On 2/16/2024 9:49 PM, Andrew Lunn wrote:
On Fri, Feb 16, 2024 at 07:59:19AM +0800, Yang Xiwen wrote:
On 2/16/2024 7:57 AM, Andrew Lunn wrote:
+ for (i = 0; i < CLK_NUM; i++) {
+ priv->clks[i] = devm_clk_get_enabled(&pdev->dev, clk_strs[i]);
+ if (IS_ERR(priv->clks[i])) {
+ dev_err(dev, "failed to get enabled clk %s: %ld\n", clk_strs[i],
+ PTR_ERR(priv->clks[i]));
+ ret = -ENODEV;
+ goto out_free_netdev;
+ }
The clk API has devm_clk_bulk_ versions. Please take a look at them, and see
if it will simplify the code.
I know this API, but it can't be used. We need to control clocks
individually in reset procedure.
/**
* struct clk_bulk_data - Data used for bulk clk operations.
*
* @id: clock consumer ID
* @clk: struct clk * to store the associated clock
*
* The CLK APIs provide a series of clk_bulk_() API calls as
* a convenience to consumers which require multiple clks. This
* structure is used to manage data for these calls.
*/
struct clk_bulk_data {
const char *id;
struct clk *clk;
};
You pass the bulk API calls an array of this structure. After the get
has completed, you can access the individual clocks via the clk
pointer. So you can use the bulk API on probe and remove when you need
to operate on all three, and the single clk API for your reset handler
etc.
seems okay. I'll implement this in next version.
Andrew
--
Regards,
Yang Xiwen
