All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Narendran Rajan" <nrajan@codeaurora.com>
To: 'Srinivas Kandagatla' <srinivas.kandagatla@linaro.org>,
	linux-arm-msm@vger.kernel.org
Cc: patches@linaro.org, linaro-kernel@lists.linaro.org
Subject: RE: [RFC PATCH 2/2] WIP: Add wrappers for qfprom access via syscon
Date: Fri, 30 Jan 2015 22:53:45 -0800	[thread overview]
Message-ID: <005701d03d22$a90565b0$fb103110$@codeaurora.com> (raw)
In-Reply-To: <1422267208-6189-1-git-send-email-srinivas.kandagatla@linaro.org>

Hi Srini,

The idea of abstracting qfrom reads in general is good, but looking at the 
Implementation this seems to add unwanted overhead.

Every read (if the driver is reading byte by byte especially) 
would involve a syscon node lookup,  memalloc and then regmap read.

To make it more efficient, two suggestions:

1) Instead of allocating memory in this library, let the clients allocate
memory
and provide a pointer (move memory management to client)
2) Provide one a time registration and then every read just specifies
relative offset
and length for every read

Usage pattern could be something like
struct qfprom *qfprom_register(struct device *dev, int index);
int devm_qfprom_get_data (struct qfprom *qfprom, int offset, int len, void
*data);

