Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752275AbbBMHmR (ORCPT ); Fri, 13 Feb 2015 02:42:17 -0500 Received: from mail-qg0-f51.google.com ([209.85.192.51]:59107 "EHLO mail-qg0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751536AbbBMHmQ (ORCPT ); Fri, 13 Feb 2015 02:42:16 -0500 MIME-Version: 1.0 In-Reply-To: <1423478845-2835-3-git-send-email-s.hauer@pengutronix.de> References: <1423478845-2835-1-git-send-email-s.hauer@pengutronix.de> <1423478845-2835-3-git-send-email-s.hauer@pengutronix.de> From: Tomasz Figa Date: Fri, 13 Feb 2015 16:41:51 +0900 Message-ID: Subject: Re: [PATCH 02/13] clk: mediatek: Add initial common clock support for Mediatek SoCs. To: Sascha Hauer Cc: Matthias Brugger , James Liao , Mike Turquette , =?UTF-8?B?WUggQ2hlbiAo6Zmz5pix6LGqKQ==?= , "linux-kernel@vger.kernel.org" , Henry Chen , Rob Herring , kernel@pengutronix.de, =?UTF-8?B?WWluZ2pvZSBDaGVuICjpmbPoi7HmtLIp?= , Eddie Huang , Lee Jones , "linux-arm-kernel@lists.infradead.org" Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 20301 Lines: 640 Hi, Let me add some suggestions inline. On Mon, Feb 9, 2015 at 7:47 PM, Sascha Hauer wrote: > From: James Liao > > This patch adds common clock support for Mediatek SoCs, including plls, > muxes and clock gates. [snip] > +static int mtk_cg_enable(struct clk_hw *hw) > +{ > + mtk_cg_clr_bit(hw); > + > + return 0; > +} > + > +static void mtk_cg_disable(struct clk_hw *hw) > +{ > + mtk_cg_set_bit(hw); > +} > + > +static int mtk_cg_enable_inv(struct clk_hw *hw) > +{ > + mtk_cg_set_bit(hw); > + > + return 0; > +} > + > +static void mtk_cg_disable_inv(struct clk_hw *hw) > +{ > + mtk_cg_clr_bit(hw); > +} Instead of duplicating the ops, couldn't you add a flag or something to mtk_clk_gate struct and then act appropriately in the ops? Also, see below. > + > +const struct clk_ops mtk_clk_gate_ops_setclr = { > + .is_enabled = mtk_cg_bit_is_cleared, > + .enable = mtk_cg_enable, > + .disable = mtk_cg_disable, > +}; > + > +const struct clk_ops mtk_clk_gate_ops_setclr_inv = { > + .is_enabled = mtk_cg_bit_is_set, > + .enable = mtk_cg_enable_inv, > + .disable = mtk_cg_disable_inv, > +}; > + > +struct clk *mtk_clk_register_gate( > + const char *name, > + const char *parent_name, > + struct regmap *regmap, > + int set_ofs, > + int clr_ofs, > + int sta_ofs, > + u8 bit, > + const struct clk_ops *ops) Instead of passing the ops here you would have some flags or even just a single bool inverted. Then the ops struct could be made static. Also it would be nice to have a kerneldoc-style comment documenting arguments of this function. Same thing applies to other structs added by this and related patches and non-static functions. also CodingStyle: I believe it is not kernel coding style to push every argument to new line, even if few of them can fit one line. Similar thing applies to other functions added by this and related patches using this convention. > +{ > + struct mtk_clk_gate *cg; > + struct clk *clk; > + struct clk_init_data init; > + > + cg = kzalloc(sizeof(*cg), GFP_KERNEL); > + if (!cg) > + return ERR_PTR(-ENOMEM); > + > + init.name = name; > + init.flags = CLK_SET_RATE_PARENT; > + init.parent_names = parent_name ? &parent_name : NULL; > + init.num_parents = parent_name ? 