linux-mtd.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
To: Boris Brezillon <bbrezillon@kernel.org>,
	David Woodhouse <dwmw2@infradead.org>,
	Brian Norris <computersforpeace@gmail.com>,
	"Marek Vasut" <marek.vasut@gmail.com>,
	Richard Weinberger <richard@nod.at>,
	"linux-mtd@lists.infradead.org" <linux-mtd@lists.infradead.org>
Cc: Mark Rutland <mark.rutland@arm.com>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
	Pawel Moll <pawel.moll@arm.com>,
	Ian Campbell <ijc+devicetree@hellion.org.uk>,
	Linus Walleij <linus.walleij@linaro.org>,
	"miquel.raynal@bootlin.com" <miquel.raynal@bootlin.com>,
	Rob Herring <robh+dt@kernel.org>,
	Kumar Gala <galak@codeaurora.org>,
	Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Subject: Re: [v2,11/15] mtd: maps: Merge physmap_of.c into physmap-core.c
Date: Tue, 2 Apr 2019 23:44:31 +0000	[thread overview]
Message-ID: <10efc9a0980f4bc2b06b4149d9a4ff90@svr-chch-ex1.atlnz.lc> (raw)
In-Reply-To: 20181019074908.13226-12-boris.brezillon@bootlin.com

Hi Boris,

I've just come across the commit from this patch.

On 19/10/18 8:49 PM, Boris Brezillon wrote:
> There's no real reason to have two separate driver for the DT and pdata
> case. Just do what we do everywhere else and handle DT and pdata
> parsing in the same driver.
> 
> Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
> Reviewed-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
> ---
> Changes in v2:
> - Add Ricardo's R-b
> - Fix 2 checks (reported by Ricardo)
> ---
>   drivers/mtd/maps/Kconfig           |   4 +-
>   drivers/mtd/maps/Makefile          |   7 +-
>   drivers/mtd/maps/physmap-core.c    | 261 +++++++++++++++++++++++---
>   drivers/mtd/maps/physmap_of_core.c | 368 -------------------------------------
>   4 files changed, 240 insertions(+), 400 deletions(-)
>   delete mode 100644 drivers/mtd/maps/physmap_of_core.c
> 
> diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
> index afb36bff13a7..5bffebacce86 100644
> --- a/drivers/mtd/maps/Kconfig
> +++ b/drivers/mtd/maps/Kconfig
> @@ -66,8 +66,8 @@ config MTD_PHYSMAP_BANKWIDTH
>   	  used internally by the CFI drivers.
>   
>   config MTD_PHYSMAP_OF
> -	tristate "Memory device in physical memory map based on OF description"
> -	depends on OF && (MTD_CFI || MTD_JEDECPROBE || MTD_ROM || MTD_RAM)
> +	bool "Memory device in physical memory map based on OF description"
> +	depends on OF && MTD_PHYSMAP

Previously we could select MTD_PHYSMAP_OF if MTD_RAM was defined. After 
this change we can't select this because MTD_PHYSMAP does not have 
MTD_RAM as dependency.

I'm just in the process of porting the latest kernel to a platform with 
parallel SRAM and NAND via a dedicated controller. I haven't got to a 
point where the platform is booting but I suspect I'll loose my SRAM.

>   	help
>   	  This provides a 'mapping' driver which allows the NOR Flash, ROM
>   	  and RAM driver code to communicate with chips which are mapped
> diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile
> index 2574909edffd..ad32b185a120 100644
> --- a/drivers/mtd/maps/Makefile
> +++ b/drivers/mtd/maps/Makefile
> @@ -18,13 +18,10 @@ obj-$(CONFIG_MTD_CK804XROM)	+= ck804xrom.o
>   obj-$(CONFIG_MTD_TSUNAMI)	+= tsunami_flash.o
>   obj-$(CONFIG_MTD_PXA2XX)	+= pxa2xx-flash.o
>   physmap-objs-y			+= physmap-core.o
> +physmap-objs-$(CONFIG_MTD_PHYSMAP_OF_VERSATILE) += physmap_of_versatile.o
> +physmap-objs-$(CONFIG_MTD_PHYSMAP_OF_GEMINI) += physmap_of_gemini.o
>   physmap-objs			:= $(physmap-objs-y)
>   obj-$(CONFIG_MTD_PHYSMAP)	+= physmap.o
> -physmap_of-objs-y		+= physmap_of_core.o
> -physmap_of-objs-$(CONFIG_MTD_PHYSMAP_OF_VERSATILE) += physmap_of_versatile.o
> -physmap_of-objs-$(CONFIG_MTD_PHYSMAP_OF_GEMINI) += physmap_of_gemini.o
> -physmap_of-objs			:= $(physmap_of-objs-y)
> -obj-$(CONFIG_MTD_PHYSMAP_OF)	+= physmap_of.o
>   obj-$(CONFIG_MTD_PISMO)		+= pismo.o
>   obj-$(CONFIG_MTD_PMC_MSP_EVM)   += pmcmsp-flash.o
>   obj-$(CONFIG_MTD_PCMCIA)	+= pcmciamtd.o
> diff --git a/drivers/mtd/maps/physmap-core.c b/drivers/mtd/maps/physmap-core.c
> index e27051bc5dc6..07af8368d173 100644
> --- a/drivers/mtd/maps/physmap-core.c
> +++ b/drivers/mtd/maps/physmap-core.c
> @@ -6,6 +6,13 @@
>    * Author: Jun Sun, jsun@mvista.com or jsun@junsun.net
>    *
>    * 031022 - [jsun] add run-time configure and partition setup
> + *
> + * Device tree support:
> + *    Copyright (C) 2006 MontaVista Software Inc.
