[chirp_users] Kenwood D74 MEM_FORMAT Oddity

[Dan Smith]

Hi Jacob,

This is a users list, and there are over 2000 people who are looking for usage content and not stuff like this. Please, let's take this to the chirp_devel list if you want to discuss further.

> Looking closely at the channel information starting at offset 0x4000, by my math each of the channels takes up 40 bytes. This, combined with the 16 bytes of padding for each "group" places 6 channels in one 256 byte page, preventing channels from spanning pages makes sense to me. Then, there are 210 groups/pages. This is where things get odd. 210 * 256 = 53760, plus the offset of 0x4000 places the end of this region at 70144 (0x11200), which overlaps with the "names" section that starts at offset 0x10000.
> 
> Am I missing something?

No, which is why chirp is logging this warning when loading a D74 image:

DeprecationWarning: Invalid negative seek from 0x11200 to 0x10000

It's been a while since I wrote that driver, but I think the last group is not "whole" and thus overlaps the start of names. If and since we never reference those memories, there's no problem. The #seekto afterwards repositions us in memory for the start of names. That's bad form (hence the warning) and should be fixed, but not technically a problem. I have this on my todo list to address, I just haven't gotten to it, since it's just hygiene. Feel free to open a bug if you want and I'll tag it when I do.

Again, if you want to discuss further, please subscribe to the devel list and reply to this there.

Thanks!

--Dan