Don't fail if the device sends leftover bytes before ACK/NACK #40
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
As discussed in JelmerT#13, if the running firmware prints out characters over UART while the port is closed, those characters will get buffered. Synch will then fail due to those buffered characters getting sent to the host before the ACK/NACK sequence. Under those conditions, the ACK/NACK does eventually appear on the serial line, but only after a random number of "junk" characters. Because `_wait_for_ack()` currently blindly reads 2 bytes, it will fail to detect the eventual ACK/NACK and it will thus signal an error. This commit changes `_wait_for_ack()` to keep reading byte-by-byte until it either encounters the ACK/NACK sequence or it times-out, instead of simply reading 2 bytes and failing under the aforementioned conditions.
g-oikonomou
force-pushed
the
bugfix/ack-after-synch
branch
from
f4285a7
to
3265353
Compare
|
Hi, checked with our platform (CC2538 + FTDI FT234XD). Solves the problem for us! |
|
@JelmerT just a gentle /cough on this and the other open pulls. Have you had a chance to have a look? |
|
I haven't pulled this one yet since I wasn't able to reproduce the problem. Not with the smartrf06 or a PL2303 (0 additional bytes). I'm going to retry with another FTDI tomorrow and report back. Even if I can't properly reproduce (it depends on OS and driver versions after all), I'll pull it anyway since this PR also didn't break anything for me. |
|
As mentioned before, I can't seem to receive any 'additional' bytes on my system. So this is 👍 for me. |
JelmerT
added a commit
that referenced
this pull request
Nov 7, 2015
Don't fail if the device sends leftover bytes before ACK/NACK
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As discussed in #13, if the device is running a firmware that prints characters over the UART, those characters will on occasion get buffered. When the bootloader receives the synch sequence (
0x55, 0x55), those buffered bytes will appear on the host's serial line before the ACK/NACK sequence.The current version of the script reads 2 bytes when expecting ACK/NACK, and it will therefore fail in the conditions discussed above (since the first 2 bytes received are buffered leftovers instead of an ACK/NACK).
This pull proposes changing
_wait_for_ackto read byte-by-byte until it encounters ACK/NACK or until it times out.To reproduce, program the device with a firmware that prints stuff: Contiki's startup sequence will do. Reset the device manually a couple of times and then perform the key press sequence to enter the bootloader. Run the script and it will likely fail under the above conditions. Then repeat the above with this patch in place and run with
-V. Observe the bold line below:whereas normally you should see:
A known downside of this approach is that, if the script runs while the firmware is running, it may erroneously interpret a normal printout as an ACK/NACK. However, this is unlikely because it can only ever happen if the ACK/NACK sequence gets encountered randomly during the
_wait_for_ack()timeout window.Tested with Zolertia RE-Mote (CC2538 + FT2232 (I think!)) and SmartRF06 + CC2650EM. Tested with 2.7.10 and 3.4.3.
Closes #13