> + *    Author: Vitaly Wool <vwool@ru.mvista.com>
> + *
> + *    Revised to handle newer style flash binding by:
> + *    Copyright (C) 2007 David Gibson, IBM Corporation.
>    */
>   
>   #include <linux/module.h>
> @@ -20,7 +27,12 @@
>   #include <linux/mtd/partitions.h>
>   #include <linux/mtd/physmap.h>
>   #include <linux/mtd/concat.h>
> +#include <linux/mtd/cfi_endian.h>
>   #include <linux/io.h>
> +#include <linux/of_device.h>
> +
> +#include "physmap_of_gemini.h"
> +#include "physmap_of_versatile.h"
>   
>   struct physmap_flash_info {
>   	unsigned int		nmaps;
> @@ -29,6 +41,10 @@ struct physmap_flash_info {
>   	struct map_info		*maps;
>   	spinlock_t		vpp_lock;
>   	int			vpp_refcnt;
> +	const char		*probe_type;
> +	const char * const	*part_types;
> +	unsigned int		nparts;
> +	const struct mtd_partition *parts;
>   };
>   
>   static int physmap_flash_remove(struct platform_device *dev)
> @@ -41,8 +57,6 @@ static int physmap_flash_remove(struct platform_device *dev)
>   	if (!info)
>   		return 0;
>   
> -	physmap_data = dev_get_platdata(&dev->dev);
> -
>   	if (info->cmtd) {
>   		err = mtd_device_unregister(info->cmtd);
>   		if (err)
> @@ -57,7 +71,8 @@ static int physmap_flash_remove(struct platform_device *dev)
>   			map_destroy(info->mtds[i]);
>   	}
>   
> -	if (physmap_data->exit)
> +	physmap_data = dev_get_platdata(&dev->dev);
> +	if (physmap_data && physmap_data->exit)
>   		physmap_data->exit(dev);
>   
>   	return 0;
> @@ -89,6 +104,172 @@ static void physmap_set_vpp(struct map_info *map, int state)
>   	spin_unlock_irqrestore(&info->vpp_lock, flags);
>   }
>   
> +#if IS_ENABLED(CONFIG_MTD_PHYSMAP_OF)
> +static const struct of_device_id of_flash_match[] = {
> +	{
> +		.compatible = "cfi-flash",
> +		.data = "cfi_probe",
> +	},
> +	{
> +		/*
> +		 * FIXME: JEDEC chips can't be safely and reliably
> +		 * probed, although the mtd code gets it right in
> +		 * practice most of the time.  We should use the
> +		 * vendor and device ids specified by the binding to
> +		 * bypass the heuristic probe code, but the mtd layer
> +		 * provides, at present, no interface for doing so
> +		 * :(.
