Let's make the raw NAND API consistent by patching all helpers to
take a nand_chip object instead of an mtd_info one.

Now is nand_release()'s turn.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Let's make the raw NAND API consistent by patching all helpers to take
a nand_chip object instead of an mtd_info one.

We start with nand_scan().
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Let's make the raw NAND API consistent by patching all helpers and
hooks to take a nand_chip object instead of an mtd_info one or
remove the mtd_info object when both are passed.

In order to do that, we first need to update the platform_nand_ctrl
hooks to take a nand_chip object instead of an mtd_info.

We add temporary plat_nand_xxx() wrappers to the do the mtd -> chip
conversion, but those will be dropped when patching nand_chip hooks to
take a nand_chip object.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NAlexander Sverdlin <alexander.sverdlin@gmail.com>
Acked-by: NAlexander Sverdlin <alexander.sverdlin@gmail.com>
Acked-by: NRobert Jarzmik <robert.jarzmik@free.fr>
Acked-by: NKrzysztof Halasa <khalasa@piap.pl>
Acked-by: NPaul Burton <paul.burton@mips.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:

struct foo {
	int stuff;
        void *entry[];
};

instance = devm_kzalloc(dev, sizeof(struct foo) + sizeof(void *) * count, GFP_KERNEL);

Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:

instance = devm_kzalloc(dev, struct_size(instance, entry, count), GFP_KERNEL);

This issue was detected with the help of Coccinelle.
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:

struct foo {
	int stuff;
        void *entry[];
};

instance = devm_kzalloc(dev, sizeof(struct foo) + sizeof(void *) * count, GFP_KERNEL);

Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:

instance = devm_kzalloc(dev, struct_size(instance, entry, count), GFP_KERNEL);
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

In preparation to remove the node name pointer from struct device_node,
convert printf users to use the %pOFn format specifier.

Cc: Boris Brezillon <boris.brezillon@bootlin.com>
Cc: Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Brian Norris <computersforpeace@gmail.com>
Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: linux-mtd@lists.infradead.org
Signed-off-by: NRob Herring <robh@kernel.org>
Acked-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

A recent commit removed the incorrect use of phys_to_dma from this
driver, but failed to remove the dma-direct.h include, so do that
now.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Use SPDX-License-Identifier instead of the license boilerplates.

This conversion makes it easier for us to scan the license, then
I notice license mismatch problems.

The license blocks in denali* indicate GPL-2.0 "only", while the
MODULE_LICENSE in denali.c and denali_dt.c is GPL-2.0 "or later"
as explained in include/linux/module.h as follows:

  "GPL"                           [GNU Public License v2 or later]
  "GPL v2"                        [GNU Public License v2]

I fixed the MODULE_LICENSE tags, assuming the license blocks are
the authors' intention.

Also, add missing MODULE_DESCRIPTION/AUTHOR to denali.c

While I am touching the license things, I added my credit to denali.c
because this driver was largely re-written by me in 2017.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Newer versions of the IFC controller use a different method of initializing the
internal SRAM: Instead of reading from flash, a bit in the NAND configuration
register has to be set in order to trigger the self-initializing process.
Signed-off-by: NKurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

The SRAM initialization might fail. If that happens further NAND operations
won't be successful. Therefore, the chip init routine should fail if the SRAM
initialization didn't work.
Signed-off-by: NKurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

A stale document about the old pxa3cc_nand.c driver is available in
Documentation/mtd/nand/. Rewrite the parts that explain the IP itself
and some non-trivial choices made in the driver directly in
marvell_nand.c to then be able to remove this file.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

This patch is a patch to support TOSHIBA MEMORY CORPORATION BENAND
memory devices.  Check the status of the built-in ECC with the Read
Status command without using the vendor specific command.  The Read
Status command only knows whether there was bitflips above the
threshold and can not get accurate bitflips.  For now, I set
max_bitflips mtd->bitflip_threshold.
Signed-off-by: NKOBAYASHI Yoshitake <yoshitake.kobayashi@toshiba.co.jp>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

The marvell_nfc_init() function fiddles with some bits of a system
controller on Armada 7K/8K. However, it does a read/modify/write
sequence on GENCONF_CLK_GATING_CTRL and GENCONF_ND_CLK_CTRL, which
isn't safe from a concurrency point of view, as the regmap lock isn't
taken accross the read/modify/write sequence. To solve this issue, use
regmap_update_bits().

While at it, since the "reg" variable is no longer needed for the
read/modify/write sequences, get rid of it for the regmap_write() to
GENCONF_SOC_DEVICE_MUX, and directly pass the value to be written as
argument.

Fixes: 02f26ecf ("mtd: nand: add reworked Marvell NAND controller driver")
Signed-off-by: NThomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Commit c120e75e ("mtd: nand: use read_oob() instead of cmdfunc()
for bad block check") removed this only user of the ->read_word()
method but kept the hook in place. Remove it now.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

In order to make sure we use ->exec_op() to wait for chip readiness
when it's available we provide an helper that does the selection
between ->exec_op(), udelay(chip->chip_delay) and nand_wait_ready()
based on what's implemented by the controller driver.

