diff mbox

[v8,2/2] regulator: add QCOM RPMh regulator driver

Message ID 35c4ea70cdf5caba560fb6f40e866ee8bc456d93.1529712888.git.collinsd@codeaurora.org (mailing list archive)
State Not Applicable, archived
Delegated to: Andy Gross
Headers show

Commit Message

David Collins June 23, 2018, 12:46 a.m. UTC
Add the QCOM RPMh regulator driver to manage PMIC regulators
which are controlled via RPMh on some Qualcomm Technologies, Inc.
SoCs.  RPMh is a hardware block which contains several
accelerators which are used to manage various hardware resources
that are shared between the processors of the SoC.  The final
hardware state of a regulator is determined within RPMh by
performing max aggregation of the requests made by all of the
processors.

Add support for PMIC regulator control via the voltage regulator
manager (VRM) and oscillator buffer (XOB) RPMh accelerators.
VRM supports manipulation of enable state, voltage, and mode.
XOB supports manipulation of enable state.

Signed-off-by: David Collins <collinsd@codeaurora.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
---
 drivers/regulator/Kconfig               |   9 +
 drivers/regulator/Makefile              |   1 +
 drivers/regulator/qcom-rpmh-regulator.c | 746 ++++++++++++++++++++++++++++++++
 3 files changed, 756 insertions(+)
 create mode 100644 drivers/regulator/qcom-rpmh-regulator.c

Comments

Doug Anderson June 25, 2018, 6:50 p.m. UTC | #1
Hi,

On Fri, Jun 22, 2018 at 5:46 PM, David Collins <collinsd@codeaurora.org> wrote:
> Add the QCOM RPMh regulator driver to manage PMIC regulators
> which are controlled via RPMh on some Qualcomm Technologies, Inc.
> SoCs.  RPMh is a hardware block which contains several
> accelerators which are used to manage various hardware resources
> that are shared between the processors of the SoC.  The final
> hardware state of a regulator is determined within RPMh by
> performing max aggregation of the requests made by all of the
> processors.
>
> Add support for PMIC regulator control via the voltage regulator
> manager (VRM) and oscillator buffer (XOB) RPMh accelerators.
> VRM supports manipulation of enable state, voltage, and mode.
> XOB supports manipulation of enable state.
>
> Signed-off-by: David Collins <collinsd@codeaurora.org>
> Reviewed-by: Douglas Anderson <dianders@chromium.org>
> Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
> ---
>  drivers/regulator/Kconfig               |   9 +
>  drivers/regulator/Makefile              |   1 +
>  drivers/regulator/qcom-rpmh-regulator.c | 746 ++++++++++++++++++++++++++++++++
>  3 files changed, 756 insertions(+)

I've been working with David's v7 patch series for a bit of time now
and things have been working well.  v8 looks good.  Feel free to add:

Tested-by: Douglas Anderson <dianders@chromium.org>

...to this patch.  Speaking of which it might be getting very close to
time for this series to land along with David's other series, AKA:

* [1/2] regulator: of: add property for allowed modes specification
  https://patchwork.kernel.org/patch/10395731/

* [2/2] regulator: of: add support for allowed modes configuration
  https://patchwork.kernel.org/patch/10395725/


Specifically I think we are mostly waiting for RPMh to land.  This is
on Andy's plate right now but I think it's happy and just waiting for
the actual push from Andy.


NOTE: I don't know if there anything special we need to do for landing
this and RPMh across two git repositories.  Offline Stephen Boyd
mentioned to me that maybe it was fine to land this in Mark's tree
without the RPMh code since it depends on "QCOM_RPMH", but the "||
COMPILE_TEST" worries me and makes me think auto-builders will start
yelling.  Andy and Mark: what do you think?


If it's helpful to have a link to the latest RPMh, here are patchwork
IDs on linux-arm-msm:

10477501 [v13,01/10] drivers: qcom: rpmh-rsc: add RPMH controller for QCOM SoCs
10477559 [v13,02/10] dt-bindings: introduce RPMH RSC bindings for Qualcomm SoCs
10477545 [v13,03/10] drivers: qcom: rpmh-rsc: log RPMH requests in FTRACE
10477543 [v13,04/10] drivers: qcom: rpmh: add RPMH helper functions
10477537 [v13,05/10] drivers: qcom: rpmh-rsc: write sleep/wake requests to TCS
10477539 [v13,06/10] drivers: qcom: rpmh-rsc: allow invalidation of
sleep/wake TCS
10477533 [v13,07/10] drivers: qcom: rpmh: cache sleep/wake state requests
10477529 [v13,08/10] drivers: qcom: rpmh: allow requests to be sent
asynchronously
10477525 [v13,09/10] drivers: qcom: rpmh: add support for batch RPMH request
10477519 [v13,10/10] drivers: qcom: rpmh-rsc: allow active requests
from wake TCS



-Doug
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Mark Brown June 26, 2018, 12:07 p.m. UTC | #2
On Mon, Jun 25, 2018 at 11:50:52AM -0700, Doug Anderson wrote:

> ...to this patch.  Speaking of which it might be getting very close to
> time for this series to land along with David's other series, AKA:

> * [1/2] regulator: of: add property for allowed modes specification
>   https://patchwork.kernel.org/patch/10395731/

> * [2/2] regulator: of: add support for allowed modes configuration
>   https://patchwork.kernel.org/patch/10395725/

Please do not submit new versions of already applied patches, please
submit incremental updates to the existing code.  Modifying existing
commits creates problems for other users building on top of those
commits so it's best practice to only change pubished git commits if
absolutely essential.
Doug Anderson June 26, 2018, 3 p.m. UTC | #3
Hi,

On Tue, Jun 26, 2018 at 5:07 AM, Mark Brown <broonie@kernel.org> wrote:
> On Mon, Jun 25, 2018 at 11:50:52AM -0700, Doug Anderson wrote:
>
>> ...to this patch.  Speaking of which it might be getting very close to
>> time for this series to land along with David's other series, AKA:
>
>> * [1/2] regulator: of: add property for allowed modes specification
>>   https://patchwork.kernel.org/patch/10395731/
>
>> * [2/2] regulator: of: add support for allowed modes configuration
>>   https://patchwork.kernel.org/patch/10395725/
>
> Please do not submit new versions of already applied patches, please
> submit incremental updates to the existing code.  Modifying existing
> commits creates problems for other users building on top of those
> commits so it's best practice to only change pubished git commits if
> absolutely essential.

Sorry, wasn't suggesting making any changes to those two patches, just
was noting the dependency.  ...but, as you said, the two dependent
patches have already landed and I just didn't notice.  :(  Sorry for
the noise.

-Doug
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Mark Brown June 26, 2018, 3:02 p.m. UTC | #4
On Tue, Jun 26, 2018 at 08:00:29AM -0700, Doug Anderson wrote:
> On Tue, Jun 26, 2018 at 5:07 AM, Mark Brown <broonie@kernel.org> wrote:

> > Please do not submit new versions of already applied patches, please
> > submit incremental updates to the existing code.  Modifying existing
> > commits creates problems for other users building on top of those
> > commits so it's best practice to only change pubished git commits if
> > absolutely essential.