> -----Original Message-----
> From: linux-arm-msm-owner@vger.kernel.org [mailto:linux-arm-msm-
> owner@vger.kernel.org] On Behalf Of Srinivas Kandagatla
> Sent: Monday, January 26, 2015 2:13 AM
> To: linux-arm-msm@vger.kernel.org
> Cc: patches@linaro.org; linaro-kernel@lists.linaro.org; Srinivas
Kandagatla
> Subject: [RFC PATCH 2/2] WIP: Add wrappers for qfprom access via syscon
> 
> Syscon fits very well to access qfprom. This also means drivers which
needs
> to access qfprom have to talk to syscon and get regmap, offset, size and
then
> do regmap reads. This will be kinda redone in every driver. Having a
wrapper
> for this would avoid lot of code duplications and also provide a higher
level
> and user friendly apis for qfprom. This patch attempt to provide such
> wrappers.
> This wrappers are easy way to use syscon for qfprom purposes.
> 
> Advantages of this approch is:
> 
> - driver need not have hardcoded qfprom offsets or have soc specific
>   compatible strings to determine the offset.
> - access multiple qfprom resources which is kinda tricky with standard
>   syscon.
> - no code duplication.
> - light weight, single call.
> - not a platform device driver level binding.
> 
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> ---
>  .../devicetree/bindings/soc/qcom/qfprom.txt        |  29 +++++
>  drivers/soc/qcom/Kconfig                           |   7 ++
>  drivers/soc/qcom/Makefile                          |   1 +
>  drivers/soc/qcom/qfprom.c                          | 134
+++++++++++++++++++++
>  4 files changed, 171 insertions(+)
>  create mode 100644
> Documentation/devicetree/bindings/soc/qcom/qfprom.txt
>  create mode 100644 drivers/soc/qcom/qfprom.c
> 
> diff --git a/Documentation/devicetree/bindings/soc/qcom/qfprom.txt
> b/Documentation/devicetree/bindings/soc/qcom/qfprom.txt
> new file mode 100644
> index 0000000..3ed7309
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/soc/qcom/qfprom.txt
> @@ -0,0 +1,29 @@
> +QCOM QFPROM
> +
> +QFPROM is basically some efuses where things like calibration data,
> +speed bins, etc are stored. This data is accessed by various drivers
> +like the cpufreq, thermal, etc.
> +
> +Required properties:
> +- compatible: must contain "qcom,qfprom" followed by "syscon"
> +- reg: Address range for QFPROM
> +- stride : register address stride.
> +	1 for byte.
> +	2 for 2 bytes
> +	3 for 3 bytes
> +	4 for a word.
> +
> +
> +Example:
> +	qfprom: qfprom@00700000 {
> +		compatible 	= "qcom,qfprom", "syscon";
> +		reg		= <0x00700000 0x1000>;
> +		stride		= <1>;
> +	};
> +
> +	tsens@34000 {
> +		compatible = "qcom,tsens-apq8064";
> +		reg = <0x34000 0x1000>;
> +		qcom,qfprom = <&qfprom 0x18 0x10>, <&qfprom 0x28
> 0x10>;
> +		qcom,qfprom-names = "calib", "backup_calib";
> +	};
> diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig index
> 012fb37..389ec3e 100644
> --- a/drivers/soc/qcom/Kconfig
> +++ b/drivers/soc/qcom/Kconfig
> @@ -19,3 +19,10 @@ config QCOM_PM
>  	  QCOM Platform specific power driver to manage cores and L2 low
> power
>  	  modes. It interface with various system drivers to put the cores
in
>  	  low power modes.
> +
> +config QCOM_QFPROM
> +	tristate "QCOM QFPROM Interface"
> +        depends on ARCH_QCOM && OF
> +        help
> +          Say y here to enable QFPROM support. The QFPROM provides access
> +          functions for QFPROM data to rest of the drivers via syscon
wrappers.
> diff --git a/drivers/soc/qcom/Makefile b/drivers/soc/qcom/Makefile index
> 20b329f..f5aff0a 100644
> --- a/drivers/soc/qcom/Makefile
> +++ b/drivers/soc/qcom/Makefile
> @@ -1,4 +1,5 @@
>  obj-$(CONFIG_QCOM_GSBI)	+=	qcom_gsbi.o
>  obj-$(CONFIG_QCOM_PM)	+=	spm.o
> +obj-$(CONFIG_QCOM_QFPROM) +=	qfprom.o
>  CFLAGS_scm.o :=$(call as-instr,.arch_extension sec,-DREQUIRES_SEC=1)
>  obj-$(CONFIG_QCOM_SCM) += scm.o scm-boot.o diff --git
> a/drivers/soc/qcom/qfprom.c b/drivers/soc/qcom/qfprom.c new file mode
> 100644 index 0000000..d00ed25
> --- /dev/null
> +++ b/drivers/soc/qcom/qfprom.c
> @@ -0,0 +1,134 @@
> +#include <linux/err.h>
> +#include <linux/of.h>
> +#include <linux/device.h>
> +#include <linux/regmap.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/soc/qcom/qfprom.h>
> +#include <linux/slab.h>
> +
> +#define QFPROM_MAX_ARGS	2
> +
> +static char *__qfprom_get_data(struct device *dev,
> +				bool devm, int idx, int *len)
> +{
> +	struct device_node *syscon_np, *np = dev->of_node;
