Quoted 6 lines

IN case of CONFIG_PM not being defined the else part will cause a compilation break as
the signature of these APIs defined in the else part do not match with the signature in
the if part.

Thanks for the catch. Will send a patch for this.
Regards,
Nishanth Menon

#4 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 40 lines
Quoted 40 lines

-----Original Message-----
From: Menon, Nishanth
Sent: Thursday, September 16, 2010 4:02 PM
To: Gopinath, Thara; Kevin Hilman; linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: RE: [PATCH 1/4] OMAP: introduce OPP layer for device-specific OPPs

Quoted 3 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap- owner at vger.kernel.org] On Behalf Of Gopinath, Thara

[...]

Quoted 6 lines
Quoted 1 line

arch/arm/plat-

Quoted 1 line
arch/arm/plat- renamed from arch/arm/plat-omap/include/plat/opp.h +0 -0

omap/include/plat/opp.h

Quoted 4 lines
Quoted 4 lines

new file mode 100644
index 0000000..997b56e

arch/arm/plat-omap/include/plat/opp.h new file +0 -0

[..]

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
arch/arm/plat-omap/include/plat/opp.h new file +3 -0
+
+#ifdef CONFIG_PM
+

[..]

Quoted 15 lines
Quoted 1 line
Quoted 1 line

+struct omap_opp *opp_find_freq_ceil(struct device *dev, unsigned long

*freq);

Quoted 10 lines

arch/arm/plat-omap/include/plat/opp.h line 87

Quoted 10 lines
arch/arm/plat-omap/include/plat/opp.h new file +10 -0
+
+int opp_add(const struct omap_opp_def *opp_def);
+
+int opp_enable(struct omap_opp *opp);
+
+int opp_disable(struct omap_opp *opp);
+
+void opp_init_cpufreq_table(struct device *dev,
+			    struct cpufreq_frequency_table **table);
+#else

Hello Kevin,

Quoted 7 lines
Quoted 7 lines
Quoted 6 lines

IN case of CONFIG_PM not being defined the else part will cause a compilation break as
the signature of these APIs defined in the else part do not match with the signature in
the if part.

Thanks for the catch. Will send a patch for this.

I learnt this the hard way by actually hitting the issue :-)!!

Regards
Thara

#5 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

In reply to Kevin Hilman

2010/9/15 Kevin Hilman <[email protected]>:

Quoted 6 lines

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. ?These are called Operating Performance Points or OPPs.
(...)
This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

OPPs are a generic concept, it's in silicon construction textbooks and all. Should this code not be made generic instead? You wouldn't make regulators or even DMA platform-specific these days, so why should OPPs be?

What in this code is actually OMAP-specific, more than that you name some functions omap_*, and how hard would it be to put it under arch/arm/common/*.c
arch/arm/include/asm/*.h

Possible even higher up in the directory hiearchy in include/linux/opp.h for the header and drivers/opp/*.c, because I think SuperH and power are not that different in this respect.

Yours,
Linus Walleij

#6 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Linus Walleij had written, on 09/16/2010 07:19 AM, the following:

Quoted 9 lines

2010/9/15 Kevin Hilman <[email protected]>:

Quoted 6 lines

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. These are called Operating Performance Points or OPPs.
(...)
This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

Quoted 4 lines

OPPs are a generic concept, it's in silicon construction textbooks and all. Should this code not be made generic instead? You wouldn't make regulators or even DMA platform-specific these days, so why should OPPs be?

As far as I see this patch :
hwmod[1] which is omap specific which inturn depends on omap_device. - this impacts opp_add function in the opp layer.

[1] http://marc.info/?l=linux-omap&m=128226580816341&w=2

Signature · 2 lines
Regards,
Nishanth Menon
#7 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 9 lines · Nishanth Menon

Linus Walleij had written, on 09/16/2010 07:19 AM, the following:

Quoted 8 lines

2010/9/15 Kevin Hilman <[email protected]>:

Quoted 6 lines

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. These are called Operating Performance Points or OPPs.
(...)
This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

Quoted 7 lines
Quoted 4 lines

OPPs are a generic concept, it's in silicon construction textbooks and all. Should this code not be made generic instead? You wouldn't make regulators or even DMA platform-specific these days, so why should OPPs be?

As far as I see this patch :
hwmod[1] which is omap specific which inturn depends on omap_device. - this impacts opp_add function in the opp layer.

Then wrap your local OPP in some clever way:

struct omap_opp {
    struct hwmod foo;
    struct opp opp;
};

container_of() is your friend.

Alternatively and not as elegant is to provide an
void *private_data; in the generic opp struct.

And similar design patterns for code and other platform-specific hooks. Overridable struct opp_ops for example, the same way we just abstracted the l2x0 operations for example.

It's not like there are no precedents in the linux kernel on how to handle a superclass of some struct, so how hard can it be?

The fact that a single struct member is OMAP-specific doesn't nix the fact that the major part of this OPP framework is generic code.

Yours,
Linus Walleij

#8 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

In reply to Linus Walleij

Hi Linus,

Quoted 9 lines · Linus Walleij

2010/9/15 Kevin Hilman <[email protected]>:

Quoted 6 lines

OMAP SOCs have a standard set of tuples consisting of frequency and voltage pairs that the device will support per voltage domain. ?These are called Operating Performance Points or OPPs.
(...)
This introduces a common handling OPP mechanism accross all OMAPs. As a start this is used for OMAP3.

Quoted 4 lines

OPPs are a generic concept, it's in silicon construction textbooks and all. Should this code not be made generic instead? You wouldn't make regulators or even DMA platform-specific these days, so why should OPPs be?

You're right.

Quoted 1 line

What in this code is actually OMAP-specific

Only the users. ;)

We currently register OPPs using an OMAP hwmod name, but we could easily change that to use a struct device instead which would make this much more generic (note we manage OPPs per-device, not just for the CPU)

The patch below[1] demonstrates quickly how easily we could remove all OMAP specific stuff from opp.[ch], and move it to the OMAP-specific code that does the opp_add()

Quoted 8 lines

more than that you name
some functions omap_*, and how hard would it be to put it under arch/arm/common/*.c
arch/arm/include/asm/*.h

Possible even higher up in the directory hiearchy in include/linux/opp.h for the header and drivers/opp/*.c, because I think SuperH and power are not that different in this respect.

