Hi Geert,
On 05.04.2017 15:15, Geert Uytterhoeven wrote:
Hi Alexey,
On Wed, Apr 5, 2017 at 1:46 PM, Alexey Firago <alexey_firago@xxxxxxxxxx> wrote:
Introduce vc5_chip_info structure to describe features of a particular
VC5 chip (id, number of FODs, number of outputs, flags).
For now flags are only used to indicate if chip has internal XTAL.
vc5_chip_info is set on probe from the matched of_device_id->data.
Also add defines to specify maximum number of FODs and clock outputs
supported by the driver.
With these changes it should be easier to extend driver to support
more VC5 models.
Signed-off-by: Alexey Firago <alexey_firago@xxxxxxxxxx>
Thanks for your patch!
Reviewed-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
--- a/drivers/clk/clk-versaclock5.c
+++ b/drivers/clk/clk-versaclock5.c
@@ -113,12 +113,30 @@
#define VC5_MUX_IN_XIN BIT(0)
#define VC5_MUX_IN_CLKIN BIT(1)
+/* Maximum number of clk_out supported by this driver */
+#define VC5_MAX_CLK_OUT_NUM 3
+
+/* Maximum number of FODs supported by this driver */
+#define VC5_MAX_FOD_NUM 2
+
+/* flags to describe chip features */
+/* chip has built-in oscilator */
+#define VC5_HAS_INTERNAL_XTAL BIT(0)
VC5_HAS_INTERNAL_OSC?
I'm fine with renaming it, but shouldn't it be consistent with the rest
of the driver (see "internal-xtal", VC5_XTAL*, etc) and IDT datasheet?
+
/* Supported IDT VC5 models. */
enum vc5_model {
IDT_VC5_5P49V5923,
IDT_VC5_5P49V5933,
};
+/* Structure to describe features of a particular VC5 model */
+struct vc5_chip_info {
+ const enum vc5_model model;
+ const int clk_fod_cnt;
+ const int clk_out_cnt;
const unsigned int (both)
Will fix.
+ u32 flags;
+};
+
struct vc5_driver_data;
struct vc5_hw_data {
@@ -132,15 +150,15 @@ struct vc5_hw_data {
struct vc5_driver_data {
struct i2c_client *client;
struct regmap *regmap;
- enum vc5_model model;
+ struct vc5_chip_info *chip_info;
const struct vc5_chip_info *chip_info;
@@ -591,7 +609,7 @@ static int vc5_probe(struct i2c_client *client,
struct vc5_driver_data *vc5;
struct clk_init_data init;
const char *parent_names[2];
- unsigned int n, idx;
+ unsigned int n, idx = 0;
int ret;
vc5 = devm_kzalloc(&client->dev, sizeof(*vc5), GFP_KERNEL);
@@ -600,7 +618,12 @@ static int vc5_probe(struct i2c_client *client,
i2c_set_clientdata(client, vc5);
vc5->client = client;
- vc5->model = (enum vc5_model)of_id->data;
+
+ vc5->chip_info = (struct vc5_chip_info *)of_id->data;
I think the cast is no longer needed when chip_info becomes const.
BTW, of_id is not really needed if you write it like:
vc5->chip_info = of_device_get_match_data(&client->dev);
+ if (!vc5->chip_info) {
This can't really happen, can it?
Ok. Will rework.
+ dev_err(&client->dev, "No device match found\n");
+ return -ENODEV;
+ }
vc5->pin_xin = devm_clk_get(&client->dev, "xin");
if (PTR_ERR(vc5->pin_xin) == -EPROBE_DEFER)
@@ -622,8 +645,8 @@ static int vc5_probe(struct i2c_client *client,
if (!IS_ERR(vc5->pin_xin)) {
vc5->clk_mux_ins |= VC5_MUX_IN_XIN;
parent_names[init.num_parents++] = __clk_get_name(vc5->pin_xin);
- } else if (vc5->model == IDT_VC5_5P49V5933) {
- /* IDT VC5 5P49V5933 has built-in oscilator. */
+ } else if (vc5->chip_info->flags & VC5_HAS_INTERNAL_XTAL) {
+ /* chip has built-in oscilator. */
The comment is no longer needed when the bit is named VC5_HAS_INTERNAL_OSC :-)
Ok. Will kill it.
Thanks for your review!
Regards,
Alexey
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
