On Mon, 22 Nov 2021 at 12:00, Sylwester Nawrocki [off-list ref] wrote:

On 22.11.2021 00:27, Sam Protsenko wrote:

quoted

/* Register Offset definitions for CMU_CORE (0x12000000) */

@@ -1014,24 +1060,12 @@ static int __init exynos850_cmu_probe(struct platform_device *pdev)
 {
      const struct samsung_cmu_info *info;
      struct device *dev = &pdev->dev;
-     struct device_node *np = dev->of_node;

      info = of_device_get_match_data(dev);
-     exynos850_init_clocks(np, info->clk_regs, info->nr_clk_regs);
-     samsung_cmu_register_one(np, info);

-     /* Keep bus clock running, so it's possible to access CMU registers */
-     if (info->clk_name) {
-             struct clk *bus_clk;
-
-             bus_clk = clk_get(dev, info->clk_name);
-             if (IS_ERR(bus_clk)) {
-                     pr_err("%s: could not find bus clock %s; err = %ld\n",
-                            __func__, info->clk_name, PTR_ERR(bus_clk));
-             } else {
-                     clk_prepare_enable(bus_clk);
-             }
-     }

quoted

+     /* Early clocks are already registered using CLK_OF_DECLARE_DRIVER() */
+     if (info != &peri_cmu_info)
+             exynos850_register_cmu(dev, dev->of_node, info);

Do you still need this test? How about just removing "samsung,exynos850-cmu-peri"
from exynos850_cmu_of_match[] ?

It'll be needed later, when I get to adding PM ops to platform driver,
as CMU_PERI is part of Power Domains. But you are right, it's not
needed right now. Will re-send this patch separately soon.

--
Regards,
Sylwester