> +		 */
> +		.compatible = "jedec-flash",
> +		.data = "jedec_probe",
> +	},
> +	{
> +		.compatible = "mtd-ram",
> +		.data = "map_ram",
> +	},
> +	{
> +		.compatible = "mtd-rom",
> +		.data = "map_rom",
> +	},
> +	{
> +		.type = "rom",
> +		.compatible = "direct-mapped"
> +	},
> +	{ /* sentinel */ },
> +};
> +MODULE_DEVICE_TABLE(of, of_flash_match);
> +
> +static const char * const of_default_part_probes[] = {
> +	"cmdlinepart", "RedBoot", "ofpart", "ofoldpart", NULL
> +};
> +
> +static const char * const *of_get_part_probes(struct platform_device *dev)
> +{
> +	struct device_node *dp = dev->dev.of_node;
> +	const char **res;
> +	int count;
> +
> +	count = of_property_count_strings(dp, "linux,part-probe");
> +	if (count < 0)
> +		return of_default_part_probes;
> +
> +	res = devm_kcalloc(&dev->dev, count + 1, sizeof(*res), GFP_KERNEL);
> +	if (!res)
> +		return NULL;
> +
> +	count = of_property_read_string_array(dp, "linux,part-probe", res,
> +					      count);
> +	if (count < 0)
> +		return NULL;
> +
> +	return res;
> +}
> +
> +static const char *of_select_probe_type(struct platform_device *dev)
> +{
> +	struct device_node *dp = dev->dev.of_node;
> +	const struct of_device_id *match;
> +	const char *probe_type;
> +
> +	match = of_match_device(of_flash_match, &dev->dev);
> +	probe_type = match->data;
> +	if (probe_type)
> +		return probe_type;
> +
> +	dev_warn(&dev->dev,
> +		 "Device tree uses obsolete \"direct-mapped\" flash binding\n");
> +
> +	of_property_read_string(dp, "probe-type", &probe_type);
> +	if (!probe_type)
> +		return NULL;
> +
> +	if (!strcmp(probe_type, "CFI")) {
> +		probe_type = "cfi_probe";
> +	} else if (!strcmp(probe_type, "JEDEC")) {
> +		probe_type = "jedec_probe";
> +	} else if (!strcmp(probe_type, "ROM")) {
> +		probe_type = "map_rom";
> +	} else {
> +		dev_warn(&dev->dev,
> +			 "obsolete_probe: don't know probe type '%s', mapping as rom\n",
> +			 probe_type);
> +		probe_type = "map_rom";
> +	}
> +
> +	return probe_type;
> +}
> +
> +static int physmap_flash_of_init(struct platform_device *dev)
> +{
> +	struct physmap_flash_info *info = platform_get_drvdata(dev);
> +	struct device_node *dp = dev->dev.of_node;
> +	const char *mtd_name = NULL;
> +	int err, swap = 0;
> +	bool map_indirect;
> +	unsigned int i;
> +	u32 bankwidth;
> +
> +	if (!dp)
> +		return -EINVAL;
> +
> +	info->probe_type = of_select_probe_type(dev);
> +
> +	info->part_types = of_get_part_probes(dev);
> +	if (!info->part_types)
> +		return -ENOMEM;
> +
> +	of_property_read_string(dp, "linux,mtd-name", &mtd_name);
> +
> +	map_indirect = of_property_read_bool(dp, "no-unaligned-direct-access");
> +
> +	err = of_property_read_u32(dp, "bank-width", &bankwidth);
> +	if (err) {
> +		dev_err(&dev->dev, "Can't get bank width from device tree\n");
> +		return err;
> +	}
> +
> +	if (of_property_read_bool(dp, "big-endian"))
> +		swap = CFI_BIG_ENDIAN;
> +	else if (of_property_read_bool(dp, "little-endian"))
> +		swap = CFI_LITTLE_ENDIAN;
> +
> +	for (i = 0; i < info->nmaps; i++) {
> +		info->maps[i].name = mtd_name;
> +		info->maps[i].swap = swap;
> +		info->maps[i].bankwidth = bankwidth;
> +		info->maps[i].device_node = dp;
> +
> +		err = of_flash_probe_gemini(dev, dp, &info->maps[i]);
> +		if (err)
> +			return err;
> +
> +		err = of_flash_probe_versatile(dev, dp, &info->maps[i]);
> +		if (err)
> +			return err;
> +
> +		/*
> +		 * On some platforms (e.g. MPC5200) a direct 1:1 mapping
> +		 * may cause problems with JFFS2 usage, as the local bus (LPB)
> +		 * doesn't support unaligned accesses as implemented in the
> +		 * JFFS2 code via memcpy(). By setting NO_XIP, the
> +		 * flash will not be exposed directly to the MTD users
> +		 * (e.g. JFFS2) any more.