> Sorry, wasn't suggesting making any changes to those two patches, just
> was noting the dependency.  ...but, as you said, the two dependent
> patches have already landed and I just didn't notice.  :(  Sorry for
> the noise.

Ah, so there's no revisions that need merging?  That's great.
Doug Anderson June 26, 2018, 6:15 p.m. UTC | #5
Hi,

On Tue, Jun 26, 2018 at 8:02 AM, Mark Brown <broonie@kernel.org> wrote:
> On Tue, Jun 26, 2018 at 08:00:29AM -0700, Doug Anderson wrote:
>> On Tue, Jun 26, 2018 at 5:07 AM, Mark Brown <broonie@kernel.org> wrote:
>
>> > Please do not submit new versions of already applied patches, please
>> > submit incremental updates to the existing code.  Modifying existing
>> > commits creates problems for other users building on top of those
>> > commits so it's best practice to only change pubished git commits if
>> > absolutely essential.
>
>> Sorry, wasn't suggesting making any changes to those two patches, just
>> was noting the dependency.  ...but, as you said, the two dependent
>> patches have already landed and I just didn't notice.  :(  Sorry for
>> the noise.
>
> Ah, so there's no revisions that need merging?  That's great.

Right.  So all that's left to do is decide if ${SUBJECT} patch is
ready to land and how to land it.  Andy has landed RPMh into his
for-next tree.  You can see it at
<https://git.kernel.org/pub/scm/linux/kernel/git/agross/linux.git/log/?h=for-next>.
I think he's still planning on re-shuffling his tree a bit.  When he
does this, do you need him to put the RPMh patches somewhere you can
merge into your tree?

Thanks!

-Doug
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Mark Brown June 27, 2018, 3:01 p.m. UTC | #6
On Tue, Jun 26, 2018 at 11:15:14AM -0700, Doug Anderson wrote:

> I think he's still planning on re-shuffling his tree a bit.  When he
> does this, do you need him to put the RPMh patches somewhere you can
> merge into your tree?

Well, I *think* there's no actual dependency here since it's a new
driver with a Kconfig dependency.  It really just needs me to get round
to trawling through what's a fairly large patch with a troubled history
now you've reviewed it.
Doug Anderson June 27, 2018, 4:28 p.m. UTC | #7
Hi,

On Wed, Jun 27, 2018 at 8:01 AM, Mark Brown <broonie@kernel.org> wrote:
> On Tue, Jun 26, 2018 at 11:15:14AM -0700, Doug Anderson wrote:
>
>> I think he's still planning on re-shuffling his tree a bit.  When he
>> does this, do you need him to put the RPMh patches somewhere you can
>> merge into your tree?
>
> Well, I *think* there's no actual dependency here since it's a new
> driver with a Kconfig dependency.  It really just needs me to get round
> to trawling through what's a fairly large patch with a troubled history
> now you've reviewed it.

OK, great.  I guess I'm confused about the "|| COMPILE_TEST" causing
problems then?  I was worried that anyone trying to do "COMPILE_TEST"
on your tree (or linuxnext if RPMh isn't there) would get failures due
to the lack of header files.  I guess if it's a problem you could just
gut the "|| COMPILE_TEST" and it could be added back in later?

Hoping my reviews saved you time overall.

-Doug
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Mark Brown June 28, 2018, 10:18 a.m. UTC | #8
On Wed, Jun 27, 2018 at 09:28:03AM -0700, Doug Anderson wrote:

> OK, great.  I guess I'm confused about the "|| COMPILE_TEST" causing
> problems then?  I was worried that anyone trying to do "COMPILE_TEST"
> on your tree (or linuxnext if RPMh isn't there) would get failures due
> to the lack of header files.  I guess if it's a problem you could just
> gut the "|| COMPILE_TEST" and it could be added back in later?

Ugh, yes - that'll break things.  In that case I can't apply this
without a signed tag from Andy's tree with the dependency stuff in.
David Collins June 28, 2018, 6:04 p.m. UTC | #9
Hello Mark,

On 06/28/2018 03:18 AM, Mark Brown wrote:
> On Wed, Jun 27, 2018 at 09:28:03AM -0700, Doug Anderson wrote:
> 
>> OK, great.  I guess I'm confused about the "|| COMPILE_TEST" causing
>> problems then?  I was worried that anyone trying to do "COMPILE_TEST"
>> on your tree (or linuxnext if RPMh isn't there) would get failures due
>> to the lack of header files.  I guess if it's a problem you could just
>> gut the "|| COMPILE_TEST" and it could be added back in later?
> 
> Ugh, yes - that'll break things.  In that case I can't apply this
> without a signed tag from Andy's tree with the dependency stuff in.
> 

Do you have any remaining concerns with the qcom-rpmh-regulator binding
and driver patches that would keep you from applying them (other than the
dependency patches being applied first)?

Thanks,
David
Mark Brown June 29, 2018, 11:06 a.m. UTC | #10
On Thu, Jun 28, 2018 at 11:04:17AM -0700, David Collins wrote:

> Do you have any remaining concerns with the qcom-rpmh-regulator binding
> and driver patches that would keep you from applying them (other than the
> dependency patches being applied first)?

To repeat I need to find the time to sit down and go through them
properly, the problems this driver has had mean I feel I need to be
extra careful there.
Mark Brown July 2, 2018, 10:28 a.m. UTC | #11
On Fri, Jun 22, 2018 at 05:46:14PM -0700, David Collins wrote:

> --- /dev/null
> +++ b/drivers/regulator/qcom-rpmh-regulator.c
> @@ -0,0 +1,746 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/* Copyright (c) 2018, The Linux Foundation. All rights reserved. */
> +

Please make the entire header block C++ so it looks intentional.

> +	cmd.data = bypassed ? PMIC4_BOB_MODE_PASS : pmic_mode;

Please just write normal if statements, the ternery operator isn't
really helping legibility.

> +static const int pmic_mode_map_pmic4_ldo[REGULATOR_MODE_STANDBY + 1] = {
> +	[REGULATOR_MODE_INVALID] = -EINVAL,
> +	[REGULATOR_MODE_STANDBY] = PMIC4_LDO_MODE_RETENTION,
> +	[REGULATOR_MODE_IDLE]    = PMIC4_LDO_MODE_LPM,
> +	[REGULATOR_MODE_NORMAL]  = -EINVAL,
> +	[REGULATOR_MODE_FAST]    = PMIC4_LDO_MODE_HPM,
> +};

This mapping is really weird, I'd expect one of the modes to correspond
to the normal operating mode of the regulator.  

> +static unsigned int rpmh_regulator_pmic4_ldo_of_map_mode(unsigned int mode)
> +{
> +	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
> +		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
> +		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
> +		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_INVALID,
> +		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
> +	};

Same here, based on that it looks like auto mode is a good map for
normal.

