No need to write deletion entry for non-existing entry.
Furthermore such a write to filled up storage will make another
writes impossible as it occupied delete nvs_ate space, but it
does not remove anything.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
During initialization NVS analyzes open sector for
find write addresses. It was possible to infinite loop
in case when there was no space for any new ATE in the sector.
This patch introduce check for space available in the sector which
fix the issue.
fixes#17891
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
NVS always leave space for deletion ate in a full sector.
Even that it was not possible to write delete nvs_ate to the
full sector. Because of that it was possible to fill up NVS,
and be able to delete nothing.
This patch introduce recognition of case
the delete ate is writing, and allow to write such nvs_ate to
opened full sector.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Previous ATE searching is accelerate by reading sector
close ate, which allow to skip reading whole storage.
ATE is already covered by crc8. That patch introduce
check on correctness offset read form close ate. This
increases meta-data integrity check level.
This also preserves against possible looping when read
incorrect or not a NVS storage image.
fixes#16899
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
There's desire to be able to customize parameters on a per-filesystem
basis, which means we need a way to override the Kconfig defaults which
are global. This also means the littlefs data structure cannot own the
cache and lookahead buffers.
Switch to using a macro to define the littlefs data structure. The
default version uses the Kconfig constants. A custom one takes
arguments providing the most likely partition-specific parameters.
Finally the user is free to bypass the helper macros and set any
parameters desired, though validation is limited and only present when
CONFIG_DEBUG is enabled.
Extend the test suite with a performance module, which confirms that
these settings have an impact proportional to the log of changes to the
cache or IO sizes.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Existing file system implementations do not provide the special "."
(current) and ".." (parent) directory entries in the readdir results.
littlefs does.
Remove these entries in the abstraction layer. This simplifies code in
higher level consumers that aren't prepared to see them. Consumers like
FUSE that need them can put them back without having to worry about
conflicts.
Closes issue #17951
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
The previous way of creating an absolute path relied on snprintf(), and
when used with newlib gcc warned that the output could be truncated
before the last format character. Rework to use code that doesn't rely
on snprintf.
See discussion at https://bugzilla.redhat.com/show_bug.cgi?id=1431678
Also ensure that cwd is always NUL-terminated, and use the utility
function to create the absolute path in cmd_trunc.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Add support for the littlefs file system in the fs shell. Update
the sample to use the same partition configuration as the littlefs
example for the SPI NOR test platform.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
littlefs is a fail-safe filesystem from ARM Mbed that has wear-leveling
capabilities.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Signed-off-by: Jim Paris <jim@bolt.io>
The sector size is 0 will pass "fs->sector_size % info.size" then start
a loop in nvs_startup() and never return. So retrun an error if the
sector size is 0.
Signed-off-by: Harry Jiang <explora26@gmail.com>
This will list all mount points in response to opendir("/").
This isn't perfect; mount points in subdirectories will show up as
their full path in this listing. But it's better than just returning
-EINVAL.
Signed-off-by: Jim Paris <jim@jtan.com>
File system API functions that operate on paths are passed both the
absolute path including the mount point prefix, and the mount point
within which the path lies.
Unfortunately it's not entirely trivial to convert an arbitrary path
within the file system space to an absolute path within its mount point,
because the path may be to the mount point itself and not end with a
directory separator. The effect is that a file system implementation
like nffs may be given an empty path when "/" is required.
Add an implementation module that does this transformation and use it to
transform paths within each filesystem wrapper.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
move misc/printk.h to sys/printk.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move misc/__assert.h to sys/__assert.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move flash.h to drivers/flash.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move crc.h to sys/crc.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move nvs/nvs.h to fs/nvs.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move fcb.h to fs/fcb.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
move fs.h to fs/fs.h and
create a shim for backward-compatibility.
No functional changes to the headers.
A warning in the shim can be controlled with CONFIG_COMPAT_INCLUDES.
Related to #16539
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
There isn't a function to handle the unmount, so I implemented based
in f_mount comments. In the comments of f_mount we actually have
that f_mount is used to mount/unmount logic, is just necessary use
NULL pointer to fs argument.
Signed-off-by: Lucas Peixoto <lucaspeixotoac@gmail.com>
This patch fixes following bug:
After first GC operation the 1st sector had become scratch
and the 2nd sector had became write sector. After that NVS
was initialize (via reboot) again - it recognized the 1st
sector as write sector and 2nd as undone GC destination sector,
therefore it cleared 2nd sector and re-run GC, which implied data loss.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Aligned format specifiers for the NVS FS. Now, the format specifier
matches the variable type for qemu_x86 types.
Signed-off-by: Kamil Piszczek <Kamil.Piszczek@nordicsemi.no>
Update the files which contain no license information with the
'Apache-2.0' SPDX license identifier. Many source files in the tree are
missing licensing information, which makes it harder for compliance
tools to determine the correct license.
By default all files without license information are under the default
license of Zephyr, which is Apache version 2.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Rename reserved function names in the subsys/ subdirectory except
for static _mod_pub_set and _mod_unbind functions in bluetooth mesh
cfg_srv.c which clash with the similarly named global functions.
Signed-off-by: Patrik Flykt <patrik.flykt@intel.com>
Remove use of select to "force" enabling other configs in subsys/fs
and subsys/net/l2. The forcing will cause infinite kconfig recursion.
Signed-off-by: Thomas Stenersen <thomas.stenersen@nordicsemi.no>
Inside cmd_read(), result of fs_seek() is not checked which might
result in reading the wrong part of file. So fails the read if
fs_seek() fails.
Fixes#13862
Fixes CID-190953
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This patch removes the free space calculation from nvs initialization.
The available space can be calculated if required using the routine
nvs_calc_free_space.