> +		 */
> +		if (map_indirect)
> +			info->maps[i].phys = NO_XIP;
> +	}
> +
> +	return 0;
> +}
> +#else /* IS_ENABLED(CONFIG_MTD_PHYSMAP_OF) */
> +#define of_flash_match NULL
> +
> +static int physmap_flash_of_init(struct platform_device *dev)
> +{
> +	return -ENOTSUPP;
> +}
> +#endif /* IS_ENABLED(CONFIG_MTD_PHYSMAP_OF) */
> +
>   static const char * const rom_probe_types[] = {
>   	"cfi_probe", "jedec_probe", "qinfo_probe", "map_rom", NULL
>   };
> @@ -97,18 +278,46 @@ static const char * const part_probe_types[] = {
>   	"cmdlinepart", "RedBoot", "afs", NULL
>   };
>   
> -static int physmap_flash_probe(struct platform_device *dev)
> +static int physmap_flash_pdata_init(struct platform_device *dev)
>   {
> +	struct physmap_flash_info *info = platform_get_drvdata(dev);
>   	struct physmap_flash_data *physmap_data;
> +	unsigned int i;
> +	int err;
> +
> +	physmap_data = dev_get_platdata(&dev->dev);
> +	if (!physmap_data)
> +		return -EINVAL;
> +
> +	info->probe_type = physmap_data->probe_type;
> +	info->part_types = physmap_data->part_probe_types ? : part_probe_types;
> +	info->parts = physmap_data->parts;
> +	info->nparts = physmap_data->nr_parts;
> +
> +	if (physmap_data->init) {
> +		err = physmap_data->init(dev);
> +		if (err)
> +			return err;
> +	}
> +
> +	for (i = 0; i < info->nmaps; i++) {
> +		info->maps[i].bankwidth = physmap_data->width;
> +		info->maps[i].pfow_base = physmap_data->pfow_base;
> +		info->maps[i].set_vpp = physmap_set_vpp;
> +	}
> +
> +	return 0;
> +}
> +
> +static int physmap_flash_probe(struct platform_device *dev)
> +{
>   	struct physmap_flash_info *info;
>   	const char * const *probe_type;
> -	const char * const *part_types;
>   	int err = 0;
>   	int i;
>   
> -	physmap_data = dev_get_platdata(&dev->dev);
> -	if (!physmap_data)
> -		return -ENODEV;
> +	if (!dev->dev.of_node && !dev_get_platdata(&dev->dev))
> +		return -EINVAL;
>   
>   	info = devm_kzalloc(&dev->dev, sizeof(*info), GFP_KERNEL);
>   	if (!info)
> @@ -132,14 +341,16 @@ static int physmap_flash_probe(struct platform_device *dev)
>   	if (!info->mtds)
>   		return -ENOMEM;
>   
> -	if (physmap_data->init) {
> -		err = physmap_data->init(dev);
> -		if (err)
> -			goto err_out;
> -	}
> -
>   	platform_set_drvdata(dev, info);
>   
> +	if (dev->dev.of_node)
> +		err = physmap_flash_of_init(dev);
> +	else
> +		err = physmap_flash_pdata_init(dev);
> +
> +	if (err)
> +		return err;
> +
>   	for (i = 0; i < info->nmaps; i++) {
>   		struct resource *res;
>   
> @@ -154,22 +365,22 @@ static int physmap_flash_probe(struct platform_device *dev)
>   			   res);
>   
>   		info->maps[i].name = dev_name(&dev->dev);
> -		info->maps[i].phys = res->start;
> +
> +		if (!info->maps[i].phys)
> +			info->maps[i].phys = res->start;
> +
>   		info->maps[i].size = resource_size(res);
> -		info->maps[i].bankwidth = physmap_data->width;
> -		info->maps[i].set_vpp = physmap_set_vpp;
> -		info->maps[i].pfow_base = physmap_data->pfow_base;
>   		info->maps[i].map_priv_1 = (unsigned long)dev;
>   
>   		simple_map_init(&info->maps[i]);
>   
>   		probe_type = rom_probe_types;
> -		if (!physmap_data->probe_type) {
> +		if (!info->probe_type) {
>   			for (; !info->mtds[i] && *probe_type; probe_type++)
>   				info->mtds[i] = do_map_probe(*probe_type,
>   							     &info->maps[i]);
>   		} else {
> -			info->mtds[i] = do_map_probe(physmap_data->probe_type,