We then use it in nand_wait_readrdy().
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

The logic to wait for chip readiness after a page read has been
duplicated in nand_do_read_ops() and nand_do_read_oob(). Provide an
helper that does the right thing and call it where appropriate.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Commit 49aa76b1 ("mtd: rawnand: do not execute nand_scan_ident()
if maxchips is zero") gave a new meaning for calling nand_scan_ident()
with maxchips=0.

It is a special usage for some drivers such as docg4, but actually
the Denali driver may pass maxchips=0 to nand_scan() when the driver
is enabled but no NAND chip is found on the board for some reasons.

If nand_scan_with_ids() is called with maxchips=0, nand_scan_ident()
is skipped, then nand_set_defaults() is skipped as well.  Thus, the
driver must set chip->controller beforehand.  Otherwise, nand_attach()
causes NULL pointer dereference.

In fact, the Denali controller knows the number of connected chips
before calling nand_scan_ident(); if DEVICE_RESET fails, there is no
chip in that chip select.  Then, denali_reset_banks() sets the maxchips
to the number of detected chips.  If no chip is found, maxchips is zero.

In this case, there is no point for calling nand_scan() because we know
it will fail for sure.  Let's make the probe function fail immediately.

Fixes: 49aa76b1 ("mtd: rawnand: do not execute nand_scan_ident() if maxchips is zero")
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

If gcc (e.g. 4.1.2) decides not to inline init_mtd_structs() and
read_id_reg(), this will cause section mismatches, and crashes:

    WARNING: drivers/mtd/nand/raw/docg4.o(.text+0xc10): Section mismatch in reference from the function docg4_attach_chip() to the function .init.text:init_mtd_structs()
    The function docg4_attach_chip() references
    the function __init init_mtd_structs().
    This is often because docg4_attach_chip lacks a __init
    annotation or the annotation of init_mtd_structs is wrong.

    WARNING: drivers/mtd/nand/raw/docg4.o(.text+0xc3e): Section mismatch in reference from the function docg4_attach_chip() to the function .init.text:read_id_reg()
    The function docg4_attach_chip() references
    the function __init read_id_reg().
    This is often because docg4_attach_chip lacks a __init
    annotation or the annotation of read_id_reg is wrong.

Fix this by dropping the now incorrect __init annotations from
init_mtd_structs() and read_id_reg().

Fixes: 66a38478 ("mtd: rawnand: docg4: convert driver to nand_scan()")
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Now that we have the logic for skipping CRC check for static UBI volumes
in the core, let's expose it to users.

This makes use of a padding byte in the volume description data
structure as a flag. This flag only tell for now whether we should skip
the CRC check of a volume.

This checks the UBI volume for which we are trying to skip the CRC check
is static.

Let's also make sure that the flags passed to verify_mkvol_req are
valid.

We voluntarily do not take into account the skip_check flag in
vol_cdev_write() as we want to make sure what we wrote was correctly
written.
Suggested-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NQuentin Schulz <quentin.schulz@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NRichard Weinberger <richard@nod.at>

Some users of static UBI volumes implement their own integrity check,
thus making the volume CRC check done at open time useless. For
instance, this is the case when one use the ubiblock + dm-verity +
squashfs combination, where dm-verity already checks integrity of the
block device but this time at the block granularity instead of verifying
the whole volume.

Skipping this test drastically improves the boot-time.
Suggested-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NQuentin Schulz <quentin.schulz@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NRichard Weinberger <richard@nod.at>
Signed-off-by: NRichard Weinberger <richard@nod.at>

We cannot do it last, otherwithse it will be skipped for dynamic
volumes.
Reported-by: NLachmann, Juergen <juergen.lachmann@harman.com>
Fixes: 34653fd8 ("ubi: fastmap: Check each mapping only once")
Signed-off-by: NRichard Weinberger <richard@nod.at>

The driver uses genalloc functions. Select GENERIC_ALLOCATOR to prevent
build errors when selected through COMPILE_TEST.

Fixes: 88a40e7d ("mtd: rawnand: atmel: Allow selection of this driver when COMPILE_TEST=y")
Reported-by: NRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Acked-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Acked-by: NRandy Dunlap <rdunlap@infradead.org>

By calling spi_mem_get_name(), the driver of the (Q)SPI controller can
set a custom name for the memory device if necessary.
This is useful to keep mtdparts compatible when controller drivers are
ported from the MTD to the SPI layer.
Suggested-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NFrieder Schrempf <frieder.schrempf@exceet.de>
Acked-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Now that it is possible to do dynamic allocations during the
identification phase, convert the onfi_params structure (which is only
needed with ONFI compliant chips) into a pointer that will be allocated
only if needed.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Commit 59b356ff ("mtd: m25p80: restore the status of SPI flash when
exiting") is the latest from a long history of attempts to add reboot
handling to handle stateful addressing modes on SPI flash. Some prior
mostly-related discussions:

http://lists.infradead.org/pipermail/linux-mtd/2013-March/046343.html
[PATCH 1/3] mtd: m25p80: utilize dedicated 4-byte addressing commands

http://lists.infradead.org/pipermail/barebox/2014-September/020682.html
[RFC] MTD m25p80 3-byte addressing and boot problem

http://lists.infradead.org/pipermail/linux-mtd/2015-February/057683.html
[PATCH 2/2] m25p80: if supported put chip to deep power down if not used

Previously, attempts to add reboot-time software reset handling were
rejected, but the latest attempt was not.