> +	if (mode >= RPMH_REGULATOR_MODE_COUNT)
> +		return -EINVAL;

Why not use ARRAY_SIZE?
David Collins July 9, 2018, 11:44 p.m. UTC | #12
Hello Mark,

On 07/02/2018 03:28 AM, Mark Brown wrote:
> On Fri, Jun 22, 2018 at 05:46:14PM -0700, David Collins wrote:
> 
>> --- /dev/null
>> +++ b/drivers/regulator/qcom-rpmh-regulator.c
>> @@ -0,0 +1,746 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/* Copyright (c) 2018, The Linux Foundation. All rights reserved. */
>> +
> 
> Please make the entire header block C++ so it looks intentional.

Sure, I'll change this.

I was trying to follow the guideline that kernel C source files should use
C style comments while at the same time following the SPDX guideline that
C++ style comments are needed for the SPDX line in C source files [1].


>> +	cmd.data = bypassed ? PMIC4_BOB_MODE_PASS : pmic_mode;
> 
> Please just write normal if statements, the ternery operator isn't
> really helping legibility.

I'll change this.


>> +static const int pmic_mode_map_pmic4_ldo[REGULATOR_MODE_STANDBY + 1] = {
>> +	[REGULATOR_MODE_INVALID] = -EINVAL,
>> +	[REGULATOR_MODE_STANDBY] = PMIC4_LDO_MODE_RETENTION,
>> +	[REGULATOR_MODE_IDLE]    = PMIC4_LDO_MODE_LPM,
>> +	[REGULATOR_MODE_NORMAL]  = -EINVAL,
>> +	[REGULATOR_MODE_FAST]    = PMIC4_LDO_MODE_HPM,
>> +};
> 
> This mapping is really weird, I'd expect one of the modes to correspond
> to the normal operating mode of the regulator.  

My thinking here was to have a consistent mapping for consumers to use
between REGULATOR_MODE_* and physical regulator modes for both LDO and
SMPS type regulators:

REGULATOR_MODE_STANDBY --> Retention (if supported)
REGULATOR_MODE_IDLE    --> Low power mode (if supported)
                           LPM for LDO and PFM for SMPS
REGULATOR_MODE_NORMAL  --> Auto HW switching between low and high power
                           mode (if supported)
REGULATOR_MODE_FAST    --> High power mode
                           HPM for LDO and PWM for SMPS

This allows a consumer to request NORMAL in typical use cases and FAST in
use cases that require low voltage ripple.  If NORMAL is not supported,
then it automatically gets upgraded to FAST by the regulator framework.

I could change it so that REGULATOR_MODE_NORMAL maps to LDO HPM mode.
However, doing so would make it so that REGULATOR_MODE_FAST requests would
fail for LDOs.  Thus, consumers would need to know if their supply is an
LDO or an SMPS (which seems undesirable).

Would it be acceptable to have both NORMAL and FAST map to LDO HPM?


>> +static unsigned int rpmh_regulator_pmic4_ldo_of_map_mode(unsigned int mode)
>> +{
>> +	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
>> +		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
>> +		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
>> +		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_INVALID,
>> +		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
>> +	};
> 
> Same here, based on that it looks like auto mode is a good map for
> normal.

LDO type regulators physically do not support AUTO mode.  That is why I
specified REGULATOR_MODE_INVALID in the mapping.


>> +	if (mode >= RPMH_REGULATOR_MODE_COUNT)
>> +		return -EINVAL;
> 
> Why not use ARRAY_SIZE?

I'll change this.


Thanks,
David

[1]:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/license-rules.rst?h=v4.18-rc4#n74
Mark Brown July 12, 2018, 4:54 p.m. UTC | #13
On Mon, Jul 09, 2018 at 04:44:14PM -0700, David Collins wrote:
> On 07/02/2018 03:28 AM, Mark Brown wrote:
> > On Fri, Jun 22, 2018 at 05:46:14PM -0700, David Collins wrote:

> >> +static const int pmic_mode_map_pmic4_ldo[REGULATOR_MODE_STANDBY + 1] = {
> >> +	[REGULATOR_MODE_INVALID] = -EINVAL,
> >> +	[REGULATOR_MODE_STANDBY] = PMIC4_LDO_MODE_RETENTION,
> >> +	[REGULATOR_MODE_IDLE]    = PMIC4_LDO_MODE_LPM,
> >> +	[REGULATOR_MODE_NORMAL]  = -EINVAL,
> >> +	[REGULATOR_MODE_FAST]    = PMIC4_LDO_MODE_HPM,
> >> +};

> > This mapping is really weird, I'd expect one of the modes to correspond
> > to the normal operating mode of the regulator.  

> My thinking here was to have a consistent mapping for consumers to use
> between REGULATOR_MODE_* and physical regulator modes for both LDO and
> SMPS type regulators:

No, that's not useful or helpful - if there's any modes I'd *really*
expect to see one of them be _NORMAL.

> This allows a consumer to request NORMAL in typical use cases and FAST in
> use cases that require low voltage ripple.  If NORMAL is not supported,
> then it automatically gets upgraded to FAST by the regulator framework.

> I could change it so that REGULATOR_MODE_NORMAL maps to LDO HPM mode.
> However, doing so would make it so that REGULATOR_MODE_FAST requests would
> fail for LDOs.  Thus, consumers would need to know if their supply is an
> LDO or an SMPS (which seems undesirable).

You've just discovered why it's a bad idea for consumers to do anything
with modes directly!  The mappings are just never going to be consistent
given the massive variation in what regulators can support, the
retention mode of one regulator might be able to deliver more power than
the normal mode of another.

> Would it be acceptable to have both NORMAL and FAST map to LDO HPM?

Having something other than a 1:1 mapping is going to lead to pain at
some point.

> >> +static unsigned int rpmh_regulator_pmic4_ldo_of_map_mode(unsigned int mode)
> >> +{
> >> +	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
> >> +		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
> >> +		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
> >> +		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_INVALID,
> >> +		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
> >> +	};

> > Same here, based on that it looks like auto mode is a good map for
> > normal.

> LDO type regulators physically do not support AUTO mode.  That is why I
> specified REGULATOR_MODE_INVALID in the mapping.

The other question here is why this is even in the table if it's not
valid (I'm not seeing a need for the MODE_COUNT define)?
David Collins July 13, 2018, 1:34 a.m. UTC | #14
On 07/12/2018 09:54 AM, Mark Brown wrote:
> On Mon, Jul 09, 2018 at 04:44:14PM -0700, David Collins wrote:
>> On 07/02/2018 03:28 AM, Mark Brown wrote:
>>> On Fri, Jun 22, 2018 at 05:46:14PM -0700, David Collins wrote:
>>>> +static unsigned int rpmh_regulator_pmic4_ldo_of_map_mode(unsigned int mode)
>>>> +{
>>>> +	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
>>>> +		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
>>>> +		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
>>>> +		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_INVALID,
>>>> +		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
>>>> +	};
> 
>>> Same here, based on that it looks like auto mode is a good map for
>>> normal.
> 
>> LDO type regulators physically do not support AUTO mode.  That is why I
>> specified REGULATOR_MODE_INVALID in the mapping.
> 
> The other question here is why this is even in the table if it's not
> valid (I'm not seeing a need for the MODE_COUNT define)?

