Граф коммитов

105 Коммитов

Автор SHA1 Сообщение Дата
Bartosz Golaszewski 165589f0cb nvmem: make the naming of arguments in nvmem_cell_get() consistent
The argument representing the cell name in the nvmem_cell_get() family
of functions is not consistend between function prototypes and
definitions. Name it 'id' in all those routines. This is in line with
other frameworks and can represent both the DT cell name from the
nvmem-cell-names property as well as the con_id field from cell
lookup entries.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:55 +02:00
Bartosz Golaszewski b1c1db9883 nvmem: use SPDX license identifiers
Use SPDX license identiefiers to core nvmem files and remove GPL 2.0
license boilerplate.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski bee1138bea nvmem: add a notifier chain
Add a blocking notifier chain with four events (add and remove for
both devices and cells) so that users can get notified about the
addition of nvmem resources they're waiting for.

We'll use this instead of the at24 setup callback in the mityomapl138
board file.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski 506157be06 nvmem: add support for cell lookups from machine code
Add a way for machine code users to associate devices with nvmem cells.

This restores the support for non-DT systems but following a different
approach. Cells must now be associated with devices using provided
routines and data structures before they can be retrieved using
nvmem_cell_get().

It's still possible to define cells statically in nvmem_config but
cells created this way still need to be associated with consumers using
lookup entries.

Note that nvmem_find() must be moved higher in the source file as we
want to call it from __nvmem_device_get() for devices that don't have
a device node.

The signature of __nvmem_device_get() is also changed as it's no longer
used to retrieve cells.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski e888d445ac nvmem: resolve cells from DT at registration time
Currently we're creating a new cell structure everytime a DT user
calls nvmem_cell_get().

Change this behavior by resolving the cells during nvmem provider
registration and adding all cells to the provider's list. Make
of_nvmem_cell_get() just parse the phandle and look the cell up
in the relevant provider's list.

Don't drop the cell in nvmem_cell_put().

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski b985f4cba6 nvmem: add support for cell info
Add new structs and routines allowing users to define nvmem cells from
machine code. This global list of entries is parsed when a provider
is registered and cells are associated with the relevant nvmem_device
struct.

A possible improvement for the future is to allow users to register
cell tables after the nvmem provider has been registered by updating
the cell list at each call to nvmem_(add|del)_cell_table().

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski c7235ee3f4 nvmem: remove the global cell list
Nvmem subsystem keeps a global list of cells that, for non-DT systems,
can only be referenced by cell name, which makes it impossible to have
more than one nvmem device with cells named the same.

This patch makes every nvmem device the owner of the list of its cells.
This effectively removes the support for non-DT systems, but it will
be reintroduced following a different approach in subsequent patches.

This isn't a problem as support for board files in nvmem is currently
broken anyway: any user that would try to get an nvmem cell from the
global cell list would remove the cell after the calling
nvmem_cell_put(). This can cause anything from a subsequent user not
being able to get the cell to double free errors if more users hold
reference to the same cell at the same time.

Fortunately there are no such users which allows us to rework this part.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski bf58e8820c nvmem: change the signature of nvmem_unregister()
We switched the nvmem framework to using kref instead of manually
checking the current number of users in nvmem_unregister() so this
function can no longer fail. We also converted all remaining users
that still checked the return value of nvmem_unregister() to using
devm_nvmem_register(). Make the routine return void.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski c1de7f43bd nvmem: use kref
Use kref for reference counting. Use an approach similar to the one
seen in the common clock subsystem: don't actually destroy the nvmem
device until the last user puts it. This way we can get rid of the
users check from nvmem_unregister().

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski fa72d847d6 nvmem: check the return value of nvmem_add_cells()
This function can fail so check its return value in nvmem_register()
and act accordingly.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:54 +02:00
Bartosz Golaszewski f9fcb7e36c nvmem: remove a stray newline
There are two empty lines between devm_nvmem_unregister() and
__nvmem_device_get(). Remove one.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:53 +02:00
Bartosz Golaszewski 1852183e14 nvmem: use list_for_each_entry_safe in nvmem_device_remove_all_cells()
Use the provided helper for iterating over list entries without having
to use the list_entry() macro.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:53 +02:00
Bartosz Golaszewski 5db652c9f3 nvmem: remove the name field from struct nvmem_device
This field is never set and is only used in a single error message.
Remove the field and use nvmem_dev_name() instead.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:53 +02:00
Bartosz Golaszewski d7b9fd1669 nvmem: provide nvmem_dev_name()
Kernel users don't have any means of checking the names of nvmem
devices. Add a routine that returns the name of the nvmem provider.

