在 2023/2/11 上午7:42, Stephen Boyd 写道:
Quoting Yinbo Zhu (2022-11-28 19:41:55)diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 903096bd87f8..4f8f1b8f796d 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -127,6 +127,7 @@ config LOONGARCH select USE_PERCPU_NUMA_NODE_ID select USER_STACKTRACE_SUPPORT select ZONE_DMA32 + select COMMON_CLKAny reason to put this at the end instead of alphabetically?
okay, I will change it as alphabetically.
config 32BITbool diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig index d79905f3e174..d13626f63739 100644 --- a/drivers/clk/Kconfig +++ b/drivers/clk/Kconfig @@ -326,6 +326,15 @@ config COMMON_CLK_LOCHNAGAR This driver supports the clocking features of the Cirrus Logic Lochnagar audio development board.+config COMMON_CLK_LOONGSON2+ bool "Clock driver for Loongson-2 SoC" + depends on COMMON_CLK && OF + help + This driver provides support for Clock Controller that base onDon't capitalize 'Clock Controller'
okay, I got it.
+ Common Clock Framework Controller (CCF) on Loongson-2 SoC. TheCCF doesn't really matter. Just say "support for clock controller on Loongson-2 SoC"
okay, I got it.
+ Clock Controller can generates and supplies clock to variousSame capitalization comment.
okay, I got it.
+ peripherals within the SoC. + config COMMON_CLK_NXP def_bool COMMON_CLK && (ARCH_LPC18XX || ARCH_LPC32XX) select REGMAP_MMIO if ARCH_LPC32XX diff --git a/drivers/clk/clk-loongson2.c b/drivers/clk/clk-loongson2.c new file mode 100644 index 000000000000..7487effceeab --- /dev/null +++ b/drivers/clk/clk-loongson2.c @@ -0,0 +1,286 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * Author: Yinbo Zhu <zhuyinbo@xxxxxxxxxxx> + * Copyright (C) 2022-2023 Loongson Technology Corporation Limited + */ + +#include <linux/clkdev.h>Drop include that isn't used.
okay, I got it.
+#include <linux/err.h> +#include <linux/init.h> +#include <linux/of.h> +#include <linux/of_address.h> +#include <dt-bindings/clock/loongson,ls2k-clk.h> +#include <linux/clk-provider.h> +#include <linux/slab.h> +#include <linux/clk.h> + +#define LOONGSON2_PLL_MULT_SHIFT 32 +#define LOONGSON2_PLL_MULT_WIDTH 10 +#define LOONGSON2_PLL_DIV_SHIFT 26 +#define LOONGSON2_PLL_DIV_WIDTH 6 +#define LOONGSON2_APB_FREQSCALE_SHIFT 20 +#define LOONGSON2_APB_FREQSCALE_WIDTH 3 +#define LOONGSON2_USB_FREQSCALE_SHIFT 16 +#define LOONGSON2_USB_FREQSCALE_WIDTH 3 +#define LOONGSON2_SATA_FREQSCALE_SHIFT 12 +#define LOONGSON2_SATA_FREQSCALE_WIDTH 3 + +static void __iomem *loongson2_pll_base; + +static struct clk_hw *loongson2_clk_register(struct device *dev, + const char *name, + const char *parent_name, + const struct clk_ops *ops, + unsigned long flags) +{ + int ret; + struct clk_hw *hw; + struct clk_init_data init; + + /* allocate the divider */ + hw = kzalloc(sizeof(*hw), GFP_KERNEL); + if (!hw) + return ERR_PTR(-ENOMEM); + + init.name = name; + init.ops = ops; + init.flags = flags; + init.parent_names = (parent_name ? &parent_name : NULL); + init.num_parents = (parent_name ? 1 : 0); + hw->init = &init; + + /* register the clock */ + ret = clk_hw_register(dev, hw); + if (ret) { + kfree(hw); + hw = ERR_PTR(ret); + } + + return hw; +} + +static struct clk_hw *loongson2_clk_pll_register(const char *name, + const char *parent, void __iomem *reg) +{ + u64 val; + u32 mult = 1, div = 1; + + val = readq((void *)reg);Drop the cast.
I will remove the (void *).