I thought that having a table would be more concise and easier to follow.
I can change this to a switch case statement.

Take care,
David
diff mbox

Patch

diff --git a/drivers/regulator/Kconfig b/drivers/regulator/Kconfig
index 2964eaea..f9ef567 100644
--- a/drivers/regulator/Kconfig
+++ b/drivers/regulator/Kconfig
@@ -682,6 +682,15 @@  config REGULATOR_QCOM_RPM
 	  Qualcomm RPM as a module. The module will be named
 	  "qcom_rpm-regulator".
 
+config REGULATOR_QCOM_RPMH
+	tristate "Qualcomm Technologies, Inc. RPMh regulator driver"
+	depends on QCOM_RPMH || COMPILE_TEST
+	help
+	  This driver supports control of PMIC regulators via the RPMh hardware
+	  block found on Qualcomm Technologies Inc. SoCs.  RPMh regulator
+	  control allows for voting on regulator state between multiple
+	  processors within the SoC.
+
 config REGULATOR_QCOM_SMD_RPM
 	tristate "Qualcomm SMD based RPM regulator driver"
 	depends on QCOM_SMD_RPM
diff --git a/drivers/regulator/Makefile b/drivers/regulator/Makefile
index bd818ce..06e76a6 100644
--- a/drivers/regulator/Makefile
+++ b/drivers/regulator/Makefile
@@ -78,6 +78,7 @@  obj-$(CONFIG_REGULATOR_MT6323)	+= mt6323-regulator.o
 obj-$(CONFIG_REGULATOR_MT6380)	+= mt6380-regulator.o
 obj-$(CONFIG_REGULATOR_MT6397)	+= mt6397-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_RPM) += qcom_rpm-regulator.o
+obj-$(CONFIG_REGULATOR_QCOM_RPMH) += qcom-rpmh-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_SMD_RPM) += qcom_smd-regulator.o
 obj-$(CONFIG_REGULATOR_QCOM_SPMI) += qcom_spmi-regulator.o
 obj-$(CONFIG_REGULATOR_PALMAS) += palmas-regulator.o
diff --git a/drivers/regulator/qcom-rpmh-regulator.c b/drivers/regulator/qcom-rpmh-regulator.c
new file mode 100644
index 0000000..422bb81
--- /dev/null
+++ b/drivers/regulator/qcom-rpmh-regulator.c
@@ -0,0 +1,746 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2018, The Linux Foundation. All rights reserved. */
+
+#define pr_fmt(fmt) "%s: " fmt, __func__
+
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/regulator/driver.h>
+#include <linux/regulator/machine.h>
+#include <linux/regulator/of_regulator.h>
+
+#include <soc/qcom/cmd-db.h>
+#include <soc/qcom/rpmh.h>
+
+#include <dt-bindings/regulator/qcom,rpmh-regulator.h>
+
+/**
+ * enum rpmh_regulator_type - supported RPMh accelerator types
+ * %VRM:	RPMh VRM accelerator which supports voting on enable, voltage,
+ *		and mode of LDO, SMPS, and BOB type PMIC regulators.
+ * %XOB:	RPMh XOB accelerator which supports voting on the enable state
+ *		of PMIC regulators.
+ */
+enum rpmh_regulator_type {
+	VRM,
+	XOB,
+};
+
+#define RPMH_REGULATOR_REG_VRM_VOLTAGE		0x0
+#define RPMH_REGULATOR_REG_ENABLE		0x4
+#define RPMH_REGULATOR_REG_VRM_MODE		0x8
+
+#define RPMH_REGULATOR_MODE_COUNT		4
+
+#define PMIC4_LDO_MODE_RETENTION		4
+#define PMIC4_LDO_MODE_LPM			5
+#define PMIC4_LDO_MODE_HPM			7
+
+#define PMIC4_SMPS_MODE_RETENTION		4
+#define PMIC4_SMPS_MODE_PFM			5
+#define PMIC4_SMPS_MODE_AUTO			6
+#define PMIC4_SMPS_MODE_PWM			7
+
+#define PMIC4_BOB_MODE_PASS			0
+#define PMIC4_BOB_MODE_PFM			1
+#define PMIC4_BOB_MODE_AUTO			2
+#define PMIC4_BOB_MODE_PWM			3
+
+/**
+ * struct rpmh_vreg_hw_data - RPMh regulator hardware configurations
+ * @regulator_type:		RPMh accelerator type used to manage this
+ *				regulator
+ * @ops:			Pointer to regulator ops callback structure
+ * @voltage_range:		The single range of voltages supported by this
+ *				PMIC regulator type
+ * @n_voltages:			The number of unique voltage set points defined
+ *				by voltage_range
+ * @hpm_min_load_uA:		Minimum load current in microamps that requires
+ *				high power mode (HPM) operation.  This is used
+ *				for LDO hardware type regulators only.
+ * @pmic_mode_map:		Array indexed by regulator framework mode
+ *				containing PMIC hardware modes.  Must be large
+ *				enough to index all framework modes supported
+ *				by this regulator hardware type.
+ * @of_map_mode:		Maps an RPMH_REGULATOR_MODE_* mode value defined
+ *				in device tree to a regulator framework mode
+ */
+struct rpmh_vreg_hw_data {
+	enum rpmh_regulator_type		regulator_type;
+	const struct regulator_ops		*ops;
+	const struct regulator_linear_range	voltage_range;
+	int					n_voltages;
+	int					hpm_min_load_uA;
+	const int				*pmic_mode_map;
+	unsigned int			      (*of_map_mode)(unsigned int mode);
+};
+
+/**
+ * struct rpmh_vreg - individual RPMh regulator data structure encapsulating a
+ *		single regulator device
+ * @dev:			Device pointer for the top-level PMIC RPMh
+ *				regulator parent device.  This is used as a
+ *				handle in RPMh write requests.
+ * @addr:			Base address of the regulator resource within
+ *				an RPMh accelerator
+ * @rdesc:			Regulator descriptor
+ * @hw_data:			PMIC regulator configuration data for this RPMh
+ *				regulator
+ * @always_wait_for_ack:	Boolean flag indicating if a request must always
+ *				wait for an ACK from RPMh before continuing even
+ *				if it corresponds to a strictly lower power
+ *				state (e.g. enabled --> disabled).
+ * @enabled:			Flag indicating if the regulator is enabled or
+ *				not
+ * @bypassed:			Boolean indicating if the regulator is in