This will be useful for future nvmem notifier subscribers - otherwise
they can't check what device is being added/removed.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-28 15:14:53 +02:00
Srinivas Kandagatla ca6ac25cec nvmem: core: return error code instead of NULL from nvmem_device_get
nvmem_device_get() should return ERR_PTR() on error or valid pointer
on success, but one of the code path seems to return NULL, so fix it.

Reported-by: Niklas Cassel <niklas.cassel@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-09-12 09:46:46 +02:00
Greg Kroah-Hartman 83cf9cd6d5 Merge 4.18-rc5 into char-misc-next
We want the char-misc fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-07-16 09:04:54 +02:00
Srinivas Kandagatla fc82975aea nvmem: core: remove unused nvmem_device ncells member
nvmem ncells can be over written by calling nvmem_add_cells()
multiple times. I see there is no real point of maintaining count
of cells when we have a list of cell.

Remove this to avoid any confusion!

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-07-07 17:30:46 +02:00
Douglas Anderson 87ed1405ef nvmem: Don't let a NULL cell_id for nvmem_cell_get() crash us
In commit ca04d9d3e1 ("phy: qcom-qusb2: New driver for QUSB2 PHY on
Qcom chips") you can see a call like:

  devm_nvmem_cell_get(dev, NULL);

Note that the cell ID passed to the function is NULL.  This is because
the qcom-qusb2 driver is expected to work only on systems where the
PHY node is hooked up via device-tree and is nameless.

This works OK for the most part.  The first thing nvmem_cell_get()
does is to call of_nvmem_cell_get() and there it's documented that a
NULL name is fine.  The problem happens when the call to
of_nvmem_cell_get() returns -EINVAL.  In such a case we'll fall back
to nvmem_cell_get_from_list() and eventually might (if nvmem_cells
isn't an empty list) crash with something that looks like:

 strcmp
 nvmem_find_cell
 __nvmem_device_get
 nvmem_cell_get_from_list
 nvmem_cell_get
 devm_nvmem_cell_get
 qusb2_phy_probe

There are several different ways we could fix this problem:

One could argue that perhaps the qcom-qusb2 driver should be changed
to use of_nvmem_cell_get() which is allowed to have a NULL name.  In
that case, we'd need to add a patche to introduce
devm_of_nvmem_cell_get() since the qcom-qusb2 driver is using devm
managed resources.

One could also argue that perhaps we could just add a name to
qcom-qusb2.  That would be OK but I believe it effectively changes the
device tree bindings, so maybe it's a no-go.

In this patch I have chosen to fix the problem by simply not crashing
when a NULL cell_id is passed to nvmem_cell_get().

NOTE: that for the qcom-qusb2 driver the "nvmem-cells" property is
defined to be optional and thus it's expected to be a common case that
we would hit this crash and this is more than just a theoretical fix.

Fixes: ca04d9d3e1 ("phy: qcom-qusb2: New driver for QUSB2 PHY on Qcom chips")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-07-07 17:30:25 +02:00
Mathieu Malaterre 50808bfcc1 nvmem: properly handle returned value nvmem_reg_read
Function nvmem_reg_read can return a non zero value indicating an error.
This returned value must be read and error propagated to
nvmem_cell_prepare_write_buffer. Silence the following gcc warning (W=1):

drivers/nvmem/core.c:1093:9: warning: variable 'rc' set but
 not used [-Wunused-but-set-variable]

Signed-off-by: Mathieu Malaterre <malat@debian.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-05-14 16:20:48 +02:00
Srinivas Kandagatla b378c77934 nvmem: core: describe add missing dev function parameter
Document dev parameter which not described in devm_nvmem_unregister
and devm_nvmem_register functions.

