Aeris A300CS patches [V3]
Dirk Hohndel
dirk at hohndel.org
Wed Oct 8 03:27:51 PDT 2014
On Tue, Oct 07, 2014 at 09:09:35PM +0200, Jef Driesen wrote:
> On 02-10-14 19:49, Dirk Hohndel wrote:
> >And here is one more, this one detects how many cylinders the A300CS
> >thinks we used during a dive. This way no unused cylinders are reported
> >back to the downloading application.
>
> For patches #1 and #2, I need to check how this will interact with the other
> two oceanic backends. For example the multipage code certainly doesn't
> require aligned read. So this causes more I/O operations for those backends.
> But since the A300CS also support the older B1 command, we don't really need
> those two patches immediately to get the A300CS to work. (That gives me some
> time to also find a solution for the Aeris F11 that plays nicely together
> with your A300CS work.)
Yep.
> So I would like to already go ahead and apply the other patches:
>
> Patch #3 and #4 look good to me. As explained before, the DTR/RTS shouldn't
> cause any problems for other devices, because if it did I would already had
> received plenty of bug reports from Windows users. One minor issues is that
> I'm pretty sure the rb_profile_end field of the layout is actually 0x3FE00
> instead of 0x40000. That's why I asked you for a full memory dump.
Sorry, busy flying to Bonaire and diving.
> For patch #5, where did you get the density value 1003 from? According to
> your patch, only the water type is stored in the data, but not the density.
> Since this is optional value, maybe we better leave this as zero?
OK
> Patch #6 is the only one where I don't understand the logic. Are they really
> using this very weird encoding, or are we missing something? Am I correct
> that if only 1,2,3 or 4 tanks are enabled, these are always the first ones?
I spent an evening playing with the simulator and their app. Trying every
single value this is the logic that I found. I think it would be a nice
feature not to get the unused tanks, but my confidence that this is 100%
correct is not 100% :-/. The problem is... the only way to get this more
widely tested is to add the code and wait if it works as expected in all
circumstances.
/D
More information about the devel
mailing list