> +			info->mtds[i] = do_map_probe(info->probe_type,
>   						     &info->maps[i]);
>   		}
>   
> @@ -197,11 +408,9 @@ static int physmap_flash_probe(struct platform_device *dev)
>   
>   	spin_lock_init(&info->vpp_lock);
>   
> -	part_types = physmap_data->part_probe_types ? : part_probe_types;
> -
> -	err = mtd_device_parse_register(info->cmtd, part_types, NULL,
> -					physmap_data->parts,
> -					physmap_data->nr_parts);
> +	mtd_set_of_node(info->cmtd, dev->dev.of_node);
> +	err = mtd_device_parse_register(info->cmtd, info->part_types, NULL,
> +					info->parts, info->nparts);
>   	if (err)
>   		goto err_out;
>   
> @@ -232,6 +441,7 @@ static struct platform_driver physmap_flash_driver = {
>   	.shutdown	= physmap_flash_shutdown,
>   	.driver		= {
>   		.name	= "physmap-flash",
> +		.of_match_table = of_flash_match,
>   	},
>   };
>   
> @@ -286,6 +496,7 @@ module_exit(physmap_exit);
>   
>   MODULE_LICENSE("GPL");
>   MODULE_AUTHOR("David Woodhouse <dwmw2@infradead.org>");
> +MODULE_AUTHOR("Vitaly Wool <vwool@ru.mvista.com>");
>   MODULE_DESCRIPTION("Generic configurable MTD map driver");
>   
>   /* legacy platform drivers can't hotplug or coldplg */
> diff --git a/drivers/mtd/maps/physmap_of_core.c b/drivers/mtd/maps/physmap_of_core.c
> deleted file mode 100644
> index ece605d78c21..000000000000
> --- a/drivers/mtd/maps/physmap_of_core.c
> +++ /dev/null
> @@ -1,368 +0,0 @@
> -/*
> - * Flash mappings described by the OF (or flattened) device tree
> - *
> - * Copyright (C) 2006 MontaVista Software Inc.
> - * Author: Vitaly Wool <vwool@ru.mvista.com>
> - *
> - * Revised to handle newer style flash binding by:
> - *   Copyright (C) 2007 David Gibson, IBM Corporation.
> - *
> - * This program is free software; you can redistribute  it and/or modify it
> - * under  the terms of  the GNU General  Public License as published by the
> - * Free Software Foundation;  either version 2 of the  License, or (at your
> - * option) any later version.
> - */
> -
> -#include <linux/module.h>
> -#include <linux/types.h>
> -#include <linux/device.h>
> -#include <linux/mtd/mtd.h>
> -#include <linux/mtd/map.h>
> -#include <linux/mtd/partitions.h>
> -#include <linux/mtd/concat.h>
> -#include <linux/mtd/cfi_endian.h>
> -#include <linux/of.h>
> -#include <linux/of_address.h>
> -#include <linux/of_platform.h>
> -#include <linux/slab.h>
> -#include "physmap_of_gemini.h"
> -#include "physmap_of_versatile.h"
> -
> -struct of_flash_list {
> -	struct mtd_info *mtd;
> -	struct map_info map;
> -};
> -
> -struct of_flash {
> -	struct mtd_info		*cmtd;
> -	int list_size; /* number of elements in of_flash_list */
> -	struct of_flash_list	list[0];
> -};
> -
> -static int of_flash_remove(struct platform_device *dev)
> -{
> -	struct of_flash *info;
> -	int i;
> -
> -	info = dev_get_drvdata(&dev->dev);
> -	if (!info)
> -		return 0;
> -	dev_set_drvdata(&dev->dev, NULL);
> -
> -	if (info->cmtd) {
> -		mtd_device_unregister(info->cmtd);
> -		if (info->cmtd != info->list[0].mtd)
> -			mtd_concat_destroy(info->cmtd);
> -	}
> -
> -	for (i = 0; i < info->list_size; i++)
> -		if (info->list[i].mtd)
> -			map_destroy(info->list[i].mtd);
> -
> -	return 0;
> -}
> -
> -static const char * const rom_probe_types[] = {
> -	"cfi_probe", "jedec_probe", "map_rom" };
> -
> -/* Helper function to handle probing of the obsolete "direct-mapped"