Fix below warnings when kernel is compiled with W=1
drivers/nvmem/core.c:579: warning: Function parameter or member
 'dev' not described in 'devm_nvmem_register'
nvmem/core.c:615: warning: Function parameter or member 'dev'
 not described in 'devm_nvmem_unregister'

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-05-14 16:20:48 +02:00
Andrew Lunn b3db17e4b8 drivers: nvmem: Export nvmem_add_cells()
Not all platforms use device tree. It is useful to be able to add
cells to a NVMEM device from code. Export nvmem_add_cells() so making
this possible.

This required changing the parameters a bit, so that just the cells
and the number of cells are passed, not the whole nvmem config
structure.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-05-14 16:20:48 +02:00
Andrey Smirnov f1f50eca5f nvmem: Introduce devm_nvmem_(un)register()
Introduce devm_nvmem_register()/devm_nvmem_unregister() to make
.remove() unnecessary in trivial drivers.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: Carlo Caione <carlo@caione.org>
Cc: Kevin Hilman <khilman@baylibre.com>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Cc: cphealy@gmail.com
Cc: linux-kernel@vger.kernel.org
Cc: linux-mediatek@lists.infradead.org
Cc: linux-rockchip@lists.infradead.org
Cc: linux-amlogic@lists.infradead.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-03-14 19:28:13 +01:00
Andrey Smirnov fd0f4906a3 nvmem: core: Allow specifying device name verbatim
Add code to allow avoid having nvmem core append a numeric suffix to
the end of the name by passing config->id of -1.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: Carlo Caione <carlo@caione.org>
Cc: Kevin Hilman <khilman@baylibre.com>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Cc: cphealy@gmail.com
Cc: linux-kernel@vger.kernel.org
Cc: linux-mediatek@lists.infradead.org
Cc: linux-rockchip@lists.infradead.org
Cc: linux-amlogic@lists.infradead.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-03-14 19:28:13 +01:00
Heiner Kallweit fc2f997035 nvmem: core: switch to device_property_present for reading property "read-only"
Switch to more generic device_property_present to consider also non-DT
properties.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-12-19 10:39:33 +01:00
Heiner Kallweit 99897efd5a nvmem: core: let stride and word_size default to 1
If the caller doesn't set stride and/or word_size in struct nvmem_config
then nvmem_register accepts this but we may face strange effects later
due to both values being 0. Therefore use 1 as default for both values.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-12-19 10:39:33 +01:00
Lukas Wunner 9f3044c38c nvmem: core: Deduplicate bus_find_device() by name matching
No need to reinvent the wheel, we have bus_find_device_by_name().

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-12-19 10:39:33 +01:00
Masahiro Yamada 17eb18d674 nvmem: set nvmem->owner to nvmem->dev->driver->owner if unset
All nvmem drivers are supposed to set the owner field of struct
nvmem_config, but this matches nvmem->dev->driver->owner.

As far as I see in drivers/nvmem/ directory, all the drivers are
the case.  So, make nvmem_register() set the nvmem's owner to the
associated driver's owner unless nvmem_config sets otherwise.

Remove .owner settings in the drivers that are now redundant.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-08 13:54:17 +01:00
Greg Kroah-Hartman 58f2c391cc Merge 4.14-rc5 into char-misc-next
We need the fixes in here to resolve merge issues and for testing.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-10-16 09:22:22 +02:00
Masahiro Yamada e701c67c30 nvmem: remove unneeded IS_ENABLED(CONFIG_NVMEM) conditional
As you see in drivers/nvmem/Makefile, this C file is compiled only
when CONFIG_NVMEM is y or m.

So, IS_ENABLED(CONFIG_NVMEM) is always evaluated to 1 in this file.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-18 16:20:26 +02:00
Masahiro Yamada f7c04f16d7 nvmem: remove inline in drivers/nvmem/core.c
These two functions are defined in .c file, but called just once
(at least for now).  So, the compiler will fold them into their
callers even without the "inline" markers.

However, this kind of optimization should not be done by hand.
It is compiler's judge after all.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-18 16:20:26 +02:00
Masahiro Yamada aad8d097c9 nvmem: add missing of_node_put() in of_nvmem_cell_get()
of_get_next_parent() increments the refcount of the returned node.
It should be put when done.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-18 16:12:26 +02:00
Guy Shapiro 38b0774c05 nvmem: core: return EFBIG on out-of-range write
When writing data that exceeds the nvmem size to a nvmem sysfs file
using the sh redirection operator >, the shell hangs, trying to
write the out-of-range bytes endlessly.

