tests-filesystem-general_filesystem declares BlockDevice and FileSystem
pointers as globals. If these are initialized to their respective
default_instance values in the declaration, the LFS init happens during
OS startup when __libc_init_array() is invoked by mbed_toolchain_init().
If the filesystem blockdevice does not currently contain a valid filesystem
(e.g. the chip has just been erased), then LFS will flag this as corruption
and abort the mounting process.
This cleanup process can take long enough (and is running pre-main) that
greentea times out waiting for the device to respond to its sync packet,
and resets the device.
To resolve this, move the initialization into the first test case
(bd_init_fs_reformat) right before it initializes and formats the
blockdevice and filesystem.
ARM microlib is not supported below features
- fflush(NULL) return `-1` instead `0`
- fread with size parameter "Zero".
- if file opened in append mode, file postion starts from the beginning.
In kvstore_init, prior to initializing the kvstore, erase the
underlying block storage device. This ensures that each test run
starts from a consistent state and avoids failures that can result
if a previous test run left the storage in an inconsistent state.
Previously Greentea tests was not initialising its storage
before asking for bd->get_program_size(), causing FlashBlockDevice to
return zero. This caused both TDBStorage's to use zero for both
parameter to SlicingBlockDevice(bd, 0, 0), effetivaly both then
used same addresses for slice. This caused SecureStore tests
to fail, because writes to internal RBP storage overwrote keys
from external storage.
Fine-tune TDBStore sizes, so that all tests can fit into storage.
At least with LPC55S69's default TDBStore configuration it's
impossible to run storage Greentea tests without exhausting the
memory reserved for storing keys.
Fixes an issue where number of keys were removed based on number of
threads which didn't have anything to do with the test case.
Fixes an issue where number of keys were assumed to be constant
but variable number was used for configuration.
The addition of trace logging during greentea tests pushes the multithreaded
read-write test beyond the limits of the stack it allocates for its threads.
The increase of 128 bytes was chosen by experimentation.
As of 722628be02, the "remainder" configuration
also uses the default location near the end of flash. Which makes the two tests
nearly identical with the exception that the "last two sectors" test correctly
handles parts with a low (possibly 1:1) erase size to program size ratio.
Therefore, change the "remainder" test to instead be a "default" test that uses
the tdb_internal_address/size values, so that it
a.) tests something meaningfully different and
b.) tests using the custom TDB address/size values if they are provided.
c.) functions correctly on devices where the default sector-based size computation
does not work (e.g. because of the low erase size to program size ratio)
and therefore a custom location and size has been specified.
The is_conf_tdb_internal variable is unused and therefore removed.
The QSPI spec allows alt to be any size that is a multiple of the
number of data lines. For example, Micron's N25Q128A uses only a
single alt cycle for all read modes (1, 2, or 4 bits depending on
how many data lines are in use).
Thus far the default position has been after the application plus two
spare sectors. For simplicity and to have a predictable location for the
TDBStore with the default configuration the location is now switched to
the end of the flash. Two last sectors to be exact.
Macro which restricted compilation to GCC_ARM is removed.
Existing read_write() test is amended to call stat() and check that correct size is returned.
toyowata
Martin Kojtal
Kyle Kearney
Wilfried Chauveau
MarceloSalazar
Tymoteusz Bloch
Veijo Pesonen
Marcin Tomczyk
Seppo Takalo
Rajkumar Kanagaraj
Evelyne Donnaes
Matthew Macovsky
Chun-Chieh Li
Michal Paszta
int_szyk
jeromecoutant
Qinghao Shi