[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d58e289b54f66c239ae09e94728716b7.sboyd@kernel.org>
Date: Wed, 09 Aug 2023 14:26:40 -0700
From: Stephen Boyd <sboyd@...nel.org>
To: niravkumar.l.rabara@...el.com
Cc: adrian.ho.yin.ng@...el.com, andrew@...n.ch, conor+dt@...nel.org, devicetree@...r.kernel.org, dinguyen@...nel.org, krzysztof.kozlowski+dt@...aro.org, linux-clk@...r.kernel.org, linux-kernel@...r.kernel.org, mturquette@...libre.com, netdev@...r.kernel.org, p.zabel@...gutronix.de, richardcochran@...il.com, robh+dt@...nel.org, wen.ping.teh@...el.com
Subject: Re: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoting niravkumar.l.rabara@...el.com (2023-07-31 18:02:33)
> diff --git a/drivers/clk/socfpga/clk-agilex.c b/drivers/clk/socfpga/clk-agilex.c
> index 74d21bd82710..3dcd0f233c17 100644
> --- a/drivers/clk/socfpga/clk-agilex.c
> +++ b/drivers/clk/socfpga/clk-agilex.c
> @@ -1,6 +1,6 @@
> // SPDX-License-Identifier: GPL-2.0
> /*
> - * Copyright (C) 2019, Intel Corporation
> + * Copyright (C) 2019-2023, Intel Corporation
> */
> #include <linux/slab.h>
> #include <linux/clk-provider.h>
> @@ -9,6 +9,7 @@
> #include <linux/platform_device.h>
>
> #include <dt-bindings/clock/agilex-clock.h>
> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h>
>
> #include "stratix10-clk.h"
>
> @@ -41,6 +42,67 @@ static const struct clk_parent_data mpu_free_mux[] = {
> .name = "f2s-free-clk", },
> };
>
> +static const struct clk_parent_data core0_free_mux[] = {
> + { .fw_name = "main_pll_c1",
> + .name = "main_pll_c1", },
We're adding support for something new? Only set .fw_name in that case,
as .name will never be used. To do even better, set only .index so that
we don't do any string comparisons.
> + { .fw_name = "peri_pll_c0",
> + .name = "peri_pll_c0", },
> + { .fw_name = "osc1",
> + .name = "osc1", },
> + { .fw_name = "cb-intosc-hs-div2-clk",
> + .name = "cb-intosc-hs-div2-clk", },
> + { .fw_name = "f2s-free-clk",
> + .name = "f2s-free-clk", },
> +};
> +
[...]
> +
> static int n5x_clk_register_c_perip(const struct n5x_perip_c_clock *clks,
> int nums, struct stratix10_clock_data *data)
> {
> @@ -535,6 +917,51 @@ static int n5x_clkmgr_init(struct platform_device *pdev)
> return 0;
> }
>
> +static int agilex5_clkmgr_init(struct platform_device *pdev)
> +{
> + struct device_node *np = pdev->dev.of_node;
> + struct device *dev = &pdev->dev;
> + struct stratix10_clock_data *clk_data;
Maybe call this stratix_data so that clk_data.clk_data is
stratix_data.clk_data.
> + struct resource *res;
> + void __iomem *base;
> + int i, num_clks;
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + base = devm_ioremap_resource(dev, res);
This is a single function call, devm_platform_ioremap_resource().
> + if (IS_ERR(base))
> + return PTR_ERR(base);
> +
> + num_clks = AGILEX5_NUM_CLKS;
> +
> + clk_data = devm_kzalloc(dev, struct_size(clk_data, clk_data.hws,
> + num_clks), GFP_KERNEL);
> + if (!clk_data)
> + return -ENOMEM;
> +
> + for (i = 0; i < num_clks; i++)
> + clk_data->clk_data.hws[i] = ERR_PTR(-ENOENT);
> +
> + clk_data->base = base;
> + clk_data->clk_data.num = num_clks;
> +
> + agilex_clk_register_pll(agilex_pll_clks, ARRAY_SIZE(agilex_pll_clks),
> + clk_data);
> +
> + agilex_clk_register_c_perip(agilex5_main_perip_c_clks,
> + ARRAY_SIZE(agilex5_main_perip_c_clks),
> + clk_data);
> +
> + agilex_clk_register_cnt_perip(agilex5_main_perip_cnt_clks,
> + ARRAY_SIZE(agilex5_main_perip_cnt_clks),
> + clk_data);
> +
> + agilex_clk_register_gate(agilex5_gate_clks,
> + ARRAY_SIZE(agilex5_gate_clks), clk_data);
> +
> + of_clk_add_hw_provider(np, of_clk_hw_onecell_get, &clk_data->clk_data);
devm? Or when is this provider removed?
> + return 0;
> +}
> +
> static int agilex_clkmgr_probe(struct platform_device *pdev)
> {
> int (*probe_func)(struct platform_device *init_func);
Powered by blists - more mailing lists