Add specification for DLT_LIN

[kayoub5]

No description provided.

[guyharris]

So do you know whether this matches what Hannes Kaelber ([email protected], at the time he requested DLT_LIN) wanted when he requested DLT_LIN in this message?

[guyharris]

So do you know whether this matches what Hannes Kaelber ([email protected], at the time he requested DLT_LIN) wanted when he requested DLT_LIN in this message?

Can I assume from "This document is based on email conversation with Hannes Kaelber" that the answer is "yes"?

[guyharris]

So that translates to this sequence of octets:

      8     7     6     5        |      4     3     2     1       
+--------------------------------+--------------------------------+
|                         msg_format_rev                          |
+-----------------------------------------------------------------+
|                            reserved                             |
+-----------------------------------------------------------------+
|                            reserved                             |
+-----------------------------------------------------------------+
|                            reserved                             |
+---------------------------------+-------------------------------+
|         payload_length          | message_type  | checksum_type |
+---------------------------------+---------------+---------------+
|                               PID                               |
+-----------------------------------------------------------------+
|                            checksum                             |
+-----------------------------------------------------------------+
|                             errors                              |
+-----------------------------------------------------------------+
|                             payload                             |
.                                                                 .
.                                                                 .
.                                                                 .

with the payload being an ISO 17987-2 N_PDU, as described in section 7.4 "Transport layer protocol data units", if message_type is 0 and with it being a 4-octet LIN Event type if message_type is 3?

Is the LIN Event type big-endian or little-endian?

[guyharris]

What is the interpretation of the msg_format_rev field (a single integral value?), what values are supported, and what value corresponds to this format?

For the checksum_type field, what does "unknown/error" indicate? Presumably the "classic" and "enhanced" checksum are as described in section 5.2.2.7 "Checksum" of ISO 17987-3:2016.

Presumably the PID field is the protected identifier field preceding the first data field of the frame, as described in section 5.2.2.5 "PID field" of ISO 17987-3:2016.

What types of error do the flags in the error field indicate?:

• what is an overflow error?
• a checksum error is presumably an invalid checksum
• a parity error is presumably an error in the parity of the protected identifier field
• is a framing error a low-level transmissions/reception error?
• is a no slave response error a master request frame not getting a slave response frame?

[kayoub5]

@guyharris Yes, this document as a copy of an email from Hannes Kaelber (the email is a translation of a German documentation he used), as far as your questions goes, I will have to re-contact him. Would you like that I put you email in CC?

Few topics I can answer:

Presumably the PID field is the protected identifier field preceding the first data field of the frame, as described in section 5.2.2.5 "PID field" of ISO 17987-3:2016.

yes

a checksum error is presumably an invalid checksum

yes

a parity error is presumably an error in the parity of the protected identifier field

yes

is a no slave response error a master request frame not getting a slave response frame?

In LIN a single Frame is co-written by the master and slave (the master writes the "header" on the bus, the slave writes the "body" on the bus (payload + CRC)), It is not separate Frames for request and response
A no slave response means that no slave wrote the payload + CRC part of the Frame.

[guyharris]

Would you like that I put you email in CC?

Yes, please.

[kayoub5]

@guyharris I don't see an email address on your profile

[kayoub5]

@guyharris Hannes Kaelber is in vacation until mid April, I will re-contact him when he is back.

[kayoub5]

@guyharris document updated with Hannes Kälber feedback, good to merge?

[infrastation]

Please rebase on the current master branch if you can.

[kayoub5]

@infrastation done

[mcr]

I guess I'm surprised that we need new CSS, but hey if it makes this prettier... great!

[kayoub5]

Anything pending for this PR to be merged?