The issue is related to the non-sequential way of description when
such fields going together sequential.
Related to esp32h2 chip for eFuses: MAC_FACTORY and MAC_EXT.
The issue is in wrong indexes of MAC_EXT.
MAC_EXT got indexes like it is joined to MAC_FACTORY.
const esp_efuse_desc_t* ESP_EFUSE_MAC_FACTORY[] = {
&MAC_FACTORY[0],
&MAC_FACTORY[1],
&MAC_FACTORY[2],
&MAC_FACTORY[3],
&MAC_FACTORY[4],
&MAC_FACTORY[5],
NULL
};
const esp_efuse_desc_t* ESP_EFUSE_MAC_EXT[] = {
&MAC_EXT[6],
&MAC_EXT[7],
NULL
};
This commit fixed it to:
const esp_efuse_desc_t* ESP_EFUSE_MAC_EXT[] = {
&MAC_EXT[0],
&MAC_EXT[1],
NULL
};
Supported a new format of efuse description using '.' in the name.
It means that RD_DIS.KEYx belongs to the range of the RD_DIS name.
RD_DIS, EFUSE_BLK0, 32, 7, Read protection
RD_DIS.KEY0, EFUSE_BLK0, 32, 1, Read protection for EFUSE_BLK4.
RD_DIS.KEY1, EFUSE_BLK0, 33, 1, Read protection for EFUSE_BLK5.
Updated:
- CSV eFuse table
- Kconfig
- efuse_table_gen.py
- UTs
- espefuse.py
efuse/c3: Fix the burn order of BLOCKs (MAX->0)
efuse/c3: Support efuse API with keys, protection, purpose bits
tool/ci: Adds test_efuse_table_on_host for esp32s3 and esp32c3
Updated:
- CI test_esp32s2beta_efuse_table_on_host.
- efuse_table_gen.py.
- esp_efuse_table.csv file and generated headers files.
- splitted esp32 and esp32s2beta parts.
- unit tests and api efuse.
laokaiyao