1 : 0; > + init.ops = ops; > + > + cg->regmap = regmap; > + cg->set_ofs = set_ofs; > + cg->clr_ofs = clr_ofs; > + cg->sta_ofs = sta_ofs; > + cg->bit = bit; > + > + cg->hw.init = &init; > + > + clk = clk_register(NULL, &cg->hw); > + if (IS_ERR(clk)) > + kfree(cg); > + > + return clk; > +} > diff --git a/drivers/clk/mediatek/clk-gate.h b/drivers/clk/mediatek/clk-gate.h > new file mode 100644 > index 0000000..a44dcbf > --- /dev/null > +++ b/drivers/clk/mediatek/clk-gate.h > @@ -0,0 +1,49 @@ > +/* > + * Copyright (c) 2014 MediaTek Inc. > + * Author: James Liao Might not be necessary, but maybe the other people (all or some of them) from signed-off-by should be added to this and other copyright statements? > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + */ > + > +#ifndef __DRV_CLK_GATE_H > +#define __DRV_CLK_GATE_H > + > +/* > + * This is a private header file. DO NOT include it except clk-*.c. > + */ I believe the above comment is unnecessary, because the file is already located in drivers/clk/mediatek. > +#include > +#include > +#include > + > +struct mtk_clk_gate { It would be nice to have a kerneldoc-style comment describing fields of this struct. > + struct clk_hw hw; > + struct regmap *regmap; > + int set_ofs; > + int clr_ofs; > + int sta_ofs; > + u8 bit; > +}; > + > +#define to_clk_gate(_hw) container_of(_hw, struct mtk_clk_gate, hw) I believe static inline is preferred to macros for such helpers, due to increased type safety. > + > +extern const struct clk_ops mtk_clk_gate_ops_setclr; > +extern const struct clk_ops mtk_clk_gate_ops_setclr_inv; > + > +struct clk *mtk_clk_register_gate( > + const char *name, > + const char *parent_name, > + struct regmap *regmap, > + int set_ofs, > + int clr_ofs, > + int sta_ofs, > + u8 bit, > + const struct clk_ops *ops); > + > +#endif /* __DRV_CLK_GATE_H */ > diff --git a/drivers/clk/mediatek/clk-mtk.c b/drivers/clk/mediatek/clk-mtk.c > new file mode 100644 > index 0000000..479857c > --- /dev/null > +++ b/drivers/clk/mediatek/clk-mtk.c > @@ -0,0 +1,155 @@ > +/* > + * Copyright (c) 2014 MediaTek Inc. > + * Author: James Liao > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + */ > + > +#include > +#include > + > +#include > +#include > +#include > +#include > + > +#include "clk-mtk.h" > +#include "clk-gate.h" > + > +void mtk_init_factors(struct mtk_fixed_factor *clks, int num, > + struct clk_onecell_data *clk_data) > +{ > + int i; > + struct clk *clk; > + > + for (i = 0; i < num; i++) { > + struct mtk_fixed_factor *ff = &clks[i]; > + > + clk = clk_register_fixed_factor(NULL, ff->name, ff->parent_name, > + CLK_SET_RATE_PARENT, ff->mult, ff->div); > + > + if (IS_ERR(clk)) { > + pr_err("Failed to register clk %s: %ld\n", > + ff->name, PTR_ERR(clk)); > + continue; > + } > + > + if (clk_data) > + clk_data->clks[ff->id] = clk; > + } > +} > + > +void mtk_init_clk_gates(struct regmap *regmap, > + struct mtk_gate *clks, int num, > + struct clk_onecell_data *clk_data) > +{ > + int i; > + struct clk *clk; > + > + for (i = 0; i < num; i++) { > + struct mtk_gate *gate = &clks[i]; > + > + clk = mtk_clk_register_gate(gate->name, gate->parent_name, > + regmap, > + gate->regs->set_ofs, > + gate->regs->clr_ofs, > + gate->regs->sta_ofs, > + gate->shift, gate->ops); > + > + if (IS_ERR(clk)) { > + pr_err("Failed to register clk %s: %ld\n", > + gate->name, PTR_ERR(clk)); > + continue; > + } > + > + if (clk_data) > + clk_data->clks[gate->id] = clk; > + } > +} > + > +struct clk_onecell_data *mtk_alloc_clk_data(unsigned int clk_num) > +{ > + int i; > + struct clk_onecell_data *clk_data; > + > + clk_data = kzalloc(sizeof(clk_data), GFP_KERNEL); Shouldn't it be sizeof(*clk_data)? > + if (!clk_data) > + return NULL; > + > + clk_data->clks = kcalloc(clk_num, sizeof(struct clk *), GFP_KERNEL); sizeof(*clk_data->clks) > + if (!clk_data->clks) { > + kfree(clk_data); > + return NULL; > + } > + > + clk_data->clk_num = clk_num; > + > + for (i = 0; i < clk_num; ++i) > + clk_data->clks[i] = ERR_PTR(-ENOENT); > + > + return clk_data; > +} > + > +struct clk *mtk_clk_register_mux( > + const char *name, > + const char **parent_names, > + u8 num_parents, > + void __iomem *base_addr, > + u8 shift, > + u8 