> - * compatible binding, which has an extra "probe-type" property
> - * describing the type of flash probe necessary. */
> -static struct mtd_info *obsolete_probe(struct platform_device *dev,
> -				       struct map_info *map)
> -{
> -	struct device_node *dp = dev->dev.of_node;
> -	const char *of_probe;
> -	struct mtd_info *mtd;
> -	int i;
> -
> -	dev_warn(&dev->dev, "Device tree uses obsolete \"direct-mapped\" "
> -		 "flash binding\n");
> -
> -	of_probe = of_get_property(dp, "probe-type", NULL);
> -	if (!of_probe) {
> -		for (i = 0; i < ARRAY_SIZE(rom_probe_types); i++) {
> -			mtd = do_map_probe(rom_probe_types[i], map);
> -			if (mtd)
> -				return mtd;
> -		}
> -		return NULL;
> -	} else if (strcmp(of_probe, "CFI") == 0) {
> -		return do_map_probe("cfi_probe", map);
> -	} else if (strcmp(of_probe, "JEDEC") == 0) {
> -		return do_map_probe("jedec_probe", map);
> -	} else {
> -		if (strcmp(of_probe, "ROM") != 0)
> -			dev_warn(&dev->dev, "obsolete_probe: don't know probe "
> -				 "type '%s', mapping as rom\n", of_probe);
> -		return do_map_probe("map_rom", map);
> -	}
> -}
> -
> -/* When partitions are set we look for a linux,part-probe property which
> -   specifies the list of partition probers to use. If none is given then the
> -   default is use. These take precedence over other device tree
> -   information. */
> -static const char * const part_probe_types_def[] = {
> -	"cmdlinepart", "RedBoot", "ofpart", "ofoldpart", NULL };
> -
> -static const char * const *of_get_probes(struct device_node *dp)
> -{
> -	const char **res;
> -	int count;
> -
> -	count = of_property_count_strings(dp, "linux,part-probe");
> -	if (count < 0)
> -		return part_probe_types_def;
> -
> -	res = kcalloc(count + 1, sizeof(*res), GFP_KERNEL);
> -	if (!res)
> -		return NULL;
> -
> -	count = of_property_read_string_array(dp, "linux,part-probe", res,
> -					      count);
> -	if (count < 0)
> -		return NULL;
> -
> -	return res;
> -}
> -
> -static void of_free_probes(const char * const *probes)
> -{
> -	if (probes != part_probe_types_def)
> -		kfree(probes);
> -}
> -
> -static const struct of_device_id of_flash_match[];
> -static int of_flash_probe(struct platform_device *dev)
> -{
> -	const char * const *part_probe_types;
> -	const struct of_device_id *match;
> -	struct device_node *dp = dev->dev.of_node;
> -	struct resource res;
> -	struct of_flash *info;
> -	const char *probe_type;
> -	const __be32 *width;
> -	int err;
> -	int i;
> -	int count;
> -	const __be32 *p;
> -	int reg_tuple_size;
> -	struct mtd_info **mtd_list = NULL;
> -	resource_size_t res_size;
> -	bool map_indirect;
> -	const char *mtd_name = NULL;
> -
> -	match = of_match_device(of_flash_match, &dev->dev);
> -	if (!match)
> -		return -EINVAL;
> -	probe_type = match->data;
> -
> -	reg_tuple_size = (of_n_addr_cells(dp) + of_n_size_cells(dp)) * sizeof(u32);
> -
> -	of_property_read_string(dp, "linux,mtd-name", &mtd_name);
> -
> -	/*
> -	 * Get number of "reg" tuples. Scan for MTD devices on area's
> -	 * described by each "reg" region. This makes it possible (including
> -	 * the concat support) to support the Intel P30 48F4400 chips which
> -	 * consists internally of 2 non-identical NOR chips on one die.
> -	 */
> -	p = of_get_property(dp, "reg", &count);
> -	if (!p || count % reg_tuple_size != 0) {
> -		dev_err(&dev->dev, "Malformed reg property on %pOF\n",