Yeah, I guess this isn't ARM specific either, so should be at a higher level.

Nishanth, can take my hack below and continue this evolution? As I demonstrate with this hack, this won't really change anything for us.

Kevin

#9 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Kevin Hilman had written, on 09/16/2010 10:08 AM, the following:
[..]

Quoted 9 lines
Quoted 8 lines

more than that you name
some functions omap_*, and how hard would it be to put it under arch/arm/common/*.c
arch/arm/include/asm/*.h

Possible even higher up in the directory hiearchy in include/linux/opp.h for the header and drivers/opp/*.c, because I think SuperH and power are not that different in this respect.

Quoted 5 lines

Yeah, I guess this isn't ARM specific either, so should be at a higher level.

Nishanth, can take my hack below and continue this evolution? As I demonstrate with this hack, this won't really change anything for us.

thanks.. The only contention ahead is: where do we want this? Is drivers/opp/opp_core.c the right place? Given that this is just a support library and not really a driver? for some reason lib/opp.c does'nt sound just right either :(

[..]

Signature · 2 lines
Regards,
Nishanth Menon
#10 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 14 lines · Nishanth Menon

Kevin Hilman had written, on 09/16/2010 10:08 AM, the following:
[..]

Quoted 12 lines
Quoted 8 lines

more than that you name
some functions omap_*, and how hard would it be to put it under arch/arm/common/*.c
arch/arm/include/asm/*.h

Possible even higher up in the directory hiearchy in include/linux/opp.h for the header and drivers/opp/*.c, because I think SuperH and power are not that different in this respect.

Yeah, I guess this isn't ARM specific either, so should be at a higher level.

Quoted 7 lines
Quoted 2 lines

Nishanth, can take my hack below and continue this evolution? As I demonstrate with this hack, this won't really change anything for us.

thanks.. The only contention ahead is: where do we want this? Is drivers/opp/opp_core.c the right place? Given that this is just a support library and not really a driver? for some reason lib/opp.c does'nt sound just right either :(

Well, since it's not really a driver, I don't think drivers/* is appropriate.

lib/opp/* seems more appropriate to me with the header at include/linux/opp.h as Linus suggested.

Kevin

#11 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

2010/9/16 Kevin Hilman <[email protected]>:

Quoted 2 lines

lib/opp/* seems more appropriate to me with the header at include/linux/opp.h as Linus suggested.

I second this. I would love to see the generic OPP stuff in lib/opp/* so we put it in the right place from the beginning and don't have to painfully refactor everything later (clk.h especially comes to mind.)

Yours,
Linus Walleij

#12 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Linus Walleij had written, on 09/16/2010 12:07 PM, the following:

Quoted 9 lines

2010/9/16 Kevin Hilman <[email protected]>:

Quoted 2 lines

lib/opp/* seems more appropriate to me with the header at include/linux/opp.h as Linus suggested.

I second this. I would love to see the generic OPP stuff in lib/opp/* so we put it in the right place from the beginning and don't have to painfully refactor everything later (clk.h especially comes to mind.)

cool.. digging a bit deeper into lib directory, I propose the following:
lib/opp.c
include/linux/opp.h

any soc specific data (for registration etc) -> goes to arch/<arch>/mach-<soc>/oppdata_xyz.c or what ever they choose

the intent being lib/ is no place to put mach or arch specific data definitions.. it is just noise..

if folks are ok with this, will post a new rev soonish..

Signature · 2 lines
Regards,
Nishanth Menon
#13 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 24 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap- owner at vger.kernel.org] On Behalf Of Menon, Nishanth
Sent: Thursday, September 16, 2010 10:40 PM
To: Linus Walleij
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm- kernel at lists.infradead.org
Subject: Re: [PATCH 1/4] OMAP: introduce OPP layer for device-specific OPPs

Linus Walleij had written, on 09/16/2010 12:07 PM, the following:

Quoted 9 lines

2010/9/16 Kevin Hilman <[email protected]>:

Quoted 2 lines

lib/opp/* seems more appropriate to me with the header at include/linux/opp.h as Linus suggested.

I second this. I would love to see the generic OPP stuff in lib/opp/* so we put it in the right place from the beginning and don't have to painfully refactor everything later (clk.h especially comes to mind.)

cool.. digging a bit deeper into lib directory, I propose the following:
lib/opp.c
include/linux/opp.h

Quoted 8 lines

any soc specific data (for registration etc) -> goes to arch/<arch>/mach-<soc>/oppdata_xyz.c or what ever they choose

the intent being lib/ is no place to put mach or arch specific data definitions.. it is just noise..

if folks are ok with this, will post a new rev soonish..

If you like may be you can take this thread over lkml just to get more wider perspective

#14 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 27 lines · Santosh
Quoted 27 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap- owner at vger.kernel.org] On Behalf Of Menon, Nishanth
Sent: Thursday, September 16, 2010 10:40 PM
To: Linus Walleij
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm- kernel at lists.infradead.org
Subject: Re: [PATCH 1/4] OMAP: introduce OPP layer for device-specific OPPs

Linus Walleij had written, on 09/16/2010 12:07 PM, the following:

Quoted 9 lines

2010/9/16 Kevin Hilman <[email protected]>:

Quoted 2 lines

lib/opp/* seems more appropriate to me with the header at include/linux/opp.h as Linus suggested.

I second this. I would love to see the generic OPP stuff in lib/opp/* so we put it in the right place from the beginning and don't have to painfully refactor everything later (clk.h especially comes to mind.)

cool.. digging a bit deeper into lib directory, I propose the following:
lib/opp.c
include/linux/opp.h

any soc specific data (for registration etc) -> goes to arch/<arch>/mach-<soc>/oppdata_xyz.c or what ever they choose

Quoted 7 lines
Quoted 5 lines

the intent being lib/ is no place to put mach or arch specific data definitions.. it is just noise..

if folks are ok with this, will post a new rev soonish..

If you like may be you can take this thread over lkml just to get more wider perspective

I propose Nishanth does the slight rework to move this to lib/ and then post to l-o, l-a-k and lkml for RFC. Note that only the generic parts should be posted first, the OMAP specific usage of this can be fixed after the generic parts are accepted.