+ *				bypass (pass-through) mode or not.  This is
+ *				only used by BOB rpmh-regulator resources.
+ * @voltage_selector:		Selector used for get_voltage_sel() and
+ *				set_voltage_sel() callbacks
+ * @mode:			RPMh VRM regulator current framework mode
+ */
+struct rpmh_vreg {
+	struct device			*dev;
+	u32				addr;
+	struct regulator_desc		rdesc;
+	const struct rpmh_vreg_hw_data	*hw_data;
+	bool				always_wait_for_ack;
+
+	int				enabled;
+	bool				bypassed;
+	int				voltage_selector;
+	unsigned int			mode;
+};
+
+/**
+ * struct rpmh_vreg_init_data - initialization data for an RPMh regulator
+ * @name:			Name for the regulator which also corresponds
+ *				to the device tree subnode name of the regulator
+ * @resource_name:		RPMh regulator resource name format string.
+ *				This must include exactly one field: '%s' which
+ *				is filled at run-time with the PMIC ID provided
+ *				by device tree property qcom,pmic-id.  Example:
+ *				"ldo%s1" for RPMh resource "ldoa1".
+ * @supply_name:		Parent supply regulator name
+ * @hw_data:			Configuration data for this PMIC regulator type
+ */
+struct rpmh_vreg_init_data {
+	const char			*name;
+	const char			*resource_name;
+	const char			*supply_name;
+	const struct rpmh_vreg_hw_data	*hw_data;
+};
+
+/**
+ * rpmh_regulator_send_request() - send the request to RPMh
+ * @vreg:		Pointer to the RPMh regulator
+ * @cmd:		Pointer to the RPMh command to send
+ * @wait_for_ack:	Boolean indicating if execution must wait until the
+ *			request has been acknowledged as complete
+ *
+ * Return: 0 on success, errno on failure
+ */
+static int rpmh_regulator_send_request(struct rpmh_vreg *vreg,
+			struct tcs_cmd *cmd, bool wait_for_ack)
+{
+	int ret;
+
+	if (wait_for_ack || vreg->always_wait_for_ack)
+		ret = rpmh_write(vreg->dev, RPMH_ACTIVE_ONLY_STATE, cmd, 1);
+	else
+		ret = rpmh_write_async(vreg->dev, RPMH_ACTIVE_ONLY_STATE, cmd,
+					1);
+
+	return ret;
+}
+
+static int _rpmh_regulator_vrm_set_voltage_sel(struct regulator_dev *rdev,
+				unsigned int selector, bool wait_for_ack)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+	struct tcs_cmd cmd = {
+		.addr = vreg->addr + RPMH_REGULATOR_REG_VRM_VOLTAGE,
+	};
+	int ret;
+
+	/* VRM voltage control register is set with voltage in millivolts. */
+	cmd.data = DIV_ROUND_UP(regulator_list_voltage_linear_range(rdev,
+							selector), 1000);
+
+	ret = rpmh_regulator_send_request(vreg, &cmd, wait_for_ack);
+	if (!ret)
+		vreg->voltage_selector = selector;
+
+	return ret;
+}
+
+static int rpmh_regulator_vrm_set_voltage_sel(struct regulator_dev *rdev,
+					unsigned int selector)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+
+	if (vreg->enabled == -EINVAL) {
+		/*
+		 * Cache the voltage and send it later when the regulator is
+		 * enabled or disabled.
+		 */
+		vreg->voltage_selector = selector;
+		return 0;
+	}
+
+	return _rpmh_regulator_vrm_set_voltage_sel(rdev, selector,
+					selector > vreg->voltage_selector);
+}
+
+static int rpmh_regulator_vrm_get_voltage_sel(struct regulator_dev *rdev)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+
+	return vreg->voltage_selector;
+}
+
+static int rpmh_regulator_is_enabled(struct regulator_dev *rdev)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+
+	return vreg->enabled;
+}
+
+static int rpmh_regulator_set_enable_state(struct regulator_dev *rdev,
+					bool enable)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+	struct tcs_cmd cmd = {
+		.addr = vreg->addr + RPMH_REGULATOR_REG_ENABLE,
+		.data = enable,
+	};
+	int ret;
+
+	if (vreg->enabled == -EINVAL &&
+	    vreg->voltage_selector != -ENOTRECOVERABLE) {
+		ret = _rpmh_regulator_vrm_set_voltage_sel(rdev,
+						vreg->voltage_selector, true);
+		if (ret < 0)
+			return ret;
+	}
+
+	ret = rpmh_regulator_send_request(vreg, &cmd, enable);
+	if (!ret)
+		vreg->enabled = enable;
+
+	return ret;
+}
+
+static int rpmh_regulator_enable(struct regulator_dev *rdev)
+{
+	return rpmh_regulator_set_enable_state(rdev, true);
+}
+
+static int rpmh_regulator_disable(struct regulator_dev *rdev)
+{
+	return rpmh_regulator_set_enable_state(rdev, false);
+}
+
+static int rpmh_regulator_vrm_set_mode_bypass(struct rpmh_vreg *vreg,
+					unsigned int mode, bool bypassed)
+{
+	struct tcs_cmd cmd = {
+		.addr = vreg->addr + RPMH_REGULATOR_REG_VRM_MODE,
+	};
+	int pmic_mode;
+
+	if (mode > REGULATOR_MODE_STANDBY)
+		return -EINVAL;
+
+	pmic_mode = vreg->hw_data->pmic_mode_map[mode];
+	if (pmic_mode < 0)
+		return pmic_mode;
+
+	cmd.data = bypassed ? PMIC4_BOB_MODE_PASS : pmic_mode;
+
+	return rpmh_regulator_send_request(vreg, &cmd, true);
+}
+
+static int rpmh_regulator_vrm_set_mode(struct regulator_dev *rdev,
+					unsigned int mode)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+	int ret;
+
+	if (mode == vreg->mode)
+		return 0;
+
+	ret = rpmh_regulator_vrm_set_mode_bypass(vreg, mode, vreg->bypassed);
+	if (!ret)
+		vreg->mode = mode;
+
+	return ret;
+}
+
+static unsigned int rpmh_regulator_vrm_get_mode(struct regulator_dev *rdev)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+
+	return vreg->mode;
+}
+
+/**