+ + mult = (val >> LOONGSON2_PLL_MULT_SHIFT) & + clk_div_mask(LOONGSON2_PLL_MULT_WIDTH); + div = (val >> LOONGSON2_PLL_DIV_SHIFT) & + clk_div_mask(LOONGSON2_PLL_DIV_WIDTH); + + return clk_hw_register_fixed_factor(NULL, name, parent, + CLK_SET_RATE_PARENT, mult, div); +} + +static unsigned long loongson2_apb_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + u64 val; + u32 mult; + unsigned long rate; + + val = readq((void *)(loongson2_pll_base + 0x50));Drop the cast.
I will remove the (void *).
+ + mult = (val >> LOONGSON2_APB_FREQSCALE_SHIFT) & + clk_div_mask(LOONGSON2_APB_FREQSCALE_WIDTH); + + rate = parent_rate * (mult + 1); + do_div(rate, 8); + + return rate; +} + +static const struct clk_ops loongson2_apb_clk_ops = { + .recalc_rate = loongson2_apb_recalc_rate, +}; + +static unsigned long loongson2_usb_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + u64 val; + u32 mult; + unsigned long rate; + + val = readq((void *)(loongson2_pll_base + 0x50));Drop the cast.
I will remove the (void *).
+ + mult = (val >> LOONGSON2_USB_FREQSCALE_SHIFT) & + clk_div_mask(LOONGSON2_USB_FREQSCALE_WIDTH); + + rate = parent_rate * (mult + 1); + do_div(rate, 8);Why is do_div() being used?
no expecial reason, I only want to get a result that rate divide 8.
+ + return rate; +} + +static const struct clk_ops loongson2_usb_clk_ops = { + .recalc_rate = loongson2_usb_recalc_rate, +}; + +static unsigned long loongson2_sata_recalc_rate(struct clk_hw *hw, + unsigned long parent_rate) +{ + u64 val; + u32 mult; + unsigned long rate; + + val = readq((void *)(loongson2_pll_base + 0x50));Drop the cast.
I will remove the (void *).
+ + mult = (val >> LOONGSON2_SATA_FREQSCALE_SHIFT) & + clk_div_mask(LOONGSON2_SATA_FREQSCALE_WIDTH); + + rate = parent_rate * (mult + 1); + do_div(rate, 8); + + return rate; +} + +static const struct clk_ops loongson2_sata_clk_ops = { + .recalc_rate = loongson2_sata_recalc_rate, +}; + +static void loongson2_check_clk_hws(struct clk_hw *clks[], unsigned int count)Inline this function.
I will add it.
+{ + unsigned int i; + + for (i = 0; i < count; i++) + if (IS_ERR(clks[i])) + pr_err("Loongson2 clk %u: register failed with %ld\n" + , i, PTR_ERR(clks[i]));Put the comma on the line above please.
okay, I got it.
+} + +static struct clk_hw *loongson2_obtain_fixed_clk_hw( + struct device_node *np, + const char *name) +{ + struct clk *clk; + + clk = of_clk_get_by_name(np, name); + if (IS_ERR(clk)) + return ERR_PTR(-ENOENT); + + return __clk_get_hw(clk);Please don't implement this whole function. Instead, use clk_parent_data and the 'fw_name' or 'index' members to assign clk parents.
at present, the child clk doesn't depend on the clk_parent_data, eg.hws[LOONGSON2_NODE_PLL] = loongson2_clk_pll_register("node_pll", "ref_100m", loongson2_pll_base);
static struct clk_hw *loongson2_clk_pll_register(const char *name,
const char *parent, void __iomem *reg)
{
u64 val;
u32 mult = 1, div = 1;
val = readq((void *)reg);
mult = (val >> LOONGSON2_PLL_MULT_SHIFT) &
clk_div_mask(LOONGSON2_PLL_MULT_WIDTH);
div = (val >> LOONGSON2_PLL_DIV_SHIFT) &
clk_div_mask(LOONGSON2_PLL_DIV_WIDTH);
return clk_hw_register_fixed_factor(NULL, name, parent,
CLK_SET_RATE_PARENT, mult, div);
}
you meaning is to use clk_parent_data to rework
loongson2_clk_pll_register and drop
loongson2_obtain_fixed_clk_hw ?
sorry, I don't get it. you meaning is that remove "goto err". Instead, add a "return".+} + +static void __init loongson2_clocks_init(struct device_node *np) +{ + struct clk_hw **hws; + struct clk_hw_onecell_data *clk_hw_data; + spinlock_t loongson2_clk_lock; + + loongson2_pll_base = of_iomap(np, 0); + + if (!loongson2_pll_base) { + pr_err("clk: unable to map loongson2 clk registers\n"); + goto err;return?