Kevin

#15 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

In reply to Kevin Hilman

Nishant,

Since you are anyways re-sending this you might as well fix these nits.

Quoted 22 lines · ext Kevin Hilman

From: Nishanth Menon<[email protected]>

arch/arm/plat-omap/opp.c new file +12 -0
new file mode 100644
index 0000000..17f93b2
@@ -0,0 +1,461 @@
+/*
+ * OMAP OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *     Nishanth Menon
+ *     Romit Dasgupta<[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.
+ */
+
 

arch/arm/plat-omap/opp.c line 107

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+/**
+ * opp_get_voltage() - Gets the voltage corresponding to an opp
+ * @opp:       opp for which voltage has to be returned for
+ *
+ * Return voltage in micro volt corresponding to the opp, else
+ * return 0
+ */
+unsigned long opp_get_voltage(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

If !opp->enabled, then is it really an invalid parameter being passed?

I'm not user if people need to know the voltage of a disabled OPP, but do we need to limit it?

Quoted 7 lines
arch/arm/plat-omap/opp.c new file +7 -0
+               pr_err("%s: Invalid parameters being passed\n", __func__);
+               return 0;
+       }
+
+       return opp->u_volt;
+}
+

arch/arm/plat-omap/opp.c line 124

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+/**
+ * opp_get_freq() - Gets the frequency corresponding to an opp
+ * @opp:       opp for which frequency has to be returned for
+ *
+ * Return frequency in hertz corresponding to the opp, else
+ * return 0
+ */
+unsigned long opp_get_freq(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

ditto.

Quoted 26 lines
arch/arm/plat-omap/opp.c new file +26 -0
+               pr_err("%s: Invalid parameters being passed\n", __func__);
+               return 0;
+       }
+
+       return opp->rate;
+}
+
+/**
+ * opp_get_opp_count() - Get number of opps enabled in the opp list
+ * @opp_type:  OPP type we want to count
+ *
+ * This functions returns the number of opps if there are any OPPs enabled,
+ * else returns corresponding error value.
+ */
+int opp_get_opp_count(struct device *dev)
+{
+       struct device_opp *dev_opp;
+
+       dev_opp = find_device_opp(dev);
+       if (IS_ERR(dev_opp))
+               return -ENODEV;
+
+       return dev_opp->enabled_opp_count;
+}
+
+/**

arch/arm/plat-omap/opp.c line 160

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+ * opp_find_freq_exact() - search for an exact frequency
+ * @opp_type:  OPP type we want to search in.
+ * @freq:      frequency to search for
+ * @enabled:   enabled/disabled OPP to search for
+ *
+ * Searches for exact match in the opp list and returns handle to the matching
+ * opp if found, else returns ERR_PTR in case of error and should be handled
+ * using IS_ERR.
+ *
+ * Note enabled is a modifier for the search. if enabled=true, then the match is

Good to add some punctuation after 'Note'.

Quoted 3 lines

+ * for exact matching frequency and is enabled. if false, the match is for exact
+ * frequency which is disabled.
+ */

cheers,
-roger

#16 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Roger Quadros had written, on 09/16/2010 08:54 AM, the following:

Quoted 1 line

Since you are anyways re-sending this you might as well fix these nits.

thanks for reviewing..

Quoted 1 line

[..]

Quoted 46 lines
Quoted 19 lines
arch/arm/plat-omap/opp.c new file +12 -0
new file mode 100644
index 0000000..17f93b2
@@ -0,0 +1,461 @@
+/*
+ * OMAP OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *     Nishanth Menon
+ *     Romit Dasgupta<[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.
+ */
+
 

arch/arm/plat-omap/opp.c line 107

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+/**
+ * opp_get_voltage() - Gets the voltage corresponding to an opp
+ * @opp:       opp for which voltage has to be returned for
+ *
+ * Return voltage in micro volt corresponding to the opp, else
+ * return 0
+ */
+unsigned long opp_get_voltage(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

If !opp->enabled, then is it really an invalid parameter being passed?

I'm not user if people need to know the voltage of a disabled OPP, but do we need to limit it?

Quoted 11 lines
arch/arm/plat-omap/opp.c new file +11 -0
+               pr_err("%s: Invalid parameters being passed\n", __func__);
+               return 0;
+       }
+
+       return opp->u_volt;
+}
+
+/**
+ * opp_get_freq() - Gets the frequency corresponding to an opp
+ * @opp:       opp for which frequency has to be returned for
+ *
Quoted 8 lines

arch/arm/plat-omap/opp.c line 124

Quoted 6 lines

+ * Return frequency in hertz corresponding to the opp, else
+ * return 0
+ */

arch/arm/plat-omap/opp.c new file +3 -0
+unsigned long opp_get_freq(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

ditto.

Yes, the intent here was for opp operational apis to function ONLY on the enabled opps - this helps to pull out any bugs in the users who might be unintentionally using bad params.

do you have any usecase you can think of where we might want to use these on a disabled opp?

Quoted 31 lines
Quoted 31 lines
arch/arm/plat-omap/opp.c new file +30 -0
+               pr_err("%s: Invalid parameters being passed\n", __func__);
+               return 0;
+       }
+
+       return opp->rate;
+}
+
+/**
+ * opp_get_opp_count() - Get number of opps enabled in the opp list
+ * @opp_type:  OPP type we want to count
+ *
+ * This functions returns the number of opps if there are any OPPs enabled,
+ * else returns corresponding error value.
+ */
+int opp_get_opp_count(struct device *dev)
+{
+       struct device_opp *dev_opp;
+
+       dev_opp = find_device_opp(dev);
+       if (IS_ERR(dev_opp))
+               return -ENODEV;
+
+       return dev_opp->enabled_opp_count;
+}
+
+/**
+ * opp_find_freq_exact() - search for an exact frequency
+ * @opp_type:  OPP type we want to search in.
+ * @freq:      frequency to search for
+ * @enabled:   enabled/disabled OPP to search for
Quoted 8 lines
Quoted 6 lines

+ *
+ * Searches for exact match in the opp list and returns handle to the matching
+ * opp if found, else returns ERR_PTR in case of error and should be handled
+ * using IS_ERR.
+ *
+ * Note enabled is a modifier for the search. if enabled=true, then the match is

Good to add some punctuation after 'Note'.