> -				dev->dev.of_node);
> -		err = -EINVAL;
> -		goto err_flash_remove;
> -	}
> -	count /= reg_tuple_size;
> -
> -	map_indirect = of_property_read_bool(dp, "no-unaligned-direct-access");
> -
> -	err = -ENOMEM;
> -	info = devm_kzalloc(&dev->dev,
> -			    sizeof(struct of_flash) +
> -			    sizeof(struct of_flash_list) * count, GFP_KERNEL);
> -	if (!info)
> -		goto err_flash_remove;
> -
> -	dev_set_drvdata(&dev->dev, info);
> -
> -	mtd_list = kcalloc(count, sizeof(*mtd_list), GFP_KERNEL);
> -	if (!mtd_list)
> -		goto err_flash_remove;
> -
> -	for (i = 0; i < count; i++) {
> -		err = -ENXIO;
> -		if (of_address_to_resource(dp, i, &res)) {
> -			/*
> -			 * Continue with next register tuple if this
> -			 * one is not mappable
> -			 */
> -			continue;
> -		}
> -
> -		dev_dbg(&dev->dev, "of_flash device: %pR\n", &res);
> -
> -		err = -EBUSY;
> -		res_size = resource_size(&res);
> -		info->list[i].map.virt = devm_ioremap_resource(&dev->dev, &res);
> -		if (IS_ERR(info->list[i].map.virt)) {
> -			err = PTR_ERR(info->list[i].map.virt);
> -			goto err_out;
> -		}
> -
> -		err = -ENXIO;
> -		width = of_get_property(dp, "bank-width", NULL);
> -		if (!width) {
> -			dev_err(&dev->dev, "Can't get bank width from device"
> -				" tree\n");
> -			goto err_out;
> -		}
> -
> -		info->list[i].map.name = mtd_name ?: dev_name(&dev->dev);
> -		info->list[i].map.phys = res.start;
> -		info->list[i].map.size = res_size;
> -		info->list[i].map.bankwidth = be32_to_cpup(width);
> -		info->list[i].map.device_node = dp;
> -
> -		if (of_property_read_bool(dp, "big-endian"))
> -			info->list[i].map.swap = CFI_BIG_ENDIAN;
> -		else if (of_property_read_bool(dp, "little-endian"))
> -			info->list[i].map.swap = CFI_LITTLE_ENDIAN;
> -
> -		err = of_flash_probe_gemini(dev, dp, &info->list[i].map);
> -		if (err)
> -			goto err_out;
> -		err = of_flash_probe_versatile(dev, dp, &info->list[i].map);
> -		if (err)
> -			goto err_out;
> -
> -		simple_map_init(&info->list[i].map);
> -
> -		/*
> -		 * On some platforms (e.g. MPC5200) a direct 1:1 mapping
> -		 * may cause problems with JFFS2 usage, as the local bus (LPB)
> -		 * doesn't support unaligned accesses as implemented in the
> -		 * JFFS2 code via memcpy(). By setting NO_XIP, the
> -		 * flash will not be exposed directly to the MTD users
> -		 * (e.g. JFFS2) any more.
> -		 */
> -		if (map_indirect)
> -			info->list[i].map.phys = NO_XIP;
> -
> -		if (probe_type) {
> -			info->list[i].mtd = do_map_probe(probe_type,
> -							 &info->list[i].map);
> -		} else {
> -			info->list[i].mtd = obsolete_probe(dev,
> -							   &info->list[i].map);
> -		}
> -
> -		/* Fall back to mapping region as ROM */
> -		if (!info->list[i].mtd) {
> -			dev_warn(&dev->dev,
> -				"do_map_probe() failed for type %s\n",
> -				 probe_type);
> -
> -			info->list[i].mtd = do_map_probe("map_rom",
> -							 &info->list[i].map);
> -		}
> -		mtd_list[i] = info->list[i].mtd;
> -
> -		err = -ENXIO;
> -		if (!info->list[i].mtd) {
> -			dev_err(&dev->dev, "do_map_probe() failed\n");
> -			goto err_out;
> -		} else {
> -			info->list_size++;
> -		}
> -		info->list[i].mtd->dev.parent = &dev->dev;
> -	}
> -
> -	err = 0;
> -	info->cmtd = NULL;
> -	if (info->list_size == 1) {
> -		info->cmtd = info->list[0].mtd;
> -	} else if (info->list_size > 1) {
> -		/*
> -		 * We detected multiple devices. Concatenate them together.