+ } + + clk_hw_data = kzalloc(struct_size(clk_hw_data, hws, LOONGSON2_CLK_END), + GFP_KERNEL); + if (WARN_ON(!clk_hw_data)) + goto err; + + clk_hw_data->num = LOONGSON2_CLK_END; + hws = clk_hw_data->hws; + + hws[LOONGSON2_REF_100M] = loongson2_obtain_fixed_clk_hw(np, + "ref_100m"); + + hws[LOONGSON2_NODE_PLL] = loongson2_clk_pll_register("node_pll", + "ref_100m", + loongson2_pll_base); + + hws[LOONGSON2_DDR_PLL] = loongson2_clk_pll_register("ddr_pll", + "ref_100m", + loongson2_pll_base + 0x10); + + hws[LOONGSON2_DC_PLL] = loongson2_clk_pll_register("dc_pll", + "ref_100m", + loongson2_pll_base + 0x20); + + hws[LOONGSON2_PIX0_PLL] = loongson2_clk_pll_register("pix0_pll", + "ref_100m", + loongson2_pll_base + 0x30); + + hws[LOONGSON2_PIX1_PLL] = loongson2_clk_pll_register("pix1_pll", + "ref_100m", + loongson2_pll_base + 0x40); + + hws[LOONGSON2_NODE_CLK] = clk_hw_register_divider(NULL, "node", + "node_pll", 0, + loongson2_pll_base + 0x8, 0, + 6, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + /* + * The hda clk divisor in the upper 32bits and the clk-prodiver + * layer code doesn't support 64bit io operation thus a conversion + * is required that subtract shift by 32 and add 4byte to the hda + * address + */ + hws[LOONGSON2_HDA_CLK] = clk_hw_register_divider(NULL, "hda", + "ddr_pll", 0, + loongson2_pll_base + 0x22, 12, + 7, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_GPU_CLK] = clk_hw_register_divider(NULL, "gpu", + "ddr_pll", 0, + loongson2_pll_base + 0x18, 22, + 6, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_DDR_CLK] = clk_hw_register_divider(NULL, "ddr", + "ddr_pll", 0, + loongson2_pll_base + 0x18, 0, + 6, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_GMAC_CLK] = clk_hw_register_divider(NULL, "gmac", + "dc_pll", 0, + loongson2_pll_base + 0x28, 22, + 6, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_DC_CLK] = clk_hw_register_divider(NULL, "dc", + "dc_pll", 0, + loongson2_pll_base + 0x28, 0, + 6, CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_APB_CLK] = loongson2_clk_register(NULL, "apb", + "gmac", + &loongson2_apb_clk_ops, 0); + + hws[LOONGSON2_USB_CLK] = loongson2_clk_register(NULL, "usb", + "gmac", + &loongson2_usb_clk_ops, 0); + + hws[LOONGSON2_SATA_CLK] = loongson2_clk_register(NULL, "sata", + "gmac", + &loongson2_sata_clk_ops, 0); + + hws[LOONGSON2_PIX0_CLK] = clk_hw_register_divider(NULL, "pix0", + "pix0_pll", 0, + loongson2_pll_base + 0x38, 0, 6, + CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + hws[LOONGSON2_PIX1_CLK] = clk_hw_register_divider(NULL, "pix1", + "pix1_pll", 0, + loongson2_pll_base + 0x48, 0, 6, + CLK_DIVIDER_ONE_BASED, + &loongson2_clk_lock); + + loongson2_check_clk_hws(hws, LOONGSON2_CLK_END); + + of_clk_add_hw_provider(np, of_clk_hw_onecell_get, clk_hw_data); + +err: + iounmap(loongson2_pll_base); +} + +CLK_OF_DECLARE(loongson2_clk, "loongson,ls2k-clk", loongson2_clocks_init);Any reason this can't be a platform driver?
For the compatible consideration of other clock controllers of Loongson-2 series in the future, the way of using dts can be
better compatible.