Thanks will do

Quoted 7 lines
Quoted 4 lines

+ * for exact matching frequency and is enabled. if false, the match is for exact
+ * frequency which is disabled.
+ */

cheers,
-roger

Signature · 2 lines
Regards,
Nishanth Menon
#17 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Quoted 62 lines · ext Nishanth Menon

Roger Quadros had written, on 09/16/2010 08:54 AM, the following:

Quoted 1 line

Since you are anyways re-sending this you might as well fix these nits.

thanks for reviewing..

Quoted 1 line

[..]

Quoted 46 lines
Quoted 19 lines
arch/arm/plat-omap/opp.c new file +12 -0
new file mode 100644
index 0000000..17f93b2
@@ -0,0 +1,461 @@
+/*
+ * OMAP OPP Interface
+ *
+ * Copyright (C) 2009-2010 Texas Instruments Incorporated.
+ *     Nishanth Menon
+ *     Romit Dasgupta<[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.
+ */
+
 

arch/arm/plat-omap/opp.c line 107

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+/**
+ * opp_get_voltage() - Gets the voltage corresponding to an opp
+ * @opp:       opp for which voltage has to be returned for
+ *
+ * Return voltage in micro volt corresponding to the opp, else
+ * return 0
+ */
+unsigned long opp_get_voltage(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

If !opp->enabled, then is it really an invalid parameter being passed?

I'm not user if people need to know the voltage of a disabled OPP, but do we need to limit it?

Quoted 11 lines
arch/arm/plat-omap/opp.c new file +11 -0
+               pr_err("%s: Invalid parameters being passed\n", __func__);
+               return 0;
+       }
+
+       return opp->u_volt;
+}
+
+/**
+ * opp_get_freq() - Gets the frequency corresponding to an opp
+ * @opp:       opp for which frequency has to be returned for
+ *
Quoted 8 lines

arch/arm/plat-omap/opp.c line 124

Quoted 6 lines

+ * Return frequency in hertz corresponding to the opp, else
+ * return 0
+ */

arch/arm/plat-omap/opp.c new file +3 -0
+unsigned long opp_get_freq(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

ditto.

Quoted 4 lines

Yes, the intent here was for opp operational apis to function ONLY on the enabled opps - this helps to pull out any bugs in the users who might be unintentionally using bad params.

OK.

Quoted 2 lines

do you have any usecase you can think of where we might want to use these on a disabled opp?

Not really. Based on what is an OPP enabled/disabled? Is it possible for an initially enabled OPP to be disabled at some point in time? What triggers this disable?
OR does an OPP enabled at boot time remain enabled throughout the power session?

regards,
-roger

#18 lore raw

PATCH 1/4 OMAP: introduce OPP layer for device-specific OPPs

Roger Quadros had written, on 09/16/2010 09:20 AM, the following:
[...]

Quoted 17 lines
Quoted 11 lines
Quoted 11 lines

arch/arm/plat-omap/opp.c line 124

Quoted 10 lines
arch/arm/plat-omap/opp.c new file +10 -0
+/**
+ * opp_get_freq() - Gets the frequency corresponding to an opp
+ * @opp:       opp for which frequency has to be returned for
+ *
+ * Return frequency in hertz corresponding to the opp, else
+ * return 0
+ */
+unsigned long opp_get_freq(const struct omap_opp *opp)
+{
+       if (unlikely(!opp || IS_ERR(opp)) || !opp->enabled) {

ditto.

Quoted 4 lines

Yes, the intent here was for opp operational apis to function ONLY on the enabled opps - this helps to pull out any bugs in the users who might be unintentionally using bad params.

OK.

Quoted 8 lines
Quoted 2 lines

do you have any usecase you can think of where we might want to use these on a disabled opp?

Not really. Based on what is an OPP enabled/disabled? Is it possible for an initially enabled OPP to be disabled at some point in time? What triggers this disable?
OR does an OPP enabled at boot time remain enabled throughout the power session?

At this point - enable/disable is done at init time - there is flexibility for board files to define their own custom OPPs (as some custom boards need to). they dont change once this initial definition is done. there are plans to do dynamic enable disable based on previous discussion in l-o - the framework to use opp layer in that manner is yet to go up yet.

Signature · 2 lines
Regards,
Nishanth Menon
#19 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

In reply to Kevin Hilman

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

---
 arch/arm/plat-omap/Makefile                   |    1 +
 arch/arm/plat-omap/include/plat/opp_twl_tps.h |   21 +++++++++++++
 arch/arm/plat-omap/opp_twl_tps.c              |   41 +++++++++++++++++++++++++
 3 files changed, 63 insertions(+), 0 deletions(-)
 create mode 100644 arch/arm/plat-omap/include/plat/opp_twl_tps.h
 create mode 100644 arch/arm/plat-omap/opp_twl_tps.c
arch/arm/plat-omap/Makefile +1 -0
index c718a0a..a88879c 100644
@@ -15,6 +15,7 @@ obj-$(CONFIG_ARCH_OMAP16XX) += ocpi.o
 # OPP support in (OMAP3+ only at the moment)
 ifdef CONFIG_PM
 obj-$(CONFIG_ARCH_OMAP3) += opp.o
+obj-$(CONFIG_TWL4030_CORE) += opp_twl_tps.o
 endif
 
 # omap_device support (OMAP2+ only at the moment)
arch/arm/plat-omap/include/plat/opp_twl_tps.h new file +21 -0
new file mode 100644
index 0000000..8784e5f
@@ -0,0 +1,21 @@
+/*
+ * opp_twl_tps.h - TWL/TPS-specific headers for the OPP code
+ *
+ * Copyright (C) 2009 Texas Instruments Incorporated.
+ *	Nishanth Menon
+ *
+ * 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.
+ *
+ * XXX This code belongs as part of some other TWL/TPS code.
+ */
+#ifndef _ARCH_ARM_PLAT_OMAP_OPP_TWL_TPS_H
+#define _ARCH_ARM_PLAT_OMAP_OPP_TWL_TPS_H
+
+#include <linux/kernel.h>
+
+unsigned long omap_twl_vsel_to_uv(const u8 vsel);
+u8 omap_twl_uv_to_vsel(unsigned long uV);
+
+#endif
arch/arm/plat-omap/opp_twl_tps.c new file +41 -0

41 lines not on this page · raw message

Signature · 1 line
1.7.2.1
#20 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 10 lines
Quoted 10 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of Kevin Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 4 lines
Quoted 4 lines

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

I have been using this code for a while now. I really do not think wee need a separate
file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage
layer is interested in these conversions. Voltage layer has a structure that can be populated with
the information required from the PMIC. We only need to add two more function pointers to this structure. This info can then be passed from the actual PMIC driver file. This will make it much more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also
in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and omap_twl_uv_to_vsel. So please consider dropping this patch from this series.