width, > + u8 gate_bit, > + spinlock_t *lock) > +{ > + struct clk *clk; > + struct clk_mux *mux; > + struct clk_gate *gate = NULL; > + struct clk_hw *gate_hw = NULL; > + const struct clk_ops *gate_ops = NULL; > + u32 mask = BIT(width) - 1; > + > + mux = kzalloc(sizeof(struct clk_mux), GFP_KERNEL); sizeof(*mux) > + if (!mux) > + return ERR_PTR(-ENOMEM); > + > + mux->reg = base_addr; > + mux->mask = mask; > + mux->shift = shift; > + mux->flags = 0; Flags field was already zeroed by kzalloc(). > + mux->lock = lock; > + > + if (gate_bit <= MAX_MUX_GATE_BIT) { > + gate = kzalloc(sizeof(struct clk_gate), GFP_KERNEL); sizeof(*gate) > + if (!gate) { > + kfree(mux); Please use goto style error path below the main code instead of duplicating roll-back actions in every error check. > + return ERR_PTR(-ENOMEM); > + } > + > + gate->reg = base_addr; > + gate->bit_idx = gate_bit; > + gate->flags = CLK_GATE_SET_TO_DISABLE; > + gate->lock = lock; > + > + gate_hw = &gate->hw; > + gate_ops = &clk_gate_ops; > + } > + > + clk = clk_register_composite(NULL, name, parent_names, num_parents, > + &mux->hw, &clk_mux_ops, > + NULL, NULL, > + gate_hw, gate_ops, > + CLK_SET_RATE_PARENT); > + > + if (IS_ERR(clk)) { > + kfree(gate); > + kfree(mux); > + } > + > + return clk; > +} > diff --git a/drivers/clk/mediatek/clk-mtk.h b/drivers/clk/mediatek/clk-mtk.h > new file mode 100644 > index 0000000..35cf9a3 > --- /dev/null > +++ b/drivers/clk/mediatek/clk-mtk.h > @@ -0,0 +1,133 @@ > +/* > + * Copyright (c) 2014 MediaTek Inc. > + * Author: James Liao > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + */ > + > +#ifndef __DRV_CLK_MTK_H > +#define __DRV_CLK_MTK_H > + > +/* > + * This is a private header file. DO NOT include it except clk-*.c. > + */ > + > +#include > +#include > +#include > +#include > + > +#define MAX_MUX_GATE_BIT 31 > +#define INVALID_MUX_GATE_BIT (MAX_MUX_GATE_BIT + 1) It might be a good idea to describe what this value is used for. > + > +struct mtk_fixed_factor { > + int id; > + const char *name; > + const char *parent_name; > + int mult; > + int div; > +}; > + > +#define FACTOR(_id, _name, _parent, _mult, _div) { \ > + .id = _id, \ > + .name = _name, \ > + .parent_name = _parent, \ > + .mult = _mult, \ > + .div = _div, \ > + } > + > +extern void mtk_init_factors(struct mtk_fixed_factor *clks, int num, > + struct clk_onecell_data *clk_data); > + > +struct mtk_mux { > + int id; > + const char *name; > + uint32_t reg; > + int shift; > + int width; > + int gate; > + const char **parent_names; > + int num_parents; > +}; > + > +#define MUX(_id, _name, _parents, _reg, _shift, _width, _gate) { \ > + .id = _id, \ > + .name = _name, \ > + .reg = _reg, \ > + .shift = _shift, \ > + .width = _width, \ > + .gate = _gate, \ > + .parent_names = (const char **)_parents, \ Hmm, it doesn't sound like a good idea to hide casts like this inside a macro. Anyway, is this cast even necessary? Your _parents argument to this macro should be always of correct type. > + .num_parents = ARRAY_SIZE(_parents), \ > + } > + > +struct mtk_pll { > + int id; > + const char *name; > + const char *parent_name; > + uint32_t reg; > + uint32_t pwr_reg; > + uint32_t en_mask; > + unsigned int flags; > + const struct clk_ops *ops; > +}; > + > +#define PLL(_id, _name, _parent, _reg, _pwr_reg, _en_mask, _flags, _ops) { \ > + .id = _id, \ > + .name = _name, \ > + .parent_name = _parent, \ > + .reg = _reg, \ > + .pwr_reg = _pwr_reg, \ > + .en_mask = _en_mask, \ > + .flags = _flags, \ > + .ops = _ops, \ > + } > + > +struct mtk_gate_regs { > + u32 sta_ofs; > + u32 clr_ofs; > + u32 set_ofs; > +}; > + > +struct mtk_gate { > + int id; > + const char *name; > + const char *parent_name; > + struct mtk_gate_regs *regs; > + int shift; > + const struct clk_ops *ops; > +}; > + > +#define GATE(_id, _name, _parent, _regs, _shift, _ops) { \ > + .id = _id, \ > + .name = _name, \ > + .parent_name = _parent, \ > + .regs = &_regs, \ Also hiding operators like & inside macros like this doesn't help with code readability and it doesn't cost that much to just add & when calling this macro. > + .shift = _shift, \ > + .ops = _ops, \ > + } > + > +void mtk_init_clk_gates(struct regmap *regmap, > + struct mtk_gate *clks, int num, > + struct clk_onecell_data *clk_data); > + > +struct clk_onecell_data *mtk_alloc_clk_data(unsigned int clk_num); > + > +struct clk *mtk_clk_register_mux( > + const char *name, > + const char **parent_names, > + u8 num_parents, > + void __iomem *base_addr, > + u8 shift, > + u8 width, > + u8 gate_bit, > + spinlock_t *lock); > + > +#endif /* __DRV_CLK_MTK_H */ > diff --git a/drivers/clk/mediatek/clk-pll.c b/drivers/clk/mediatek/clk-pll.c > new file mode 100644 > index 0000000..59dee83 > --- /dev/null > +++ b/drivers/clk/mediatek/clk-pll.c > @@ -0,0 +1,63 @@ > +/* > + * Copyright (c) 2014 MediaTek Inc. > + * Author: James Liao > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + */ > + > +#include > +#include > +#include > + > +#include "clk-mtk.h" > +#include "clk-pll.h" > + > +struct clk *mtk_clk_register_pll( > + const char *name, > + const char *parent_name, > + u32 *base_addr, > + u32 *pwr_addr, > + u32 en_mask, > + u32 flags, > + const struct clk_ops *ops, > + spinlock_t *lock) > +{ > + struct mtk_clk_pll *pll; > + struct clk_init_data init; > + struct clk *clk; > + > + pr_debug("%s(): name: %s\n", __func__, name); > + > + if (!lock) > + return ERR_PTR(-EINVAL); Hmm, this check seems to be slightly inconsistent. Why you need to check lock, but not name, parent_name and other arguments of this function? Also bailing out without any error message isn't really the best practice. Anyway, if this function is defined to require the lock argument to be non-NULL then this is not a natural error condition but rather a mistake of the author of code calling this function with lock == NULL. IMHO either BUG_ON(!lock) or just remove this check, but I'd like to know Mike's thoughts on this before you make this change. > + > + pll = kzalloc(sizeof(*pll), GFP_KERNEL); > + if (!pll) > + return ERR_PTR(-ENOMEM); > + > + pll->base_addr = base_addr; > + pll->pwr_addr = pwr_addr; > + pll->en_mask = en_mask; > + pll->flags = flags; > + pll->lock = lock; > + pll->hw.init = &init; > + > + init.name = name; > + init.ops = ops; > + init.parent_names = &parent_name; > + init.num_parents = 1; > + > + clk = clk_register(NULL, &pll->hw); > + > + if (IS_ERR(clk)) > + kfree(pll); > + > + return clk; > +} > diff --git a/drivers/clk/mediatek/clk-pll.h b/drivers/clk/mediatek/clk-pll.h > new file mode 100644 > index 0000000..341d2fe > --- /dev/null > +++ b/drivers/clk/mediatek/clk-pll.h > @@ -0,0 +1,52 @@ > +/* > + * Copyright (c) 2014 MediaTek Inc. > + * Author: James Liao > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License version 2 as > + * published by the Free Software Foundation. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + */ > + > +#ifndef __DRV_CLK_PLL_H > +#define __DRV_CLK_PLL_H > + > +/* > + * This is a private header file. DO NOT include it except clk-*.c. > + */ > + > +#include > +#include > +#include > + > +struct mtk_clk_pll { > + struct clk_hw hw; > + void __iomem *base_addr; > + void __iomem *pwr_addr; > + u32 en_mask; > + u32 flags; > + spinlock_t *lock; > +}; > + > +#define to_mtk_clk_pll(_hw) container_of(_hw, struct mtk_clk_pll, hw) Static inline please. Best regards, Tomasz -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/