+ * rpmh_regulator_vrm_set_load() - set the regulator mode based upon the load
+ *		current requested
+ * @rdev:		Regulator device pointer for the rpmh-regulator
+ * @load_uA:		Aggregated load current in microamps
+ *
+ * This function is used in the regulator_ops for VRM type RPMh regulator
+ * devices.
+ *
+ * Return: 0 on success, errno on failure
+ */
+static int rpmh_regulator_vrm_set_load(struct regulator_dev *rdev, int load_uA)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+	unsigned int mode;
+
+	if (load_uA >= vreg->hw_data->hpm_min_load_uA)
+		mode = REGULATOR_MODE_FAST;
+	else
+		mode = REGULATOR_MODE_IDLE;
+
+	return rpmh_regulator_vrm_set_mode(rdev, mode);
+}
+
+static int rpmh_regulator_vrm_set_bypass(struct regulator_dev *rdev,
+				bool enable)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+	int ret;
+
+	if (vreg->bypassed == enable)
+		return 0;
+
+	ret = rpmh_regulator_vrm_set_mode_bypass(vreg, vreg->mode, enable);
+	if (!ret)
+		vreg->bypassed = enable;
+
+	return ret;
+}
+
+static int rpmh_regulator_vrm_get_bypass(struct regulator_dev *rdev,
+				bool *enable)
+{
+	struct rpmh_vreg *vreg = rdev_get_drvdata(rdev);
+
+	*enable = vreg->bypassed;
+
+	return 0;
+}
+
+static const struct regulator_ops rpmh_regulator_vrm_ops = {
+	.enable			= rpmh_regulator_enable,
+	.disable		= rpmh_regulator_disable,
+	.is_enabled		= rpmh_regulator_is_enabled,
+	.set_voltage_sel	= rpmh_regulator_vrm_set_voltage_sel,
+	.get_voltage_sel	= rpmh_regulator_vrm_get_voltage_sel,
+	.list_voltage		= regulator_list_voltage_linear_range,
+	.set_mode		= rpmh_regulator_vrm_set_mode,
+	.get_mode		= rpmh_regulator_vrm_get_mode,
+};
+
+static const struct regulator_ops rpmh_regulator_vrm_drms_ops = {
+	.enable			= rpmh_regulator_enable,
+	.disable		= rpmh_regulator_disable,
+	.is_enabled		= rpmh_regulator_is_enabled,
+	.set_voltage_sel	= rpmh_regulator_vrm_set_voltage_sel,
+	.get_voltage_sel	= rpmh_regulator_vrm_get_voltage_sel,
+	.list_voltage		= regulator_list_voltage_linear_range,
+	.set_mode		= rpmh_regulator_vrm_set_mode,
+	.get_mode		= rpmh_regulator_vrm_get_mode,
+	.set_load		= rpmh_regulator_vrm_set_load,
+};
+
+static const struct regulator_ops rpmh_regulator_vrm_bypass_ops = {
+	.enable			= rpmh_regulator_enable,
+	.disable		= rpmh_regulator_disable,
+	.is_enabled		= rpmh_regulator_is_enabled,
+	.set_voltage_sel	= rpmh_regulator_vrm_set_voltage_sel,
+	.get_voltage_sel	= rpmh_regulator_vrm_get_voltage_sel,
+	.list_voltage		= regulator_list_voltage_linear_range,
+	.set_mode		= rpmh_regulator_vrm_set_mode,
+	.get_mode		= rpmh_regulator_vrm_get_mode,
+	.set_bypass		= rpmh_regulator_vrm_set_bypass,
+	.get_bypass		= rpmh_regulator_vrm_get_bypass,
+};
+
+static const struct regulator_ops rpmh_regulator_xob_ops = {
+	.enable			= rpmh_regulator_enable,
+	.disable		= rpmh_regulator_disable,
+	.is_enabled		= rpmh_regulator_is_enabled,
+};
+
+/**
+ * rpmh_regulator_init_vreg() - initialize all attributes of an rpmh-regulator
+ * vreg:		Pointer to the individual rpmh-regulator resource
+ * dev:			Pointer to the top level rpmh-regulator PMIC device
+ * node:		Pointer to the individual rpmh-regulator resource
+ *			device node
+ * pmic_id:		String used to identify the top level rpmh-regulator
+ *			PMIC device on the board
+ * pmic_rpmh_data:	Pointer to a null-terminated array of rpmh-regulator
+ *			resources defined for the top level PMIC device
+ *
+ * Return: 0 on success, errno on failure
+ */
+static int rpmh_regulator_init_vreg(struct rpmh_vreg *vreg, struct device *dev,
+			struct device_node *node, const char *pmic_id,
+			const struct rpmh_vreg_init_data *pmic_rpmh_data)
+{
+	struct regulator_config reg_config = {};
+	char rpmh_resource_name[20] = "";
+	const struct rpmh_vreg_init_data *rpmh_data;
+	struct regulator_init_data *init_data;
+	struct regulator_dev *rdev;
+	int ret;
+
+	vreg->dev = dev;
+
+	for (rpmh_data = pmic_rpmh_data; rpmh_data->name; rpmh_data++)
+		if (!strcmp(rpmh_data->name, node->name))
+			break;
+
+	if (!rpmh_data->name) {
+		dev_err(dev, "Unknown regulator %s\n", node->name);
+		return -EINVAL;
+	}
+
+	scnprintf(rpmh_resource_name, sizeof(rpmh_resource_name),
+		rpmh_data->resource_name, pmic_id);
+
+	vreg->addr = cmd_db_read_addr(rpmh_resource_name);
+	if (!vreg->addr) {
+		dev_err(dev, "%s: could not find RPMh address for resource %s\n",
+			node->name, rpmh_resource_name);
+		return -ENODEV;
+	}
+
+	vreg->rdesc.name = rpmh_data->name;
+	vreg->rdesc.supply_name = rpmh_data->supply_name;
+	vreg->hw_data = rpmh_data->hw_data;
+
+	vreg->enabled = -EINVAL;
+	vreg->voltage_selector = -ENOTRECOVERABLE;
+	vreg->mode = REGULATOR_MODE_INVALID;
+
+	if (rpmh_data->hw_data->n_voltages) {
+		vreg->rdesc.linear_ranges = &rpmh_data->hw_data->voltage_range;
+		vreg->rdesc.n_linear_ranges = 1;
+		vreg->rdesc.n_voltages = rpmh_data->hw_data->n_voltages;
+	}
+
+	vreg->always_wait_for_ack = of_property_read_bool(node,
+						"qcom,always-wait-for-ack");
+
+	vreg->rdesc.owner	= THIS_MODULE;
+	vreg->rdesc.type	= REGULATOR_VOLTAGE;
+	vreg->rdesc.ops		= vreg->hw_data->ops;
+	vreg->rdesc.of_map_mode	= vreg->hw_data->of_map_mode;
+
+	init_data = of_get_regulator_init_data(dev, node, &vreg->rdesc);