Fix the problem by returning EFBIG described in man 2 write.

Similar change was done for binary sysfs files on commit
0936896056

Signed-off-by: Guy Shapiro <guy.shapiro@mobi-wize.com>
Cc: linux-api@vger.kernel.org
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-09-18 16:12:26 +02:00
Dan Carpenter fd08611317 nvmem: core: remove unneeded NULL check
"p" is the list iterator so it can't be NULL.  Static checkers complain
about this unnecessary check because we dereference the list iterator to
get the next item in the list so we'd be in trouble if it really was
NULL.  I have removed the check.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-08-28 17:33:23 +02:00
Leonard Crestez d026d70a2e nvmem: core: Add nvmem_cell_read_u32
This function does a quick and easy read of an u32 value without any
kind of resource management code on the consumer side.

Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
Reviewed-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-08-28 17:33:23 +02:00
Rob Herring 5f214ccdd1 nvmem: Convert to using %pOF instead of full_name
Now that we have a custom printf format specifier, convert users of
full_name to use %pOF instead. This is preparation to remove storing
of the full path string for each node.

Signed-off-by: Rob Herring <robh@kernel.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-08-28 17:33:22 +02:00
Heiner Kallweit 666d6a3623 nvmem: core: add locking to nvmem_find_cell
Adding entries to nvmem_cells and deleting entries from it is
protected by nvmem_cells_mutex. Therefore this mutex should
also protect iterating over the list.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-06-09 12:08:27 +02:00
Andrey Smirnov 79fbf0468b nvmem: core: Call put_device() in nvmem_unregister()
Call put_device() in nvmem_unregister() to make sure nvmem_release
gets called freeing up allocated resources.

Cc: cphealy@gmail.com
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-06-09 12:08:27 +02:00
Johan Hovold 3360acdf83 nvmem: core: fix leaks on registration errors
Make sure to deregister and release the nvmem device and underlying
memory on registration errors.

Note that the private data must be freed using put_device() once the
struct device has been initialised.

Also note that there's a related reference leak in the deregistration
function as reported by Mika Westerberg which is being fixed separately.

Fixes: b6c217ab9b ("nvmem: Add backwards compatibility support for older EEPROM drivers.")
Fixes: eace75cfdc ("nvmem: Add a simple NVMEM framework for nvmem providers")
Cc: stable <stable@vger.kernel.org>     # 4.3
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Acked-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-06-09 12:08:27 +02:00
Aban Bedel 5253193d54 nvmem: core: Allow allocating several anonymous nvmem devices
Currently the nvmem core expect the config to provide a name and ID
that are then used to create the device name. When no device name is
given 'nvmem' is used. However if there is several such anonymous
devices they all get named 'nvmem0', which doesn't work.

To fix this problem use the ID from the config only when the config
also provides a name. When no name is provided take the uinque ID of
the nvmem device instead.

Signed-off-by: Aban Bedel <albeu@free.fr>
Reviewed-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-08 17:51:48 +02:00
Vivek Gautam fd0c478c0d nvmem: core: Allow getting nvmem cell with a NULL cell id
The nvmem cell with a NULL cell name/id should be the one
with no accompanying 'nvmem-cell-names' property, and thus
will be the cell at index 0 in the device tree.
So, we default to index 0 and update the cell index only when
nvmem cell name id exists.

Suggested-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Vivek Gautam <vivek.gautam@codeaurora.org>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-01-25 11:49:39 +01:00
Vivek Gautam 29143268ec nvmem: core: Correct a bunch of function documentations
Correct the documentation for arguments to a number
of functions.

Signed-off-by: Vivek Gautam <vivek.gautam@codeaurora.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-01-25 11:49:39 +01:00
Vivek Gautam 3b4a687726 nvmem: core: Allow ignoring length when reading a cell
nvmem_cell_read() API fills in the argument 'len' with
the number of bytes read from the cell. Many users don't
care about this length value. So allow users to pass a
NULL pointer to this len field.