Regards
Thara

Quoted 108 lines
Quoted 108 lines

---

arch/arm/plat-omap/Makefile                   |    1 +
arch/arm/plat-omap/include/plat/opp_twl_tps.h |   21 +++++++++++++
arch/arm/plat-omap/opp_twl_tps.c              |   41 +++++++++++++++++++++++++
3 files changed, 63 insertions(+), 0 deletions(-)

create mode 100644 arch/arm/plat-omap/include/plat/opp_twl_tps.h create mode 100644 arch/arm/plat-omap/opp_twl_tps.c

arch/arm/plat-omap/Makefile +0 -0
index c718a0a..a88879c 100644
@@ -15,6 +15,7 @@ obj-$(CONFIG_ARCH_OMAP16XX) += ocpi.o
 # OPP support in (OMAP3+ only at the moment)

ifdef CONFIG_PM
obj-$(CONFIG_ARCH_OMAP3) += opp.o
+obj-$(CONFIG_TWL4030_CORE) += opp_twl_tps.o
endif

# omap_device support (OMAP2+ only at the moment)
arch/arm/plat- renamed from arch/arm/plat-omap/include/plat/opp_twl_tps.h +0 -0

omap/include/plat/opp_twl_tps.h
new file mode 100644
index 0000000..8784e5f

arch/arm/plat-omap/include/plat/opp_twl_tps.h new file +21 -0
@@ -0,0 +1,21 @@
+/*
+ * opp_twl_tps.h - TWL/TPS-specific headers for the OPP code
+ *
+ * Copyright (C) 2009 Texas Instruments Incorporated.
+ *	Nishanth Menon
+ *
+ * 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.
+ *
+ * XXX This code belongs as part of some other TWL/TPS code.
+ */
+#ifndef _ARCH_ARM_PLAT_OMAP_OPP_TWL_TPS_H
+#define _ARCH_ARM_PLAT_OMAP_OPP_TWL_TPS_H
+
+#include <linux/kernel.h>
+
+unsigned long omap_twl_vsel_to_uv(const u8 vsel);
+u8 omap_twl_uv_to_vsel(unsigned long uV);
+
+#endif
arch/arm/plat-omap/opp_twl_tps.c new file +41 -1
new file mode 100644
index 0000000..112f106
@@ -0,0 +1,41 @@
+/*
+ * opp_twl_tps.c - TWL/TPS-specific functions for the OPP code
+ *
+ * Copyright (C) 2009 Texas Instruments Incorporated.
+ * Nishanth Menon
+ * Copyright (C) 2009 Nokia Corporation
+ * Paul Walmsley
+ *
+ * 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.
+ *
+ * XXX This code should be part of some other TWL/TPS code.
+ */
+
+#include <plat/opp_twl_tps.h>
+
+/**
+ * omap_twl_vsel_to_vdc - convert TWL/TPS VSEL value to microvolts DC
+ * @vsel: TWL/TPS VSEL value to convert
+ *
+ * Returns the microvolts DC that the TWL/TPS family of PMICs should
+ * generate when programmed with @vsel.
+ */
+unsigned long omap_twl_vsel_to_uv(const u8 vsel)
+{
+	return (((vsel * 125) + 6000)) * 100;
+}
+
+/**
+ * omap_twl_uv_to_vsel - convert microvolts DC to TWL/TPS VSEL value
+ * @uv: microvolts DC to convert
+ *
+ * Returns the VSEL value necessary for the TWL/TPS family of PMICs to
+ * generate an output voltage equal to or greater than @uv microvolts DC.
+ */
+u8 omap_twl_uv_to_vsel(unsigned long uv)
+{
+	/* Round up to higher voltage */
+	return DIV_ROUND_UP(uv - 600000, 12500);
+}
--

1.7.2.1

--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo at vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

#21 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 05:40 AM, the following:

Quoted 11 lines
Quoted 11 lines
Quoted 11 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of Kevin Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

From: Paul Walmsley <[email protected]>

Quoted 7 lines
Quoted 2 lines
Quoted 2 lines

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage

This split introduces a PMIC level abstraction already. Do you have a suggestion which file it should go to? It is definitely not part of opp.c, not part of other existing twl files as well. the job of this file was to introduce conversion routines which can be used by any layer (voltage layer if need be - it used to be srf and smartreflex before).. in fact one of your voltage layer patches introduces capability for 6030 as well
http://marc.info/?l=linux-omap&m=128213020927919&w=2

Quoted 3 lines

layer is interested in these conversions. Voltage layer has a structure that can be populated with the information required from the PMIC. We only need to add two more function pointers to this structure. This info can then be passed from the actual PMIC driver file. This

will make it much

Quoted 2 lines

more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and omap_twl_uv_to_vsel.

I think the problem is with the voltage layer (which has not been posted upstream) which is using hard coded function pointer. What the patchset should have done is to introduce function pointers registration from twl_tps.c to voltage layer and voltage layer should ideally been using function pointers by itself.

Quoted 1 line

So please consider dropping this patch from this series.

I think I disagree - rationale for having this separated as a pmic specific file is still sound, only the implementation of the future framework should have changed (it should be using function pointers instead of hardcoded function names). in fact I can add additional suggestion for the voltage layer: the pmic selection should be done from a board file - This will allow voltage layer to handle numerous PMICs and combination of PMICs controlling various domains as well.. the only neat way to handle it is ofcourse using function pointers.

PS: Suggestion
- please fix your mailer to round off for 70/80 chars..
- might be good to point folks to rfc patchset for voltage layer to give context.