+	if (!init_data)
+		return -ENOMEM;
+
+	if (rpmh_data->hw_data->regulator_type == XOB &&
+	    init_data->constraints.min_uV &&
+	    init_data->constraints.min_uV == init_data->constraints.max_uV) {
+		vreg->rdesc.fixed_uV = init_data->constraints.min_uV;
+		vreg->rdesc.n_voltages = 1;
+	}
+
+	reg_config.dev		= dev;
+	reg_config.init_data	= init_data;
+	reg_config.of_node	= node;
+	reg_config.driver_data	= vreg;
+
+	rdev = devm_regulator_register(dev, &vreg->rdesc, &reg_config);
+	if (IS_ERR(rdev)) {
+		ret = PTR_ERR(rdev);
+		dev_err(dev, "%s: devm_regulator_register() failed, ret=%d\n",
+			node->name, ret);
+		return ret;
+	}
+
+	dev_dbg(dev, "%s regulator registered for RPMh resource %s @ 0x%05X\n",
+		node->name, rpmh_resource_name, vreg->addr);
+
+	return 0;
+}
+
+static const int pmic_mode_map_pmic4_ldo[REGULATOR_MODE_STANDBY + 1] = {
+	[REGULATOR_MODE_INVALID] = -EINVAL,
+	[REGULATOR_MODE_STANDBY] = PMIC4_LDO_MODE_RETENTION,
+	[REGULATOR_MODE_IDLE]    = PMIC4_LDO_MODE_LPM,
+	[REGULATOR_MODE_NORMAL]  = -EINVAL,
+	[REGULATOR_MODE_FAST]    = PMIC4_LDO_MODE_HPM,
+};
+
+static unsigned int rpmh_regulator_pmic4_ldo_of_map_mode(unsigned int mode)
+{
+	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
+		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
+		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
+		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_INVALID,
+		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
+	};
+
+	if (mode >= RPMH_REGULATOR_MODE_COUNT)
+		return -EINVAL;
+
+	return of_mode_map[mode];
+}
+
+static const int pmic_mode_map_pmic4_smps[REGULATOR_MODE_STANDBY + 1] = {
+	[REGULATOR_MODE_INVALID] = -EINVAL,
+	[REGULATOR_MODE_STANDBY] = PMIC4_SMPS_MODE_RETENTION,
+	[REGULATOR_MODE_IDLE]    = PMIC4_SMPS_MODE_PFM,
+	[REGULATOR_MODE_NORMAL]  = PMIC4_SMPS_MODE_AUTO,
+	[REGULATOR_MODE_FAST]    = PMIC4_SMPS_MODE_PWM,
+};
+
+static unsigned int rpmh_regulator_pmic4_smps_of_map_mode(unsigned int mode)
+{
+	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
+		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_STANDBY,
+		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
+		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_NORMAL,
+		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
+	};
+
+	if (mode >= RPMH_REGULATOR_MODE_COUNT)
+		return -EINVAL;
+
+	return of_mode_map[mode];
+}
+
+static const int pmic_mode_map_pmic4_bob[REGULATOR_MODE_STANDBY + 1] = {
+	[REGULATOR_MODE_INVALID] = -EINVAL,
+	[REGULATOR_MODE_STANDBY] = -EINVAL,
+	[REGULATOR_MODE_IDLE]    = PMIC4_BOB_MODE_PFM,
+	[REGULATOR_MODE_NORMAL]  = PMIC4_BOB_MODE_AUTO,
+	[REGULATOR_MODE_FAST]    = PMIC4_BOB_MODE_PWM,
+};
+
+static unsigned int rpmh_regulator_pmic4_bob_of_map_mode(unsigned int mode)
+{
+	static const unsigned int of_mode_map[RPMH_REGULATOR_MODE_COUNT] = {
+		[RPMH_REGULATOR_MODE_RET]  = REGULATOR_MODE_INVALID,
+		[RPMH_REGULATOR_MODE_LPM]  = REGULATOR_MODE_IDLE,
+		[RPMH_REGULATOR_MODE_AUTO] = REGULATOR_MODE_NORMAL,
+		[RPMH_REGULATOR_MODE_HPM]  = REGULATOR_MODE_FAST,
+	};
+
+	if (mode >= RPMH_REGULATOR_MODE_COUNT)
+		return -EINVAL;
+
+	return of_mode_map[mode];
+}
+
+static const struct rpmh_vreg_hw_data pmic4_pldo = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_drms_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(1664000, 0, 255, 8000),
+	.n_voltages = 256,
+	.hpm_min_load_uA = 10000,
+	.pmic_mode_map = pmic_mode_map_pmic4_ldo,
+	.of_map_mode = rpmh_regulator_pmic4_ldo_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_pldo_lv = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_drms_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(1256000, 0, 127, 8000),
+	.n_voltages = 128,
+	.hpm_min_load_uA = 10000,
+	.pmic_mode_map = pmic_mode_map_pmic4_ldo,
+	.of_map_mode = rpmh_regulator_pmic4_ldo_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_nldo = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_drms_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(312000, 0, 127, 8000),
+	.n_voltages = 128,
+	.hpm_min_load_uA = 30000,
+	.pmic_mode_map = pmic_mode_map_pmic4_ldo,
+	.of_map_mode = rpmh_regulator_pmic4_ldo_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_hfsmps3 = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(320000, 0, 215, 8000),
+	.n_voltages = 216,
+	.pmic_mode_map = pmic_mode_map_pmic4_smps,
+	.of_map_mode = rpmh_regulator_pmic4_smps_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_ftsmps426 = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(320000, 0, 258, 4000),
+	.n_voltages = 259,
+	.pmic_mode_map = pmic_mode_map_pmic4_smps,
+	.of_map_mode = rpmh_regulator_pmic4_smps_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_bob = {
+	.regulator_type = VRM,
+	.ops = &rpmh_regulator_vrm_bypass_ops,
+	.voltage_range = REGULATOR_LINEAR_RANGE(1824000, 0, 83, 32000),
+	.n_voltages = 84,
+	.pmic_mode_map = pmic_mode_map_pmic4_bob,
+	.of_map_mode = rpmh_regulator_pmic4_bob_of_map_mode,
+};
+
+static const struct rpmh_vreg_hw_data pmic4_lvs = {
+	.regulator_type = XOB,
+	.ops = &rpmh_regulator_xob_ops,
+	/* LVS hardware does not support voltage or mode configuration. */
