The commit changes the fs_mount to not allow mounting same system,
with the same private data pointer, at two different mount paths.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
As of today <zephyr/zephyr.h> is 100% equivalent to <zephyr/kernel.h>.
This patch proposes to then include <zephyr/kernel.h> instead of
<zephyr/zephyr.h> since it is more clear that you are including the
Kernel APIs and (probably) nothing else. <zephyr/zephyr.h> sounds like a
catch-all header that may be confusing. Most applications need to
include a bunch of other things to compile, e.g. driver headers or
subsystem headers like BT, logging, etc.
The idea of a catch-all header in Zephyr is probably not feasible
anyway. Reason is that Zephyr is not a library, like it could be for
example `libpython`. Zephyr provides many utilities nowadays: a kernel,
drivers, subsystems, etc and things will likely grow. A catch-all header
would be massive, difficult to keep up-to-date. It is also likely that
an application will only build a small subset. Note that subsystem-level
headers may use a catch-all approach to make things easier, though.
NOTE: This patch is **NOT** removing the header, just removing its usage
in-tree. I'd advocate for its deprecation (add a #warning on it), but I
understand many people will have concerns.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Add a bunch of missing "zephyr/" prefixes to #include statements in
various test and test framework files.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
In order to bring consistency in-tree, migrate all tests to the new
prefix <zephyr/...>. Note that the conversion has been scripted, refer
to #45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The fs_statvfs is supposed to return, as all VFS functions, -ENOTSUP
error when underlying file-system driver does not implement the
API call.
The fs_statvfs was returning 0 for success and when API call is not
implemented, which means it is indistinguishable whether stat
structure has been filled by diver with any data or not touched at all.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
Convert remaining tests and samples to using find_package() instead of
literally including the CMake boilerplate code.
Signed-off-by: Henrik Brix Andersen <henrik@brixandersen.dk>
Move to CMake 3.20.0.
At the Toolchain WG it was decided to move to CMake 3.20.0.
The main reason for increasing CMake version is better toolchain
support.
Better toolchain support is added in the following CMake versions:
- armclang, CMake 3.15
- Intel oneAPI, CMake 3.20
- IAR, CMake 3.15 and 3.20
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>
Fixes problem when fs_opendir invoked on fs_dir_t object,
which is already holding information on opened directory,
overwrites references to other memory objects within
the fs_dir_t object causing resource leak.
If fs_opendir is invoked on already used fs_dir_t object,
it will return
-EBUSY.
Note: The change requires that all fs_dir_t objects should be
initialized to 0.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
That whether a FS implementation allow to open
same directory multiple time is FS implementation
dependent. It shouldn't be tested. Such test was removed.
However need to test whether API transfers failures from
underlying FS properly.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
The commit adds initializations of fs_dir_t variables in preparation
for fs_opendir function change that will require fs_dir_t object, passed
to the function, to be initialized before first usage.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
The fs_dir_t_init() function has been added that should be used
for initialization of fs_dir_t structures before passing them
to fs_open and other functions.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
Fixes problem when fs_open invoked on fs_file_t object, which is already
holding information on opened file, overwrites references to other
memory objects within the fs_file_t object causing resource leak.
If fs_open is invoked on already used fs_file_t object, it will return
-EBUSY.
Note: The change requires that all fs_file_t objects should be
initialized to 0.
Fixes: #29478
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The fs_file_t_init() function has been added that should be used
for initialization of fs_file_t structures before passing them
to fs_open and other functions.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The commit fixes three things:
- the order of zassert_* checks has been often incorrect, and some
checks of values have been done after some other operations has
been performed;
- some operations have been missing checks at all;
- most of zassert_* messages have been given ret parameter to be
printed but it has been missing from format string.
This addresses:
Coverity CID :215714
GH Issue: #31668
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The fs_seek and fs_tell are expected to return -ENOTUSP if file system
driver lacks implementation of said funcions.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The commit changes error handling by fs_unmount; the function will
return -EINVAL if mount point, described by mp, is not mounted or
-ENNOTSUP when unmounting is not supported by the driver; in the second
case it will also log error.
Additionally to the above changes, checks for correct mnt_path and
mnt_path, within fs_unmount, have been removed as they are not needed;
only the fs_mount_t->fs pointer is needed to decide whether system is
mounted or not.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The commit moves fs_mount parameter verification above mutex lock.
The list of mount points is now checked before attempting to obtain
file system API pointer.
All modifications to mount point data structure, given as a parameter
to the fs_mount, are only applied after every other operation needed
have completed successfully, immediately before adding the mount point
to the list of mount points.
The fs_mount will a warning when mounted file system does not support
unmount.
When a file system does not provide mount function, the -ENOTSUP error
will be returned instead of -EINVAL.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The struct fs_file_system_t is only useful when defining file system
drivers and is not required for typical application development,
that is why it has been moved to separate file fs_sys.h.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
FS_TYPE_END is no longer used as there is no hard-coded limit on the
number of file systems. Replace with the limit used when only in-tree
file systems are enabled.
The test for unsupported file systems is no longer valid. Instead
verify that attempting to register a file system when there is no room
for it fails as expected.
Signed-off-by: Peter A. Bigot <pab@pabigot.com>
Anas Nashif