This patch also removes the locked state of nvs, it is not possible to
get in a locked state.
This patch adds an extra check on the sector_size configuration and only
allows operation on nvs when nvs has been initialized.
This patch also solves issue #13369, the usage of FLASH_ERASE_BLOCK_SIZE
has been replaced with the flash page api.
Changes:
Removed locked state and free_space from the nvs structure.
nvs_reinit(): has been replaced with by an internal only function
_nvs_startup().
nvs_write(): removed the possibility to place the file system in a
locked state, if to many gc operations are required it will return
-ENOSPC.
ssize_t nvs_calc_free_space(): introduced, calculates the free space
that is available in the nvs file system.
Removed define LOG_LEVEL.
Rebased to current master.
Signed-off-by: Laczen JMS <laczenjms@gmail.com>
This patch removes the free space calculation from nvs initialization.
The available space can be calculated if required using the routine
nvs_calc_free_space.
This patch also removes the locked state of nvs, it is not possible to
get in a locked state.
Changes:
Removed locked state and free_space from the nvs structure.
nvs_reinit(): has been replaced with by an internal only function
_nvs_startup().
nvs_write(): removed the possibility to place the file system in a
locked state, if to many gc operations are required it will return
-ENOSPC.
ssize_t nvs_calc_free_space(): introduced, calculates the free space
that is available in the nvs file system.
Signed-off-by: Laczen JMS <laczenjms@gmail.com>
It is planned to deprecate SHELL_CREATE_STATIC_SUBCMD_SET macro
which is replaced by SHELL_STATIC_SUBCMD_SET_CREATE.
Additionally, removed irrelevant comments about alphabetical
ordering which is no longer needed.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
There are issues using lowercase min and max macros when compiling a C++
application with a third-party toolchain such as GNU ARM Embedded when
using some STL headers i.e. <chrono>.
This is because there are actual C++ functions called min and max
defined in some of the STL headers and these macros interfere with them.
By changing the macros to UPPERCASE, which is consistent with almost all
other pre-processor macros this naming conflict is avoided.
All files that use these macros have been updated.
Signed-off-by: Carlos Stuart <carlosstuart1970@gmail.com>
Commit 41f86c3db2 ("nvs: fix warnings in logger") wrongly changed the
"%d" into "%x" while it was only supposed to suppress the warning.
This patch switches back the format string to "%x".
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
When compiling NVS with NEWLIB_LIBC=y, GCC outputs the following
warning:
In file included from $ZEPHYR/include/logging/log.h:11:0,
from $ZEPHYR/subsys/fs/nvs/nvs.c:17:
$ZEPHYR/subsys/fs/nvs/nvs.c: In function 'nvs_init':
$ZEPHYR/subsys/fs/nvs/nvs.c:748:10: warning: format '%lx' expects
argument of type 'long unsigned int', but argument 3 has type 'u32_t
{aka unsigned int}' [-Wformat=]
LOG_INF("alloc wra: %d, %" PRIx32 "",
^
fs->ate_wra and fs->data_wra are both defined as u32_t, so they need to
be printed with '%d'.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Following the recent NVS changes, the following warning now appear with
GCC 7.3 when building with -O2:
ZEPHYROOT/subsys/fs/nvs/nvs.c: In function 'nvs_reinit':
ZEPHYROOT/subsys/fs/nvs/nvs.c:92:36: warning: 'addr' may be used
uninitialized in this function [-Wmaybe-uninitialized]
offset += fs->sector_size * (addr >> ADDR_SECT_SHIFT);
~~~~~~^~~~~~~~~~~~~~~~~~~
ZEPHYROOT/subsys/fs/nvs/nvs.c:606:8: note: 'addr' was declared here
u32_t addr;
^~~~
This was already reported by Coverity earlier as CID:187903.
In practice this can only happen if fs->sector_count equals 0, which is
not possible as checked in nvs_init(). At least in the GCC case, it
believes that k_mutex_lock(&fs->nvs_lock, K_FOREVER) could modify
fs->sector_count.
Workaround the issue by initializing addr to 0.
Fixes#9767
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
The current NVS code checks for an empty ATE using th
_nvs_flash_cmp_const() function. This function loads the data and
compare them to a value. This means that when executed multiple on the
same area, the data get reloaded multiple time. This might have a
noticeable performance impact with an SPI flash.
Instead define a function _nvs_ate_cmp_const to compare an already read
struct nvs_ate with a constant value. Then replace the calls to
_nvs_flash_cmp_const() on struct nvs_ate by _nvs_flash_ate_rd() followed
by _nvs_ate_cmp_const(). This also has the advantage of explicitly
checking for errors instead of testing the error and the result of the
comparison at the same time.
Tested on a Nucleo L432KC board with the nvs sample. The maximum
initialization time (ie just before running the first garbage collector)
goes down to 6213 µs from 7350 µs.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
NVS with modified flash layout. At the end of a sector a special ate is
placed that points to the last ate that was written. This special ate
is written when a sector is closed. This allows nvs to travel through
the fs much quicker as it doesn't have to search for the last ate in
a sector.
This modification also speeds up the nvs_init procedure that was very
slow on external (spi) flash.
Remark: As the layout of data in flash is changed old data in the flash
cannot be recovered. It is advised to erase the nvs flash area before
using the changed nvs.
Modification after review by @nvlsianpu applied
Modification after review by @aurel32:
_nvs_prev_ate(): provide a backup search of a valid ate when the sector
close_ate has a bad CRC8. Tested on nrf81522 by making flash writing
bad data to the sector closing ate. Also validated that if a valid ate
is overwritten the filesystem keeps working.
_nvs_gc(): return error if _nvs_flash_cmp_const() is < 0.
Signed-off-by: Laczen JMS <laczenjms@gmail.com>
