From what I can tell, these radios, there is an extra 1 byte write/ack sequence at the end of the big blob of memory in the middle.
The 8800/8900 have this similarly set. Not sure why 7900 wasnt set here.

While fixing that, I'm seriously considering refactoring 7800 class, and I really want to use yaesu_common, but it's a beast of a chore.
And I dont have 7800, 8800, or 8900 to test with - so would be afraid of breaking those ...
You want me to resend a patch with some distillation of the first sentence above?

From: Dan Smith <dsmith@danplanet.com>
To: chirp_devel@intrepid.danplanet.com
Sent: Wednesday, January 8, 2014 6:46 PM
Subject: Re: [chirp_devel] [PATCH] [ft-7900] fix upload error after clone #597

> # HG changeset patch
> # User Jens Jensen <kd4tjx@yahoo.com>
> # Date 1388972740 21600
> #      Sun Jan 05 19:45:40 2014 -0600
> # Node ID 0331c22551fc539ad794c0d90079a3507f04dc59
> # Parent  2d995594800cceebb8da5491a43bcf57c12f5164
> [ft-7900] fix upload error after clone #597


Can you add some more detail to the commit message here? It's not
obvious from looking at the diff what the problem was (aside from the
fact that _block_lengths was apparently missing).

Thanks!

--Dan