Signature · 2 lines
Regards,
Nishanth Menon
#22 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 29 lines
Quoted 29 lines

-----Original Message-----
From: Menon, Nishanth
Sent: Thursday, September 16, 2010 5:45 PM
To: Gopinath, Thara
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 05:40 AM, the following:

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of

Kevin

Quoted 15 lines
Quoted 10 lines
Quoted 10 lines

Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage

Quoted 8 lines
Quoted 8 lines

This split introduces a PMIC level abstraction already. Do you have a suggestion which file it should go to? It is definitely not part of opp.c, not part of other existing twl files as well. the job of this file was to introduce conversion routines which can be used by any layer (voltage layer if need be - it used to be srf and smartreflex before).. in fact one of your voltage layer patches introduces capability for 6030 as well
http://marc.info/?l=linux-omap&m=128213020927919&w=2

Yes one of my patches has introduces this coz I had no other way to add OMAP4 support. But I still do not understand why cant these APIs be implemented in twl-core.c or twl4030-power.c?

Quoted 16 lines
Quoted 16 lines
Quoted 3 lines

layer is interested in these conversions. Voltage layer has a structure that can be populated with the information required from the PMIC. We only need to add two more function pointers to this

structure.

Quoted 1 line

This info can then be passed from the actual PMIC driver file. This

will make it much

Quoted 2 lines

more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and

omap_twl_uv_to_vsel.
I think the problem is with the voltage layer (which has not been posted upstream) which is using hard coded function pointer. What the patchset should have done is to introduce function pointers registration from twl_tps.c to voltage layer and voltage layer should ideally been using function pointers by itself.

Quoted 1 line

So please consider dropping this patch from this series.

Quoted 8 lines
Quoted 8 lines

I think I disagree - rationale for having this separated as a pmic specific file is still sound, only the implementation of the future framework should have changed (it should be using function pointers instead of hardcoded function names). in fact I can add additional suggestion for the voltage layer: the pmic selection should be done from a board file - This will allow voltage layer to handle numerous PMICs and combination of PMICs controlling various domains as well.. the only neat way to handle it is ofcourse using function pointers.

Exactly if voltage layer has to use a func ptr, why not let them be populated by the PMIC driver files directly?

Quoted 4 lines
Quoted 4 lines

PS: Suggestion
- please fix your mailer to round off for 70/80 char.

Will try.

Quoted 2 lines
Quoted 2 lines

- might be good to point folks to rfc patchset for voltage layer to give context.

https://patchwork.kernel.org/patch/119429/

Quoted 4 lines
Quoted 4 lines

--
Regards,
Nishanth Menon

#23 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 08:51 AM, the following:

Quoted 35 lines
Quoted 26 lines
Quoted 26 lines

-----Original Message-----
From: Menon, Nishanth
Sent: Thursday, September 16, 2010 5:45 PM
To: Gopinath, Thara
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 05:40 AM, the following:

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of

Kevin

Quoted 14 lines
Quoted 10 lines
Quoted 10 lines

Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage

Quoted 8 lines
Quoted 8 lines

This split introduces a PMIC level abstraction already. Do you have a suggestion which file it should go to? It is definitely not part of opp.c, not part of other existing twl files as well. the job of this file was to introduce conversion routines which can be used by any layer (voltage layer if need be - it used to be srf and smartreflex before).. in fact one of your voltage layer patches introduces capability for 6030 as well
http://marc.info/?l=linux-omap&m=128213020927919&w=2

Quoted 3 lines

Yes one of my patches has introduces this coz I had no other way to add OMAP4 support. But I still do not understand why cant these APIs be implemented in twl-core.c or twl4030-power.c?

Why there? Twl power does regulator operations not conversion operations. core is not the place either as it is function independent.

Quoted 25 lines
Quoted 16 lines
Quoted 16 lines
Quoted 3 lines

layer is interested in these conversions. Voltage layer has a structure that can be populated with the information required from the PMIC. We only need to add two more function pointers to this

structure.

Quoted 1 line

This info can then be passed from the actual PMIC driver file. This

will make it much

Quoted 2 lines

more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and

omap_twl_uv_to_vsel.
I think the problem is with the voltage layer (which has not been posted upstream) which is using hard coded function pointer. What the patchset should have done is to introduce function pointers registration from twl_tps.c to voltage layer and voltage layer should ideally been using function pointers by itself.

Quoted 1 line

So please consider dropping this patch from this series.

Quoted 8 lines
Quoted 8 lines

I think I disagree - rationale for having this separated as a pmic specific file is still sound, only the implementation of the future framework should have changed (it should be using function pointers instead of hardcoded function names). in fact I can add additional suggestion for the voltage layer: the pmic selection should be done from a board file - This will allow voltage layer to handle numerous PMICs and combination of PMICs controlling various domains as well.. the only neat way to handle it is ofcourse using function pointers.

Quoted 2 lines

Exactly if voltage layer has to use a func ptr, why not let them be populated by the PMIC driver files directly?

Again, we are going into the details of how voltage layer will be implemented.. What suggestion do you do when on board x vdd1 is driven by PMIC1, vdd2 by PMIC2? it will have to be somesort of different registration mechanism. but nothing prevents the framework from registering the pointers defined in twl_tps.c

Quoted 11 lines
Quoted 4 lines
Quoted 4 lines

PS: Suggestion
- please fix your mailer to round off for 70/80 char.

Will try.

Quoted 2 lines
Quoted 2 lines

- might be good to point folks to rfc patchset for voltage layer to give context.

https://patchwork.kernel.org/patch/119429/

thanks..

Signature · 2 lines
Regards,
Nishanth Menon
#24 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 45 lines
Quoted 45 lines

-----Original Message-----
From: Menon, Nishanth
Sent: Thursday, September 16, 2010 7:37 PM
To: Gopinath, Thara
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 08:51 AM, the following:

Quoted 35 lines
Quoted 34 lines
Quoted 34 lines

-----Original Message-----
From: Menon, Nishanth
Sent: Thursday, September 16, 2010 5:45 PM
To: Gopinath, Thara
Cc: Kevin Hilman; linux-omap at vger.kernel.org; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/16/2010 05:40 AM, the following:

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

-----Original Message-----
From: linux-omap-owner at vger.kernel.org [mailto:linux-omap-owner at vger.kernel.org] On Behalf Of

Kevin

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

Hilman
Sent: Thursday, September 16, 2010 3:27 AM
To: linux-omap at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Subject: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 8 lines
Quoted 4 lines
Quoted 4 lines

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage

This split introduces a PMIC level abstraction already. Do you have a suggestion which file it should go to? It is definitely not part of opp.c, not part of other existing twl files as well. the job of this file was to introduce conversion routines which can be used by any layer (voltage layer if need be - it used to be srf and smartreflex before).. in fact one of your voltage layer patches introduces capability for 6030 as well
http://marc.info/?l=linux-omap&m=128213020927919&w=2

Quoted 5 lines
Quoted 5 lines
Quoted 3 lines

Yes one of my patches has introduces this coz I had no other way to add OMAP4 support. But I still do not understand why cant these APIs be implemented in twl-core.c or twl4030-power.c?

