linux-tegra.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH V3 1/2] regmap: Support for caching in reg_raw_write()
@ 2012-02-10 16:00 Laxman Dewangan
  2012-02-10 16:00 ` [PATCH V3 2/2] regmap: add regmap_bulk_write() for register write Laxman Dewangan
       [not found] ` <1328889628-13151-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
  0 siblings, 2 replies; 4+ messages in thread
From: Laxman Dewangan @ 2012-02-10 16:00 UTC (permalink / raw)
  To: broonie-yzvPICuk2AATkU/dhu1WVueM+bqZidxxQQ4Iyu8u01E,
	gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r,
	lars-Qo5EllUWu/uELgA04lAiVw, linux-kernel-u79uwXL29TY76Z2rM5mHXA
  Cc: linux-tegra-u79uwXL29TY76Z2rM5mHXA,
	ldewangan-DDmLM1+adcrQT0dZR+AlfA

Adding support for caching of data into the
non-volatile register from the call of reg_raw_write().

This will allow the larger block of data write into multiple
register without worrying whether register is cached or not
through reg_raw_write().

Signed-off-by: Laxman Dewangan <ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
Changes from V1: 
- Add register number and return value in error case.
- Some variable usage cleanups..
Changes from V2:
- Fixing register number during error printing.
 drivers/base/regmap/regmap.c |   26 +++++++++++++++++++++-----
 1 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index 80129c0..ba92217 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -323,6 +323,26 @@ static int _regmap_raw_write(struct regmap *map, unsigned int reg,
 			if (!map->writeable_reg(map->dev, reg + i))
 				return -EINVAL;
 
+	if (!map->cache_bypass && map->format.parse_val) {
+		unsigned int ival;
+		int val_bytes = map->format.val_bytes;
+		for (i = 0; i < val_len / val_bytes; i++) {
+			memcpy(map->work_buf, val + (i * val_bytes), val_bytes);
+			ival = map->format.parse_val(map->work_buf);
+			ret = regcache_write(map, reg + i, ival);
+			if (ret) {
+				dev_err(map->dev,
+				   "Error in caching of register: %u ret: %d\n",
+					reg + i, ret);
+				return ret;
+			}
+		}
+		if (map->cache_only) {
+			map->cache_dirty = true;
+			return 0;
+		}
+	}
+
 	map->format.format_reg(map->work_buf, reg);
 
 	u8[0] |= map->write_flag_mask;
@@ -373,7 +393,7 @@ int _regmap_write(struct regmap *map, unsigned int reg,
 	int ret;
 	BUG_ON(!map->format.format_write && !map->format.format_val);
 
-	if (!map->cache_bypass) {
+	if (!map->cache_bypass && map->format.format_write) {
 		ret = regcache_write(map, reg, val);
 		if (ret != 0)
 			return ret;
@@ -450,12 +470,8 @@ EXPORT_SYMBOL_GPL(regmap_write);
 int regmap_raw_write(struct regmap *map, unsigned int reg,
 		     const void *val, size_t val_len)
 {
-	size_t val_count = val_len / map->format.val_bytes;
 	int ret;
 
-	WARN_ON(!regmap_volatile_range(map, reg, val_count) &&
-		map->cache_type != REGCACHE_NONE);
-
 	mutex_lock(&map->lock);
 
 	ret = _regmap_raw_write(map, reg, val, val_len);
-- 
1.7.1.1

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* [PATCH V3 2/2] regmap: add regmap_bulk_write() for register write
  2012-02-10 16:00 [PATCH V3 1/2] regmap: Support for caching in reg_raw_write() Laxman Dewangan
@ 2012-02-10 16:00 ` Laxman Dewangan
  2012-02-11 21:47   ` Mark Brown
       [not found] ` <1328889628-13151-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
  1 sibling, 1 reply; 4+ messages in thread
From: Laxman Dewangan @ 2012-02-10 16:00 UTC (permalink / raw)
  To: broonie, gregkh, lars, linux-kernel; +Cc: linux-tegra, ldewangan

The bulk_write() supports the data transfer to multi
register which takes the data into cpu_endianness format
and does formatting of data to device format before
sending to device.
The transfer can be completed in single transfer or multiple
transfer based on data formatting.

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
---
Changes from V1: 
- Change commit message.
- Byte swapping in input data packet.
Changes from V2:
- Remove formatting if val_byte is 1.
- Allocating proper size of memory for val_byte != 1
 drivers/base/regmap/regmap.c |   61 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/regmap.h       |    2 +
 2 files changed, 63 insertions(+), 0 deletions(-)

diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index ba92217..a1656a7 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -482,6 +482,67 @@ int regmap_raw_write(struct regmap *map, unsigned int reg,
 }
 EXPORT_SYMBOL_GPL(regmap_raw_write);
 