Quick summary of the problem:
Some systems (e.g., boot ROM or bootloader) assume that they can read
initial boot code from their SPI flash using 3-byte addressing. If the
flash is left in 4-byte mode after reset, these systems won't boot. The
above patch provided a shutdown/remove hook to attempt to reset the
addressing mode before we reboot. Notably, this patch misses out on
huge classes of unexpected reboots (e.g., crashes, watchdog resets).

Unfortunately, it is essentially impossible to solve this problem 100%:
if your system doesn't know how to reset the SPI flash to power-on
defaults at initialization time, no amount of software can really rescue
you -- there will always be a chance of some unexpected reset that
leaves your flash in an addressing mode that your boot sequence didn't
expect.

While it is not directly harmful to perform hacks like the
aforementioned commit on all 4-byte addressing flash, a
properly-designed system should not need the hack -- and in fact,
providing this hack may mask the fact that a given system is indeed
broken. So this patch attempts to apply this unsound hack more narrowly,
providing a strong suggestion to developers and system designers that
this is truly a hack. With luck, system designers can catch their errors
early on in their development cycle, rather than applying this hack long
term. But apparently enough systems are out in the wild that we still
have to provide this hack.

Document a new device tree property to denote systems that do not have a
proper hardware (or software) reset mechanism, and apply the hack (with
a loud warning) only in this case.
Signed-off-by: NBrian Norris <computersforpeace@gmail.com>
Reviewed-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

wait_for_completion_timeout returns an unsigned long not an int, so
let's check its return value directly instead of storing it in ret,
and avoid checking for negative values since this cannot happen.
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

GPIO consumers now include <linux/gpio/consumer.h> instead of
<linux/gpio.h> if they can.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

This driver doesn't specify parsers so it can use that little helper.
Signed-off-by: NRafał Miłecki <rafal@milecki.pl>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

The format specifier "%p" can leak kernel addresses.
Use "%pK" instead.
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@st.com>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Implement suspend/resume hooks.
Signed-off-by: NClaudiu Beznea <claudiu.beznea@microchip.com>
Reviewed-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Thanks to the migration of all drivers to use nand_scan() and the
related nand_controller_ops, we can now allocate data during the
detection phase. Let's do it first for the NAND model parameter which
is allocated in nand_detect().
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Both nand_scan_ident() and nand_scan_tail() helpers used to be called
directly from controller drivers that needed to tweak some ECC-related
parameters before nand_scan_tail(). This separation prevented dynamic
allocations during the phase of NAND identification, which was
inconvenient.

All controller drivers have been moved to use nand_scan(), in
conjunction with the chip->ecc.[attach|detach]_chip() hooks that
actually do the required tweaking sequence between both ident/tail
calls, allowing programmers to use dynamic allocation as they need all
across the scanning sequence.

Declare nand_scan_[ident|tail]() statically now.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Two helpers have been added to the core to do all kind of controller
side configuration/initialization between the detection phase and the
final NAND scan. Implement these hooks so that we can convert the driver
to just use nand_scan() instead of the nand_scan_ident() +
nand_scan_tail() pair.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

A comment in the probe declares that values are assigned to ecc.size
and ecc.bytes, but these values will be overwritten. This is not
entirely right as they are overwritten only if
mtd->writesize >= 512. Let's clarify this by moving these assignations
to txx9ndfmc_nand_scan().
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Two helpers have been added to the core to do all kind of controller
side configuration/initialization between the detection phase and the
final NAND scan. Implement these hooks so that we can convert the driver
to just use nand_scan() instead of the nand_scan_ident() +
nand_scan_tail() pair.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Two helpers have been added to the core to do all kind of controller
side configuration/initialization between the detection phase and the
final NAND scan. Implement these hooks so that we can convert the driver
to just use nand_scan() instead of the nand_scan_ident() +
nand_scan_tail() pair.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Prepare the migration to nand_scan() by moving both calls to
nand_scan_ident() and nand_scan_tail() in a single spot.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

An error after nand_scan_tail() should trigger a nand_cleanup(), not a
nand_release() as mtd_device_register() (or one of its variants) has not
been called and there is no need to deregister any MTD device yet.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>

Two helpers have been added to the core to do all kind of controller
side configuration/initialization between the detection phase and the
final NAND scan. Implement these hooks so that we can convert the driver
to just use nand_scan() instead of the nand_scan_ident() +
nand_scan_tail() pair.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Some driver (eg. docg4) will need to handle themselves the
identification phase. As part of the migration to use nand_scan()
everywhere (which will unconditionnaly call nand_scan_ident()), we add
a condition at the start of nand_scan_with_ids() to jump over
nand_scan_ident() if the maxchips parameters is zero, meaning that the
driver does not want the core to handle this phase.
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@bootlin.com>