+};
+
+#define RPMH_VREG(_name, _resource_name, _hw_data, _supply_name) \
+{ \
+	.name		= _name, \
+	.resource_name	= _resource_name, \
+	.hw_data	= _hw_data, \
+	.supply_name	= _supply_name, \
+}
+
+static const struct rpmh_vreg_init_data pm8998_vreg_data[] = {
+	RPMH_VREG("smps1",  "smp%s1",  &pmic4_ftsmps426, "vdd-s1"),
+	RPMH_VREG("smps2",  "smp%s2",  &pmic4_ftsmps426, "vdd-s2"),
+	RPMH_VREG("smps3",  "smp%s3",  &pmic4_hfsmps3,   "vdd-s3"),
+	RPMH_VREG("smps4",  "smp%s4",  &pmic4_hfsmps3,   "vdd-s4"),
+	RPMH_VREG("smps5",  "smp%s5",  &pmic4_hfsmps3,   "vdd-s5"),
+	RPMH_VREG("smps6",  "smp%s6",  &pmic4_ftsmps426, "vdd-s6"),
+	RPMH_VREG("smps7",  "smp%s7",  &pmic4_ftsmps426, "vdd-s7"),
+	RPMH_VREG("smps8",  "smp%s8",  &pmic4_ftsmps426, "vdd-s8"),
+	RPMH_VREG("smps9",  "smp%s9",  &pmic4_ftsmps426, "vdd-s9"),
+	RPMH_VREG("smps10", "smp%s10", &pmic4_ftsmps426, "vdd-s10"),
+	RPMH_VREG("smps11", "smp%s11", &pmic4_ftsmps426, "vdd-s11"),
+	RPMH_VREG("smps12", "smp%s12", &pmic4_ftsmps426, "vdd-s12"),
+	RPMH_VREG("smps13", "smp%s13", &pmic4_ftsmps426, "vdd-s13"),
+	RPMH_VREG("ldo1",   "ldo%s1",  &pmic4_nldo,      "vdd-l1-l27"),
+	RPMH_VREG("ldo2",   "ldo%s2",  &pmic4_nldo,      "vdd-l2-l8-l17"),
+	RPMH_VREG("ldo3",   "ldo%s3",  &pmic4_nldo,      "vdd-l3-l11"),
+	RPMH_VREG("ldo4",   "ldo%s4",  &pmic4_nldo,      "vdd-l4-l5"),
+	RPMH_VREG("ldo5",   "ldo%s5",  &pmic4_nldo,      "vdd-l4-l5"),
+	RPMH_VREG("ldo6",   "ldo%s6",  &pmic4_pldo,      "vdd-l6"),
+	RPMH_VREG("ldo7",   "ldo%s7",  &pmic4_pldo_lv,   "vdd-l7-l12-l14-l15"),
+	RPMH_VREG("ldo8",   "ldo%s8",  &pmic4_nldo,      "vdd-l2-l8-l17"),
+	RPMH_VREG("ldo9",   "ldo%s9",  &pmic4_pldo,      "vdd-l9"),
+	RPMH_VREG("ldo10",  "ldo%s10", &pmic4_pldo,      "vdd-l10-l23-l25"),
+	RPMH_VREG("ldo11",  "ldo%s11", &pmic4_nldo,      "vdd-l3-l11"),
+	RPMH_VREG("ldo12",  "ldo%s12", &pmic4_pldo_lv,   "vdd-l7-l12-l14-l15"),
+	RPMH_VREG("ldo13",  "ldo%s13", &pmic4_pldo,      "vdd-l13-l19-l21"),
+	RPMH_VREG("ldo14",  "ldo%s14", &pmic4_pldo_lv,   "vdd-l7-l12-l14-l15"),
+	RPMH_VREG("ldo15",  "ldo%s15", &pmic4_pldo_lv,   "vdd-l7-l12-l14-l15"),
+	RPMH_VREG("ldo16",  "ldo%s16", &pmic4_pldo,      "vdd-l16-l28"),
+	RPMH_VREG("ldo17",  "ldo%s17", &pmic4_nldo,      "vdd-l2-l8-l17"),
+	RPMH_VREG("ldo18",  "ldo%s18", &pmic4_pldo,      "vdd-l18-l22"),
+	RPMH_VREG("ldo19",  "ldo%s19", &pmic4_pldo,      "vdd-l13-l19-l21"),
+	RPMH_VREG("ldo20",  "ldo%s20", &pmic4_pldo,      "vdd-l20-l24"),
+	RPMH_VREG("ldo21",  "ldo%s21", &pmic4_pldo,      "vdd-l13-l19-l21"),
+	RPMH_VREG("ldo22",  "ldo%s22", &pmic4_pldo,      "vdd-l18-l22"),
+	RPMH_VREG("ldo23",  "ldo%s23", &pmic4_pldo,      "vdd-l10-l23-l25"),
+	RPMH_VREG("ldo24",  "ldo%s24", &pmic4_pldo,      "vdd-l20-l24"),
+	RPMH_VREG("ldo25",  "ldo%s25", &pmic4_pldo,      "vdd-l10-l23-l25"),
+	RPMH_VREG("ldo26",  "ldo%s26", &pmic4_nldo,      "vdd-l26"),
+	RPMH_VREG("ldo27",  "ldo%s27", &pmic4_nldo,      "vdd-l1-l27"),
+	RPMH_VREG("ldo28",  "ldo%s28", &pmic4_pldo,      "vdd-l16-l28"),
+	RPMH_VREG("lvs1",   "vs%s1",   &pmic4_lvs,       "vin-lvs-1-2"),
+	RPMH_VREG("lvs2",   "vs%s2",   &pmic4_lvs,       "vin-lvs-1-2"),
+	{},
+};
+
+static const struct rpmh_vreg_init_data pmi8998_vreg_data[] = {
+	RPMH_VREG("bob",    "bob%s1",  &pmic4_bob,       "vdd-bob"),
+	{},
+};
+
+static const struct rpmh_vreg_init_data pm8005_vreg_data[] = {
+	RPMH_VREG("smps1",  "smp%s1",  &pmic4_ftsmps426, "vdd-s1"),
+	RPMH_VREG("smps2",  "smp%s2",  &pmic4_ftsmps426, "vdd-s2"),
+	RPMH_VREG("smps3",  "smp%s3",  &pmic4_ftsmps426, "vdd-s3"),
+	RPMH_VREG("smps4",  "smp%s4",  &pmic4_ftsmps426, "vdd-s4"),
+	{},
+};
+
+static int rpmh_regulator_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	const struct rpmh_vreg_init_data *vreg_data;
+	struct device_node *node;
+	struct rpmh_vreg *vreg;
+	const char *pmic_id;
+	int ret;
+
+	vreg_data = of_device_get_match_data(dev);
+	if (!vreg_data)
+		return -ENODEV;
+
+	ret = of_property_read_string(dev->of_node, "qcom,pmic-id", &pmic_id);
+	if (ret < 0) {
+		dev_err(dev, "qcom,pmic-id missing in DT node\n");
+		return ret;
+	}
+
+	for_each_available_child_of_node(dev->of_node, node) {
+		vreg = devm_kzalloc(dev, sizeof(*vreg), GFP_KERNEL);
+		if (!vreg) {
+			of_node_put(node);
+			return -ENOMEM;
+		}
+
+		ret = rpmh_regulator_init_vreg(vreg, dev, node, pmic_id,
+						vreg_data);
+		if (ret < 0) {
+			of_node_put(node);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static const struct of_device_id rpmh_regulator_match_table[] = {
+	{
+		.compatible = "qcom,pm8998-rpmh-regulators",
+		.data = pm8998_vreg_data,
+	},
+	{
+		.compatible = "qcom,pmi8998-rpmh-regulators",
+		.data = pmi8998_vreg_data,
+	},
+	{
+		.compatible = "qcom,pm8005-rpmh-regulators",
+		.data = pm8005_vreg_data,
+	},
+	{}
+};
+MODULE_DEVICE_TABLE(of, rpmh_regulator_match_table);
+
+static struct platform_driver rpmh_regulator_driver = {
+	.driver = {
+		.name = "qcom-rpmh-regulator",
+		.of_match_table	= of_match_ptr(rpmh_regulator_match_table),
+	},
+	.probe = rpmh_regulator_probe,
+};
+module_platform_driver(rpmh_regulator_driver);
+
+MODULE_DESCRIPTION("Qualcomm RPMh regulator driver");
+MODULE_LICENSE("GPL v2");