- In case of AXI-DMA, the DMA descriptors need to be 8 bytes aligned
lldesc_t do not satify this condition thus we need to replace it with
dma_descriptor_t (align(4) and align(8)) in esp_crypto_shared_gdma.
- Added new shared gdma start API that supports the dma_descriptor_t
DMA descriptor.
- Added some generic dma descriptor macros and helper functions
- replace lldesc_t with dma_descriptor_t
- Even if the config MBEDTLS_HARDWARE_AES is enabled, we now support fallback
to software implementation of GCM operations when non-AES ciphers are used.
- Add ECDSA peripheral chapter and instructions to program efuse key block
- Update security guide for ECDSA peripheral mention for device identity
- Link with ESP-TLS guide about using ECDSA peripheral in TLS connection
For ESP32-H2 case, the hardware k mode is always enforced through
efuse settings (done in startup code).
For ESP32-P4 case, the software k mode is not supported in the peripheral
itself and code was redundant.
Set max TLS version in the SSL context during setup phase. Dynamic
buffer feature overrides the `mbedtls_ssl_setup` API and hence
this change is required per upstream 3.5.0 codebase change.
Add changes to use fast_pbkdf2 as default for PMK calculations.
fast_pbkdf2 is significantly faster than current implementations
for esp chips.
Also removes unnecessary code for pbkdf-sha256 and pbkdf-sha512.
Updates config to define the new MBEDTLS_SSL_CID_TLS1_3_PAD_GRANULARITY
option, which replaced the previously used
MBEDTLS_SSL_CID_PADDING_GRANULARITY. The old option is continuing to be
used as the new one exceeds the maximum length for an option name in
esp-idf.
See https://github.com/Mbed-TLS/mbedtls/pull/4490 for more information.
Signed-off-by: Daniel Mangum <georgedanielmangum@gmail.com>
For certain data lengths, the last input descriptor was not getting appended
correctly and hence the EOF flag in the DMA descriptor link list was
set at incorrect location. This was resulting in the peripheral being
stalled expecting more data and eventually the code used to timeout
waiting for the AES completion interrupt.
Required configs for this issue:
CONFIG_MBEDTLS_HARDWARE_AES
CONFIG_SOC_AES_SUPPORT_DMA
This observation is similar to the issue reported in:
https://github.com/espressif/esp-idf/issues/10647
To recreate this issue, start the AES-GCM DMA operation with data length
12280 bytes and this should stall the operation forever.
In this fix, we are tracing the entire descriptor list and then appending the
extra bytes descriptor at correct position (as the last node).
DMA operation completion must wait until the last DMA descriptor
ownership has been changed to hardware, that is hardware is completed
the write operation for entire data. Earlier for the hardware GCM case,
the first DMA descriptor was checked and it could have resulted in some
race condition for non interrupt (MBEDTLS_AES_USE_INTERRUPT disabled) case.
SHA hardware DMA mode calculation had off-by-one error for specific
input lengths. This was causing last chunk of the input data not being
fed to the hardware accelerator and hence resulting in an incorrect
final result.
Closes: https://github.com/espressif/esp-idf/issues/11915