+/*
+ * regmap_bulk_write(): Write multiple registers to the device
+ *
+ * @map: Register map to write to
+ * @reg: First register to be write from
+ * @val: Block of data to be written, in native register size for device
+ * @val_count: Number of registers to write
+ *
+ * This function is intended to be used for writing a large block of
+ * data to be device either in single transfer or multiple transfer.
+ *
+ * A value of zero will be returned on success, a negative errno will
+ * be returned in error cases.
+ */
+int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
+		     size_t val_count)
+{
+	int ret = 0, i;
+	unsigned int ival;
+	size_t val_bytes = map->format.val_bytes;
+
+	mutex_lock(&map->lock);
+
+	if (map->format.parse_val) {
+		void *wval;
+		/* No formatting is require if val_byte is 1 */
+		if (val_bytes == 1)
+			wval = (void *)val;
+		else {
+			wval = kmemdup(val, val_count * val_bytes, GFP_KERNEL);
+			if (!wval) {
+				mutex_unlock(&map->lock);
+				dev_err(map->dev,
+					"Error in memory allocation\n");
+				return -ENOMEM;
+			}
+			for (i = 0; i < val_count * val_bytes; i += val_bytes)
+				map->format.parse_val(wval + i);
+		}
+		ret = _regmap_raw_write(map, reg, wval, val_bytes * val_count);
+
+		if (val_bytes != 1)
+			kfree(wval);
+	} else {
+		for (i = 0; i < val_count; i++) {
+			memcpy(&ival, val + (i * val_bytes), val_bytes);
+			ret = _regmap_write(map, reg + i, ival);
+			if (ret) {
+				dev_err(map->dev,
+				    "Error in register %u write, ret %d\n",
+						reg + i, ret);
+				break;
+			}
+		}
+	}
+
+	mutex_unlock(&map->lock);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(regmap_bulk_write);
+
 static int _regmap_raw_read(struct regmap *map, unsigned int reg, void *val,
 			    unsigned int val_len)
 {
diff --git a/include/linux/regmap.h b/include/linux/regmap.h
index a6ed6e6..0925e24 100644
--- a/include/linux/regmap.h
+++ b/include/linux/regmap.h
@@ -135,6 +135,8 @@ int regmap_reinit_cache(struct regmap *map,
 int regmap_write(struct regmap *map, unsigned int reg, unsigned int val);
 int regmap_raw_write(struct regmap *map, unsigned int reg,
 		     const void *val, size_t val_len);
+int regmap_bulk_write(struct regmap *map, unsigned int reg, const void *val,
+			size_t val_count);
 int regmap_read(struct regmap *map, unsigned int reg, unsigned int *val);
 int regmap_raw_read(struct regmap *map, unsigned int reg,
 		    void *val, size_t val_len);
-- 
1.7.1.1

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH V3 2/2] regmap: add regmap_bulk_write() for register write
  2012-02-10 16:00 ` [PATCH V3 2/2] regmap: add regmap_bulk_write() for register write Laxman Dewangan
@ 2012-02-11 21:47   ` Mark Brown
  0 siblings, 0 replies; 4+ messages in thread
From: Mark Brown @ 2012-02-11 21:47 UTC (permalink / raw)
  To: Laxman Dewangan; +Cc: gregkh, lars, linux-kernel, linux-tegra

[-- Attachment #1: Type: text/plain, Size: 589 bytes --]

On Fri, Feb 10, 2012 at 09:30:28PM +0530, Laxman Dewangan wrote:

> +		/* No formatting is require if val_byte is 1 */
> +		if (val_bytes == 1)
> +			wval = (void *)val;
> +		else {

Please don't combine branches with and without braces.

> +			wval = kmemdup(val, val_count * val_bytes, GFP_KERNEL);
> +			if (!wval) {
> +				mutex_unlock(&map->lock);
> +				dev_err(map->dev,
> +					"Error in memory allocation\n");
> +				return -ENOMEM;

As a style thing a goto error pattern is clearer for stuff like this -
there's so many branches that it's complex to keep track of the
unwinding.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH V3 1/2] regmap: Support for caching in reg_raw_write()
       [not found] ` <1328889628-13151-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
@ 2012-02-11 23:05   ` Mark Brown
  0 siblings, 0 replies; 4+ messages in thread
From: Mark Brown @ 2012-02-11 23:05 UTC (permalink / raw)
  To: Laxman Dewangan
  Cc: gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r,
	lars-Qo5EllUWu/uELgA04lAiVw, linux-kernel-u79uwXL29TY76Z2rM5mHXA,
	linux-tegra-u79uwXL29TY76Z2rM5mHXA

[-- Attachment #1: Type: text/plain, Size: 468 bytes --]

On Fri, Feb 10, 2012 at 09:30:27PM +0530, Laxman Dewangan wrote:
> Adding support for caching of data into the
> non-volatile register from the call of reg_raw_write().
> 
> This will allow the larger block of data write into multiple
> register without worrying whether register is cached or not
> through reg_raw_write().

Applied, thanks.  We really should refactor later on to not do the
encode/decode dance if we don't have to but this is fine for now.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2012-02-11 23:05 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-02-10 16:00 [PATCH V3 1/2] regmap: Support for caching in reg_raw_write() Laxman Dewangan
2012-02-10 16:00 ` [PATCH V3 2/2] regmap: add regmap_bulk_write() for register write Laxman Dewangan
2012-02-11 21:47   ` Mark Brown
     [not found] ` <1328889628-13151-1-git-send-email-ldewangan-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
2012-02-11 23:05   ` [PATCH V3 1/2] regmap: Support for caching in reg_raw_write() Mark Brown

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).