Why there? Twl power does regulator operations not conversion operations. core is not the place either as it is function independent.

Hello Nishant,

Why do you say core is not the place. For me core is exactly the place. It is the PMIC driver file.

Regards
Thara

#25 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Gopinath, Thara had written, on 09/17/2010 09:57 AM, the following:
[..]

Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 15 lines
Quoted 7 lines
Quoted 4 lines
Quoted 4 lines

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage

This split introduces a PMIC level abstraction already. Do you have a suggestion which file it should go to? It is definitely not part of opp.c, not part of other existing twl files as well. the job of this file was to introduce conversion routines which can be used by any layer (voltage layer if need be - it used to be srf and smartreflex before).. in fact one of your voltage layer patches introduces capability for 6030 as well
http://marc.info/?l=linux-omap&m=128213020927919&w=2

Quoted 8 lines
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines

Yes one of my patches has introduces this coz I had no other way to add OMAP4 support. But I still do not understand why cant these APIs be implemented in twl-core.c or twl4030-power.c?

Why there? Twl power does regulator operations not conversion operations. core is not the place either as it is function independent.

Why do you say core is not the place. For me core is exactly the place. It is the PMIC driver file.

See [1] - they are all generic without a specific domain implementation. if we are so touchy about the location we want these api implementation, I guess core might be a compromise enough.. personally even though I believe core should remain specific functionality independent, I dont think I have very strong opinions about adding helpers there.. i would'nt put registration there though.. That should be in board files..

[1]
http://git.kernel.org/?p=linux/kernel/git/khilman/linux-omap-pm.git;a=blob;f=drivers/mfd/twl4030-core.c;h=769b34bd48e445880ac0920423d9b73eabaf4cb7;hb=HEAD

Signature · 2 lines
Regards,
Nishanth Menon
#26 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

In reply to Gopinath, Thara

"Gopinath, Thara" <[email protected]> writes:

[...]

Quoted 7 lines
Quoted 4 lines
Quoted 4 lines

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

Quoted 10 lines

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage layer is interested in these conversions. Voltage layer has a structure that can be populated with the information required from the PMIC. We only need to add two more function pointers to this structure. This info can then be passed from the actual PMIC driver file. This will make it much more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and omap_twl_uv_to_vsel.

Feel free to post an RFC series for what you're proposing.

IMHO, even the voltage layer should not care about these conversions. The voltage layer should care about voltages, and the PMIC layer should handle these conversions. IOW, this stuff is best abstracted in the PMIC code, or board- or SoC-specific hooks that can be registered with the PMIC.

Quoted 1 line

So please consider dropping this patch from this series.

If you pick this up as part of your voltage series (or more likely as a pre-requisite of your voltage series) we can surely drop it from this series, at it doesn't really belong with the OPP series.

There are no upstream users of this yet, so dropping it is easy. :)

Kevin

#27 lore raw

PATCH 2/4 OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

Quoted 39 lines
Quoted 39 lines

-----Original Message-----
From: Kevin Hilman [mailto:khilman at deeprootsystems.com]
Sent: Thursday, September 16, 2010 10:46 PM
To: Gopinath, Thara
Cc: linux-omap at vger.kernel.org; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH 2/4] OMAP: OPP: twl/tps: Introduce TWL/TPS-specific code

"Gopinath, Thara" <[email protected]> writes:

[...]

Quoted 7 lines
Quoted 4 lines
Quoted 4 lines

From: Paul Walmsley <[email protected]>

The OPP layer code should be independent of the PMIC, introduce the TWL/TPS-specific code out to its own file.

Hello Kevin,

Quoted 10 lines

I have been using this code for a while now. I really do not think wee need a separate file for implementing the vsel to voltage in (uV) and vice versa formulas. Today only voltage layer is interested in these conversions. Voltage layer has a structure that can be populated with the information required from the PMIC. We only need to add two more function pointers to this structure. This info can then be passed from the actual PMIC driver file. This will make it much more simpler for OMAP4 where we have different formulas between different revisions of PMIC. Also in the omap voltage code we will no longer have to hard code omap_twl_vsel_to_uv and omap_twl_uv_to_vsel.

Feel free to post an RFC series for what you're proposing.

IMHO, even the voltage layer should not care about these conversions. The voltage layer should care about voltages, and the PMIC layer should handle these conversions. IOW, this stuff is best abstracted in the PMIC code, or board- or SoC-specific hooks that can be registered with the PMIC.

Quoted 5 lines
Quoted 5 lines
Quoted 1 line

So please consider dropping this patch from this series.

If you pick this up as part of your voltage series (or more likely as a pre-requisite of your voltage series) we can surely drop it from this series, at it doesn't really belong with the OPP series.

Hello Kevin/Nishant

My suggestion is introduce two new APIs in drivers/twl-core.c. Then once voltage layer support is added call into an API provided by voltage layer to register two APIs. If both Nishant and you are ok with this I will post this change as par of my voltage and smartreflex series.

Regards
Thara

#28 lore raw

PATCH 3/4 OMAP3: remove OPP interfaces from OMAP PM layer

In reply to Kevin Hilman

With new OPP layer, OPP users will access OPP API directly instead of using OMAP PM layer, so remove all notions of OPPs from the OMAP PM layer.

