Gedare Bloom commented: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5057#note_108974


As this is a protocol, it would help to describe it in terms of a grammar 
and/or a state machine. In addition, the rationale for the field sizes should 
be explained, and what that means. The reasoning for using variable versus 
fixed framing would be helpful. The simplicity of this protocol seems to lend 
itself well to having a few well-defined packet formats. Here there are 6 
different kinds of messages, with 5 different formats. It seems this could be 
made more regular to simplify processing.

It's not intuitive to me what a 12-bit base64url encoded value even is, are you 
truncating/compressing them somehow? Or, is it the 12-bit number gets base64url 
encoded (into 3 UTF-8 characters ==> 24 bits in the bitstream)? It would be 
helpful to see the description of the packets in terms of their actual contents 
(e.g., 3 ASCII characters in base64url encoded, 11 ASCII characters, etc.) 
instead of their underlying data types (12-bit number, 64-bit signal value).

Do signals and channels have semantic meaning in the protocol? If so, please 
explain what those are. If not, why not just define a generic way of passing 
data?

-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/issues/5057#note_108974
You're receiving this email because of your account on gitlab.rtems.org.


_______________________________________________
bugs mailing list
[email protected]
http://lists.rtems.org/mailman/listinfo/bugs

Reply via email to