Signed-off-by: Vivek Gautam <vivek.gautam@codeaurora.org>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-01-25 11:49:39 +01:00
Brian Norris b577fafc43 nvmem: fix nvmem_cell_read() return type doc
nvmem_cell_read() returns void *, not char *. This is a cleanup that got
left out of commit a6c5091250 ("nvmem: Declare nvmem_cell_read()
consistently").

Signed-off-by: Brian Norris <briannorris@chromium.org>
Fixes: a6c5091250 ("nvmem: Declare nvmem_cell_read() consistently")
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-01-04 18:22:47 +01:00
Arnd Bergmann 287980e49f remove lots of IS_ERR_VALUE abuses
Most users of IS_ERR_VALUE() in the kernel are wrong, as they
pass an 'int' into a function that takes an 'unsigned long'
argument. This happens to work because the type is sign-extended
on 64-bit architectures before it gets converted into an
unsigned type.

However, anything that passes an 'unsigned short' or 'unsigned int'
argument into IS_ERR_VALUE() is guaranteed to be broken, as are
8-bit integers and types that are wider than 'unsigned long'.

Andrzej Hajda has already fixed a lot of the worst abusers that
were causing actual bugs, but it would be nice to prevent any
users that are not passing 'unsigned long' arguments.

This patch changes all users of IS_ERR_VALUE() that I could find
on 32-bit ARM randconfig builds and x86 allmodconfig. For the
moment, this doesn't change the definition of IS_ERR_VALUE()
because there are probably still architecture specific users
elsewhere.

Almost all the warnings I got are for files that are better off
using 'if (err)' or 'if (err < 0)'.
The only legitimate user I could find that we get a warning for
is the (32-bit only) freescale fman driver, so I did not remove
the IS_ERR_VALUE() there but changed the type to 'unsigned long'.
For 9pfs, I just worked around one user whose calling conventions
are so obscure that I did not dare change the behavior.

I was using this definition for testing:

 #define IS_ERR_VALUE(x) ((unsigned long*)NULL == (typeof (x)*)NULL && \
       unlikely((unsigned long long)(x) >= (unsigned long long)(typeof(x))-MAX_ERRNO))

which ends up making all 16-bit or wider types work correctly with
the most plausible interpretation of what IS_ERR_VALUE() was supposed
to return according to its users, but also causes a compile-time
warning for any users that do not pass an 'unsigned long' argument.

I suggested this approach earlier this year, but back then we ended
up deciding to just fix the users that are obviously broken. After
the initial warning that caused me to get involved in the discussion
(fs/gfs2/dir.c) showed up again in the mainline kernel, Linus
asked me to send the whole thing again.

[ Updated the 9p parts as per Al Viro  - Linus ]

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Andrzej Hajda <a.hajda@samsung.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: https://lkml.org/lkml/2016/1/7/363
Link: https://lkml.org/lkml/2016/5/27/486
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> # For nvmem part
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-27 15:26:11 -07:00
Srinivas Kandagatla 795ddd18d3 nvmem: core: remove regmap dependency
nvmem uses regmap_raw_read/write apis to read/write data from providers,
regmap raw apis stopped working with recent kernels which removed raw
accessors on mmio bus. This resulted in broken nvmem for providers
which are based on regmap mmio bus. This issue can be fixed temporarly
by moving to other regmap apis, but we might hit same issue in future.
Moving to interfaces based on read/write callbacks from providers would
be more robust.

This patch removes regmap dependency from nvmem and introduces
read/write callbacks from the providers.

Without this patch nvmem providers like qfprom based on regmap mmio
bus would not work.

Reported-by: Rajendra Nayak <rjendra@qti.qualcomm.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-05-01 14:01:00 -07:00
Andrew Lunn b6c217ab9b nvmem: Add backwards compatibility support for older EEPROM drivers.
Older drivers made an 'eeprom' file available in the /sys device
directory. Have the NVMEM core provide this to retain backwards
compatibility.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-03-01 16:55:48 -08:00
Andrew Lunn 811b0d6538 nvmem: Add flag to export NVMEM to root only
Legacy AT24, AT25 EEPROMs are exported in sys so that only root can
read the contents. The EEPROMs may contain sensitive information. Add
a flag so the provide can indicate that NVMEM should also restrict
access to root only.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-03-01 16:55:48 -08:00
Rasmus Villemoes dfdf141429 nvmem: core: fix error path in nvmem_add_cells()
The current code fails to nvmem_cell_drop(cells[0]) - even worse, if
the loop above fails already at i==0, we'll enter an essentially
infinite loop doing nvmem_cell_drop on cells[-1], cells[-2], ... which
is unlikely to end well.