---
 arch/arm/mach-omap2/io.c                  |    3 +-
 arch/arm/plat-omap/include/plat/omap-pm.h |   31 +++++++++-------------------
 arch/arm/plat-omap/omap-pm-noop.c         |   11 +---------
 3 files changed, 12 insertions(+), 33 deletions(-)
arch/arm/mach-omap2/io.c +1 -2
index b9ea70b..c09bf10 100644
@@ -323,8 +323,7 @@ void __init omap2_init_common_hw(struct omap_sdrc_params *sdrc_cs0,
 		omap2430_hwmod_init();
 	else if (cpu_is_omap34xx())
 		omap3xxx_hwmod_init();
-	/* The OPP tables have to be registered before a clk init */
-	omap_pm_if_early_init(mpu_opps, dsp_opps, l3_opps);
+	omap_pm_if_early_init();
 
 	if (cpu_is_omap2420())
 		omap2420_clk_init();
arch/arm/plat-omap/include/plat/omap-pm.h +10 -21
index 728fbb9..c5b533d 100644
@@ -19,24 +19,7 @@
 #include <linux/clk.h>
 
 #include "powerdomain.h"
-
-/**
- * struct omap_opp - clock frequency-to-OPP ID table for DSP, MPU
- * @rate: target clock rate
- * @opp_id: OPP ID
- * @min_vdd: minimum VDD1 voltage (in millivolts) for this OPP
- *
- * Operating performance point data.  Can vary by OMAP chip and board.
- */
-struct omap_opp {
-	unsigned long rate;
-	u8 opp_id;
-	u16 min_vdd;
-};
-
-extern struct omap_opp *mpu_opps;
-extern struct omap_opp *dsp_opps;
-extern struct omap_opp *l3_opps;
+#include <plat/opp.h>
 
 /*
  * agent_id values for use with omap_pm_set_min_bus_tput():
@@ -59,9 +42,11 @@ extern struct omap_opp *l3_opps;
  * framework starts.  The "_if_" is to avoid name collisions with the
  * PM idle-loop code.
  */
-int __init omap_pm_if_early_init(struct omap_opp *mpu_opp_table,
-				 struct omap_opp *dsp_opp_table,
-				 struct omap_opp *l3_opp_table);
+#ifdef CONFIG_OMAP_PM_NONE
+#define omap_pm_if_early_init() 0
+#else
+int __init omap_pm_if_early_init(void);
+#endif
 
 /**
  * omap_pm_if_init - OMAP PM init code called after clock fw init
@@ -69,7 +54,11 @@ int __init omap_pm_if_early_init(struct omap_opp *mpu_opp_table,
  * The main initialization code.  OPP tables are passed in here.  The
  * "_if_" is to avoid name collisions with the PM idle-loop code.
  */
+#ifdef CONFIG_OMAP_PM_NONE
+#define omap_pm_if_init() 0
+#else
 int __init omap_pm_if_init(void);
+#endif
 
 /**
  * omap_pm_if_exit - OMAP PM exit code
arch/arm/plat-omap/omap-pm-noop.c +1 -10
index e129ce8..ca75abb 100644
@@ -26,10 +26,6 @@
 
 #include <plat/powerdomain.h>
 
-struct omap_opp *dsp_opps;
-struct omap_opp *mpu_opps;
-struct omap_opp *l3_opps;
-
 /*
  * Device-driver-originated constraints (via board-*.c files)
  */
@@ -308,13 +304,8 @@ int omap_pm_get_dev_context_loss_count(struct device *dev)
 
 
 /* Should be called before clk framework init */
-int __init omap_pm_if_early_init(struct omap_opp *mpu_opp_table,
-				 struct omap_opp *dsp_opp_table,
-				 struct omap_opp *l3_opp_table)
+int __init omap_pm_if_early_init(void)
 {
-	mpu_opps = mpu_opp_table;
-	dsp_opps = dsp_opp_table;
-	l3_opps = l3_opp_table;
 	return 0;
 }
 
Signature · 1 line
1.7.2.1
#29 lore raw

PATCH 4/4 OMAP3: OPP: add OPP table data and initialization

In reply to Kevin Hilman

From: Nishanth Menon <[email protected]>

Add OPP data for OMAP34xx and OMAP36xx and initialization functions to populate OPP tables based on current SoC.

---
 arch/arm/mach-omap2/Makefile       |    2 +
 arch/arm/mach-omap2/opp3xxx_data.c |  126 ++++++++++++++++++++++++++++++++++++
 arch/arm/mach-omap2/pm.c           |    6 ++
 arch/arm/mach-omap2/pm.h           |    1 +
 4 files changed, 135 insertions(+), 0 deletions(-)
 create mode 100644 arch/arm/mach-omap2/opp3xxx_data.c
arch/arm/mach-omap2/Makefile +2 -0
index 88d3a1e..33acca9 100644
@@ -57,6 +57,8 @@ obj-$(CONFIG_PM_DEBUG)			+= pm-debug.o
 AFLAGS_sleep24xx.o			:=-Wa,-march=armv6
 AFLAGS_sleep34xx.o			:=-Wa,-march=armv7-a
 
+obj-$(CONFIG_ARCH_OMAP3)		+= opp3xxx_data.o
+
 endif
 
 # PRCM
arch/arm/mach-omap2/opp3xxx_data.c new file +126 -0

126 lines not on this page · raw message

arch/arm/mach-omap2/pm.c +6 -0
index 68f9f2e..4ea59df 100644
@@ -18,6 +18,8 @@
 #include <plat/omap_device.h>
 #include <plat/common.h>
 
+#include "pm.h"
+
 static struct omap_device_pm_latency *pm_lats;
 
 static struct device *mpu_dev;
@@ -76,6 +78,10 @@ static void omap2_init_processor_devices(void)
 static int __init omap2_common_pm_init(void)
 {
 	omap2_init_processor_devices();
+
+	if (cpu_is_omap34xx())
+		omap3_pm_init_opp_table();
+
 	omap_pm_if_init();
 
 	return 0;
arch/arm/mach-omap2/pm.h +1 -0
index 3de6ece..9c9f943 100644
@@ -22,6 +22,7 @@ extern void omap_sram_idle(void);
 extern int omap3_can_sleep(void);
 extern int set_pwrdm_state(struct powerdomain *pwrdm, u32 state);
 extern int omap3_idle_init(void);
+extern int omap3_pm_init_opp_table(void);
 
 struct cpuidle_params {
 	u8  valid;
Signature · 1 line
1.7.2.1