> +	struct regmap *rm;
> +	struct of_phandle_args args;
> +	int rc, stride = 4;
> +	u32 offset, size;
> +	char *data;
> +
> +	if (!np)
> +		return ERR_PTR(-EINVAL);
> +
> +	rc = of_parse_phandle_with_fixed_args(np, "qcom,qfprom",
> +					      QFPROM_MAX_ARGS, idx, &args);
> +	if (rc)
> +		return ERR_PTR(rc);
> +
> +	syscon_np = args.np;
> +
> +	of_property_read_u32(syscon_np, "stride", &stride);
> +
> +	if (stride >= 4)
> +		stride = 4;
> +
> +	if (args.args_count < QFPROM_MAX_ARGS) {
> +		dev_err(dev, "Insufficient qfprom arguments %d\n",
> +			args.args_count);
> +		return ERR_PTR(-EINVAL);
> +	}
> +
> +	rm = syscon_node_to_regmap(syscon_np);
> +	if (IS_ERR(rm))
> +		return ERR_CAST(rm);
> +
> +	offset = args.args[0];
> +	size = args.args[1];
> +
> +	of_node_put(syscon_np);
> +
> +	if (devm)
> +		data = devm_kzalloc(dev, size, GFP_KERNEL | GFP_ATOMIC);
> +	else
> +		data = kzalloc(size, GFP_KERNEL | GFP_ATOMIC);
> +
> +	if (!data)
> +		return ERR_PTR(-ENOMEM);
> +
> +	rc = regmap_bulk_read(rm, offset, data, size/stride);
> +	if (rc < 0) {
> +		if (devm)
> +			devm_kfree(dev, data);
> +		else
> +			kfree(data);
> +
> +		return ERR_PTR(rc);
> +	}
> +
> +	*len = size;
> +
> +	return data;
> +}
> +
> +static char *__qfprom_get_data_byname(struct device *dev,
> +				       bool devm, const char *name, int
*len) {
> +	int index = 0;
> +
> +	if (name)
> +		index = of_property_match_string(dev->of_node,
> +						 "qcom,qfprom-names",
> name);
> +
> +	return __qfprom_get_data(dev, devm, index, len); }
> +
> +char *devm_qfprom_get_data_byname(struct device *dev,
> +					  const char *name, int *len)
> +{
> +	return __qfprom_get_data_byname(dev, true, name, len); }
> +EXPORT_SYMBOL_GPL(devm_qfprom_get_data_byname);
> +
> +char *devm_qfprom_get_data(struct device *dev,
> +				   int index, int *len)
> +{
> +	return __qfprom_get_data(dev, true, index, len); }
> +EXPORT_SYMBOL_GPL(devm_qfprom_get_data);
> +
> +/**
> + * qfprom_get_data_byname(): Reads qfprom data by name
> + *
> + * @dev: device which is requesting qfprom data
> + * @index: name of qfprom resources specified "qcom,qfprom-names" DT
> property.
> + * @len: length of data read from qfprom.
> + *
> + * The return value will be an ERR_PTR() on error or a valid pointer
> + * to a data buffer. The buffer should be freed by the user once its
> + * finished working with it kfree.
> + **/
> +char *qfprom_get_data_byname(struct device *dev,
> +				     const char *name, int *len)
> +{
> +	return __qfprom_get_data_byname(dev, false, name, len); }
> +EXPORT_SYMBOL_GPL(qfprom_get_data_byname);
> +
> +/**
> + * qfprom_get_data(): Reads qfprom data from the index
> + *
> + * @dev: device which is requesting qfprom data
> + * @index: index into qfprom resources specified "qcom,qfprom" DT
> property.
> + * @len: length of data read from qfprom.
> + *
> + * The return value will be an ERR_PTR() on error or a valid pointer
> + * to a data buffer. The buffer should be freed by the user once its
> + * finished working with it kfree.
> + **/
> +char *qfprom_get_data(struct device *dev,
> +				   int index, int *len)
> +{
> +	return __qfprom_get_data(dev, false, index, len); }
> +EXPORT_SYMBOL_GPL(qfprom_get_data);
> --
> 1.9.1
> 
> --
> 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

--Naren

      reply	other threads:[~2015-01-31  6:53 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-26 10:12 [RFC PATCH 0/2] QFPROM wrappers via syscon Srinivas Kandagatla
2015-01-26 10:13 ` [RFC PATCH 1/2] WIP: mfd: syscon: Add register stride to DT bindings Srinivas Kandagatla
2015-01-26 10:13 ` [RFC PATCH 2/2] WIP: Add wrappers for qfprom access via syscon Srinivas Kandagatla
2015-01-31  6:53   ` Narendran Rajan [this message]

Reply instructions:

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

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

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

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

  git send-email \
    --in-reply-to='005701d03d22$a90565b0$fb103110$@codeaurora.com' \
    --to=nrajan@codeaurora.com \
    --cc=linaro-kernel@lists.linaro.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=patches@linaro.org \
    --cc=srinivas.kandagatla@linaro.org \
    /path/to/YOUR_REPLY

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

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