- As the PMA entry that made some memory regions cacheable was
assigned the highest priority, some intermediate inaccessible
memory regions bypassed protection.
- Added tests for the same
- Verified that even after changing the priority of the PMA entry,
a write operation at SOC_IRAM_LOW + 0x40 (a random RAM cached address)
still needs the same number (29) of CPU cycles.
Fixed memory leak in emac_esp_new_dma function.
Polished ESP EMAC cache management.
Added emac_periph definitions based on SoC features and improved(generalized) ESP EMAC GPIO
initialization.
Added ESP EMAC GPIO reservation.
Added check for frame error condition indicated by EMAC DMA and created a target test.
The implicit promise of heap_alloc_caps() and friends is that the memory it
returns is fit for the purpose as requested in the caps field. Before
this commit, that did not happen; e.g. DMA-capable memory wass returned
from a correct region, but not aligned/sized to something the DMA subsystem
can handle.
This commit adds an API to the esp_mm component that is then used by the
heap component to adjust allocation alignment, caps and size dependent on
the hardware requirement of the requested allocation caps.
1. Fix wrong reason code in 'WIFI_EVENT_AP_STADISCONNECTED' event
2. cleanup in softAP for disconnecting connected station
3. Update examples to display reason while processing WIFI_EVENT_AP_STADISCONNECTED event
This adds a simple test that tries to run idf_size.py help and check
if the process does not exit with error. This is just to make sure
that idf_size.py can be used with minimum required python version.
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
Previous 6caa4a17ac ("fix: display correct help in the idf_size.py wrapper")
introduced a regression, because it uses exit_on_error parameter for
argparse.ArgumentParser, which was added in python3.9, making
idf_size.py incompatible with idf.py minimal required python3.8.
The objective is to inspect the arguments of idf_size.py using a wrapper
argparse to determine whether the legacy or refactored version should be
initiated, while always displaying help for the underlying version. The
exit_on_error function was previously utilized to prevent argparse from
exiting and displaying help/usage. This replaces exit_on_error with a
workaround that makes the --format argument optional. Since this is the
sole instance where the wrapper argparse might fail, it achieves the
same outcome as using exit_on_error.
Fixes: 6caa4a17ac ("fix: display correct help in the idf_size.py wrapper")
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
Currently the wrapper tries to figure out which version of
the esp-idf-size should be started. The legacy version is
used if explicitly requested by the -l/--legacy option or
if json format is specified. This works fine, but if help
is requested, it is printed for the wrapper as shown bellow
$ idf_size.py -h
usage: idf_size.py [-h] [--format FORMAT] [-l]
options:
-h, --help show this help message and exit
--format FORMAT
-l, --legacy
This is not convenient and the full help from the underlying
version should be displayed.
Fix this by only peeking into the args to figure out if legacy or
refactored version should be started and always spawn the underlying
esp_idf_size python module. This is done by using exit_on_error=False and
add_help=False for the ArgumentParser. When help for refactored version
is requested a note as following is printed to notify users that the
legacy version can still be used.
$ idf_size.py -h
Note: legacy esp_idf_size version can be invoked by specifying the -l/--legacy
option or by setting the ESP_IDF_SIZE_LEGACY environment variable.
Signed-off-by: Frantisek Hrbata <frantisek.hrbata@espressif.com>
LP core was unable to boot when system was in deepsleep.
This was due to lp uart init in LP rom using XTAL as clk source,
which is normally powered down during sleep.
This would cause lp uart to get stuck while printing ROM output,
and the app would never boot.
Also fixed wrong wakeup cause used by HP core for ULP wake up
