diff mbox

[11/20] regmap: _regmap_raw_read: Add handling of busses without bus->read()

Message ID 1439374365-20623-12-git-send-email-mpa@pengutronix.de (mailing list archive)
State New, archived
Headers show

Commit Message

Markus Pargmann Aug. 12, 2015, 10:12 a.m. UTC
Handle easy reads by using bus->reg_read(). Return with an error value
if a read is requested that can not be handled with reg_read().

Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
---
 drivers/base/regmap/regmap.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

Comments

Mark Brown Aug. 12, 2015, 11:27 a.m. UTC | #1
On Wed, Aug 12, 2015 at 12:12:36PM +0200, Markus Pargmann wrote:

> +	/*
> +	 * There are busses that do not have a read function as it is optional.
> +	 * Use their reg_read function instead if the requested number of bytes
> +	 * is correct.
> +	 */
> +	if (!map->bus->read) {
> +		/*
> +		 * bus_reg_read() does not support reading values that are not
> +		 * exactly the size of format.val_bytes
> +		 */
> +		if (val_len != map->format.val_bytes)
> +			return -EINVAL;
> +		return _regmap_bus_reg_read(map, reg, val);
> +	}

No, this makes no sense - if the device doesn't have a read operation
then it doesn't support a raw data stream and hence can't support raw
access sensibly.  Callers that want to access a lot of registers at once
without knowing what the wire format for the device is should be using
the bulk or multi interfaces.
Markus Pargmann Aug. 12, 2015, 12:34 p.m. UTC | #2
On Wed, Aug 12, 2015 at 12:27:07PM +0100, Mark Brown wrote:
> On Wed, Aug 12, 2015 at 12:12:36PM +0200, Markus Pargmann wrote:
> 
> > +	/*
> > +	 * There are busses that do not have a read function as it is optional.
> > +	 * Use their reg_read function instead if the requested number of bytes
> > +	 * is correct.
> > +	 */
> > +	if (!map->bus->read) {
> > +		/*
> > +		 * bus_reg_read() does not support reading values that are not
> > +		 * exactly the size of format.val_bytes
> > +		 */
> > +		if (val_len != map->format.val_bytes)
> > +			return -EINVAL;
> > +		return _regmap_bus_reg_read(map, reg, val);
> > +	}
> 
> No, this makes no sense - if the device doesn't have a read operation
> then it doesn't support a raw data stream and hence can't support raw
> access sensibly.  Callers that want to access a lot of registers at once
> without knowing what the wire format for the device is should be using
> the bulk or multi interfaces.

Yes okay. Then I will reduce this patch to the following and put it
into regmap_read() instead?
	if (!map->bus->read) {
		return -EINVAL;

Is -EINVAL the right thing to return or would you prefer -ENOTSUPP?

Thanks,

Markus
Mark Brown Aug. 14, 2015, 4:34 p.m. UTC | #3
On Wed, Aug 12, 2015 at 02:34:12PM +0200, Markus Pargmann wrote:

> Is -EINVAL the right thing to return or would you prefer -ENOTSUPP?

-ENOTSUPP is going to be better.
diff mbox

Patch

diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index 87f15fb60bc5..3b663350c573 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -2071,6 +2071,21 @@  static int _regmap_raw_read(struct regmap *map, unsigned int reg, void *val,
 
 	WARN_ON(!map->bus);
 
+	/*
+	 * There are busses that do not have a read function as it is optional.
+	 * Use their reg_read function instead if the requested number of bytes
+	 * is correct.
+	 */
+	if (!map->bus->read) {
+		/*
+		 * bus_reg_read() does not support reading values that are not
+		 * exactly the size of format.val_bytes
+		 */
+		if (val_len != map->format.val_bytes)
+			return -EINVAL;
+		return _regmap_bus_reg_read(map, reg, val);
+	}
+
 	range = _regmap_range_lookup(map, reg);
 	if (range) {
 		ret = _regmap_select_page(map, &reg, range,