> -		 */
> -		info->cmtd = mtd_concat_create(mtd_list, info->list_size,
> -					       dev_name(&dev->dev));
> -	}
> -	if (info->cmtd == NULL)
> -		err = -ENXIO;
> -
> -	if (err)
> -		goto err_out;
> -
> -	info->cmtd->dev.parent = &dev->dev;
> -	mtd_set_of_node(info->cmtd, dp);
> -	part_probe_types = of_get_probes(dp);
> -	if (!part_probe_types) {
> -		err = -ENOMEM;
> -		goto err_out;
> -	}
> -	mtd_device_parse_register(info->cmtd, part_probe_types, NULL,
> -			NULL, 0);
> -	of_free_probes(part_probe_types);
> -
> -	kfree(mtd_list);
> -
> -	return 0;
> -
> -err_out:
> -	kfree(mtd_list);
> -err_flash_remove:
> -	of_flash_remove(dev);
> -
> -	return err;
> -}
> -
> -static const struct of_device_id of_flash_match[] = {
> -	{
> -		.compatible	= "cfi-flash",
> -		.data		= (void *)"cfi_probe",
> -	},
> -	{
> -		/* FIXME: JEDEC chips can't be safely and reliably
> -		 * probed, although the mtd code gets it right in
> -		 * practice most of the time.  We should use the
> -		 * vendor and device ids specified by the binding to
> -		 * bypass the heuristic probe code, but the mtd layer
> -		 * provides, at present, no interface for doing so
> -		 * :(. */
> -		.compatible	= "jedec-flash",
> -		.data		= (void *)"jedec_probe",
> -	},
> -	{
> -		.compatible     = "mtd-ram",
> -		.data           = (void *)"map_ram",
> -	},
> -	{
> -		.compatible     = "mtd-rom",
> -		.data           = (void *)"map_rom",
> -	},
> -	{
> -		.type		= "rom",
> -		.compatible	= "direct-mapped"
> -	},
> -	{ },
> -};
> -MODULE_DEVICE_TABLE(of, of_flash_match);
> -
> -static struct platform_driver of_flash_driver = {
> -	.driver = {
> -		.name = "of-flash",
> -		.of_match_table = of_flash_match,
> -	},
> -	.probe		= of_flash_probe,
> -	.remove		= of_flash_remove,
> -};
> -
> -module_platform_driver(of_flash_driver);
> -
> -MODULE_LICENSE("GPL");
> -MODULE_AUTHOR("Vitaly Wool <vwool@ru.mvista.com>");
> -MODULE_DESCRIPTION("Device tree based MTD map driver");
> 


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

  reply	other threads:[~2019-04-02 23:45 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-19  7:48 [PATCH v2 00/15] mtd: maps: physmap cleanups Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 01/15] mtd: maps: physmap: Add SPDX header Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 02/15] mtd: maps: physmap: Rename ->map and ->mtd into ->maps and ->mtds Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 03/15] mtd: maps: physmap: Use platform_get_resource() to retrieve iomem resources Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 04/15] mtd: maps: physmap: Use dev_notice() and a %pR specifier Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 05/15] mtd: maps: physmap: Use devm_ioremap_resource() Boris Brezillon
2018-10-19  7:48 ` [PATCH v2 06/15] mtd: maps: physmap: Remove the MAX_RESOURCES limitation Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 07/15] mtd: maps: physmap: Check mtd_device_{parse_register, unregister}() ret code Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 08/15] mtd: maps: physmap: Return -ENOMEM directly when info allocation fails Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 09/15] mtd: maps: physmap: Fix coding style issues reported by checkpatch Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 10/15] mtd: maps: Prepare merging of physmap and physmap_of Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 11/15] mtd: maps: Merge physmap_of.c into physmap-core.c Boris Brezillon
2019-04-02 23:44   ` Chris Packham [this message]
2019-04-03  2:02     ` [PATCH] mtd: maps: Allow MTD_PHYSMAP with MTD_RAM Chris Packham
2019-05-01 18:02     ` [v2,11/15] mtd: maps: Merge physmap_of.c into physmap-core.c Li Yang
2018-10-19  7:49 ` [PATCH v2 12/15] mtd: maps: Rename physmap_of_{versatile, gemini} into physmap-{versatile, gemini} Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 13/15] dt-binding: mtd: physmap: Document the addr-gpios property Boris Brezillon
2018-10-24 18:38   ` Rob Herring
2018-10-31 10:49   ` Linus Walleij
2018-10-19  7:49 ` [PATCH v2 14/15] mtd: maps: Merge gpio-addr-flash.c into physmap-core.c Boris Brezillon
2018-10-19  7:49 ` [PATCH v2 15/15] mtd: maps: physmap: Invert logic on if/else branch Boris Brezillon
2018-10-22 11:45 ` [PATCH v2 00/15] mtd: maps: physmap cleanups Linus Walleij
2018-10-22 13:41   ` Boris Brezillon
2018-11-05 21:27 ` Boris Brezillon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=10efc9a0980f4bc2b06b4149d9a4ff90@svr-chch-ex1.atlnz.lc \
    --to=chris.packham@alliedtelesis.co.nz \
    --cc=bbrezillon@kernel.org \
    --cc=computersforpeace@gmail.com \
    --cc=devicetree@vger.kernel.org \
    --cc=dwmw2@infradead.org \
    --cc=galak@codeaurora.org \
    --cc=ijc+devicetree@hellion.org.uk \
    --cc=linus.walleij@linaro.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=marek.vasut@gmail.com \
    --cc=mark.rutland@arm.com \
    --cc=miquel.raynal@bootlin.com \
    --cc=pawel.moll@arm.com \
    --cc=ricardo.ribalda@gmail.com \
    --cc=richard@nod.at \
    --cc=robh+dt@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).