When doing a bulk read from a device which lacks raw I/O support we fall
back to doing register at a time reads but we still use the raw
formatters in order to render the data into the word size used by the
device (since bulk reads still operate on the device word size rather
than unsigned ints). This means that devices without raw formatting
such as those that provide reg_read() are not supported. Provide
handling for them by copying the values read into native endian values
of the appropriate size.
Signed-off-by: Mark Brown <[email protected]>
---
This is tested with my "hey, look - it compiles!" test plan. Based on
looking at the users and the other ways formatting is used this is the
interface that makes most sense to me, we'll need similar handling for
the write case. Does this solve the problems you are seeing?
drivers/base/regmap/regmap.c | 29 ++++++++++++++++++++++++++++-
1 file changed, 28 insertions(+), 1 deletion(-)
diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index c13b1f2..62e3d7c 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -2364,7 +2364,34 @@ int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val,
&ival);
if (ret != 0)
return ret;
- map->format.format_val(val + (i * val_bytes), ival, 0);
+
+ if (map->format.format_val) {
+ map->format.format_val(val + (i * val_bytes), ival, 0);
+ } else {
+ /* Devices providing read and write
+ * operations can use the bulk I/O
+ * functions if they define a val_bytes,
+ * we assume that the values are native
+ * endian.
+ */
+ u32 *u32 = val;
+ u16 *u16 = val;
+ u8 *u8 = val;
+
+ switch (map->format.val_bytes) {
+ case 4:
+ u32[i] = ival;
+ break;
+ case 2:
+ u16[i] = ival;
+ break;
+ case 1:
+ u8[i] = ival;
+ break;
+ default:
+ return -EINVAL;
+ }
+ }
}
}
--
2.5.0
Hi Mark,
Yes, this patch solved the issue.It avoid the null function and just
need to handle the native endian case.
Thanks.
Henry
On Fri, 2015-08-28 at 20:11 +0100, Mark Brown wrote:
> When doing a bulk read from a device which lacks raw I/O support we fall
> back to doing register at a time reads but we still use the raw
> formatters in order to render the data into the word size used by the
> device (since bulk reads still operate on the device word size rather
> than unsigned ints). This means that devices without raw formatting
> such as those that provide reg_read() are not supported. Provide
> handling for them by copying the values read into native endian values
> of the appropriate size.
>
> Signed-off-by: Mark Brown <[email protected]>
> ---
>
> This is tested with my "hey, look - it compiles!" test plan. Based on
> looking at the users and the other ways formatting is used this is the
> interface that makes most sense to me, we'll need similar handling for
> the write case. Does this solve the problems you are seeing?
>
> drivers/base/regmap/regmap.c | 29 ++++++++++++++++++++++++++++-
> 1 file changed, 28 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
> index c13b1f2..62e3d7c 100644
> --- a/drivers/base/regmap/regmap.c
> +++ b/drivers/base/regmap/regmap.c
> @@ -2364,7 +2364,34 @@ int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val,
> &ival);
> if (ret != 0)
> return ret;
> - map->format.format_val(val + (i * val_bytes), ival, 0);
> +
> + if (map->format.format_val) {
> + map->format.format_val(val + (i * val_bytes), ival, 0);
> + } else {
> + /* Devices providing read and write
> + * operations can use the bulk I/O
> + * functions if they define a val_bytes,
> + * we assume that the values are native
> + * endian.
> + */
> + u32 *u32 = val;
> + u16 *u16 = val;
> + u8 *u8 = val;
> +
> + switch (map->format.val_bytes) {
> + case 4:
> + u32[i] = ival;
> + break;
> + case 2:
> + u16[i] = ival;
> + break;
> + case 1:
> + u8[i] = ival;
> + break;
> + default:
> + return -EINVAL;
> + }
> + }
> }
> }
>