Also, we're not freeing the temporary backing array cells on the error
path.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-02-11 19:23:28 -08:00
Srinivas Kandagatla 313a72ff98 nvmem: core: return error for non word aligned access
nvmem providers have restrictions on register strides, so return error
when users attempt to read/write buffers with sizes which are less
than word size.

Without this patch the userspace would continue to try as it does not
get any error from the nvmem core, resulting in a hang or endless loop
in userspace.

Reported-by: Ariel D'Alessandro <ariel@vanguardiasur.com.ar>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2016-02-07 23:07:21 -08:00
Axel Lin ace2217065 nvmem: core: Fix memory leak in nvmem_cell_write
A tmp buffer is allocated if cell->bit_offset || cell->nbits.
So the tmp buffer needs to be freed at the same condition to avoid leak.

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-10-04 12:09:43 +01:00
Axel Lin cbf854ab36 nvmem: core: Handle shift bits in-place if cell->nbits is non-zero
It's pointless to test (cell->bit_offset || cell->bit_offset).
nvmem_shift_read_buffer_in_place() should be called when
(cell->bit_offset || cell->nbits).

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-10-04 12:09:43 +01:00
ZhengShunQian 7c806883e1 nvmem: core: fix the out-of-range leak in read/write()
The position to read/write must be less than max
register size.

Signed-off-by: ZhengShunQian <zhengsq@rock-chips.com>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-10-04 12:09:43 +01:00
Srinivas Kandagatla e2a5402ec7 nvmem: Add nvmem_device based consumer apis.
This patch adds read/write apis which are based on nvmem_device. It is
common that the drivers like omap cape manager or qcom cpr driver to
access bytes directly at particular offset in the eeprom and not from
nvmem cell info in DT. These driver would need to get access to the nvmem
directly, which is what these new APIS provide.

These wrapper apis would help such users to avoid code duplication in
there drivers and also avoid them reading a big eeprom blob and parsing
it internally in there driver.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Rajendra Nayak <rnayak@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-08-05 13:43:44 -07:00
Srinivas Kandagatla 69aba7948c nvmem: Add a simple NVMEM framework for consumers
This patch adds just consumers part of the framework just to enable easy
review.

Up until now, nvmem drivers were stored in drivers/misc, where they all
had to duplicate pretty much the same code to register a sysfs file,
allow in-kernel users to access the content of the devices they were
driving, etc.

This was also a problem as far as other in-kernel users were involved,
since the solutions used were pretty much different from on driver to
another, there was a rather big abstraction leak.

This introduction of this framework aims at solving this. It also
introduces DT representation for consumer devices to go get the data they
require (MAC Addresses, SoC/Revision ID, part numbers, and so on) from
the nvmems.

Having regmap interface to this framework would give much better
abstraction for nvmems on different buses.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
[Maxime Ripard: intial version of the framework]
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Rajendra Nayak <rnayak@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-08-05 13:43:12 -07:00
Srinivas Kandagatla eace75cfdc nvmem: Add a simple NVMEM framework for nvmem providers
This patch adds just providers part of the framework just to enable easy
review.

Up until now, NVMEM drivers like eeprom were stored in drivers/misc,
where they all had to duplicate pretty much the same code to register
a sysfs file, allow in-kernel users to access the content of the devices
they were driving, etc.

This was also a problem as far as other in-kernel users were involved,
since the solutions used were pretty much different from on driver to
another, there was a rather big abstraction leak.

This introduction of this framework aims at solving this. It also
introduces DT representation for consumer devices to go get the data
they require (MAC Addresses, SoC/Revision ID, part numbers, and so on)
from the nvmems.

Having regmap interface to this framework would give much better
abstraction for nvmems on different buses.

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
[Maxime Ripard: intial version of eeprom framework]
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Rajendra Nayak <rnayak@codeaurora.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2015-08-05 13:43:12 -07:00