Age | Commit message (Collapse) | Author |
|
|
|
|
|
People have asked for this at various times.
|
|
And add a testcase for packet encapsulation to ensure we don't
have further endian issues in the future.
|
|
Should be pointing to the CRC location in the *unescaped* packet
buffer since the CRC is subject to escaping/unescaping. Previous
code pointed to the wrong location in the escaped packet buffer,
which was often pointing to the write place if there weren't many
escaped bytes in the input buffer, but was still wrong.
|
|
Rename and document the encapsulate/decapsulate functions, CRC-check
the incoming packet, and make callers aware of the difference in how
big the decapsulated packet is versus how many bytes they should
discard from the buffer (since the decapsulated packet is at least
3 bytes shorter than the incoming packet due to the CRC + framing).
|
|
|
|
Bits for CRC calculation and frame escaping/unescaping.
|