minor airspeed sensor startup improvements

[dagar]

Still a mess, but slightly better than it was before. I tested an ms4525 and ms5525.

For the driver runner we need to do a better job handling trying to start multiple sensors. Right now the error messages are fairly generic and unhelpful. They're all external and optional.

ERROR [sdp3x_airspeed] reset failed
INFO  [sdp3x_airspeed] trying SDP3X 2
ERROR [sdp3x_airspeed] reset failed
WARN  [sdp3x_airspeed] no SDP3X airspeed sensor connected
INFO  [sdp3x_airspeed] trying SDP3X 2
WARN  [sdp3x_airspeed] no SDP3X airspeed sensor connected
WARN  [ms5525_airspeed] no MS5525 airspeed sensor connected
WARN  [ms4525_airspeed] MS4525 start failed
trying ms4525 bus 2 
WARN  [ms4525_airspeed] MS4525 start failed
WARN  [ets_airspeed] no ETS airspeed sensor connected
WARN  [ets_airspeed] no ETS airspeed sensor connected

[davids5]

Should we leave this as is to avoid the rebase when FMUv1 comes out?

[dagar]

Yes I suppose so.

[davids5]

I would think we could check one Bosch sensor first. If it is present do the only Bosch path. I do not have the HW to validate this, do you?

[dagar]

I only have a couple standard pixracers. Looking at the code again, the bmi055 gyro chip select is actually shared with the hmc, although grouping it like this shouldn't matter.

Also the external i2c bmm150 isn't actually pixracer specific, but I think it was thrown in at the same time for development/testing.

At this point it's largely cosmetic for keeping the console clean during boot (and slightly faster).

[davids5]

@dagar is this a complete picture of the shared usage or is is incomplete?

[davids5]

@dagar Other than noted for the Bosch start up the rest of this makes good sense.

Do you want to decouple that part?

[dagar]

Rebased and tested a couple different devices.

[davids5]

@dagar - I will have bosch HW in a few days- I think the easiest way to vet this is to compare the boot on V4 and V4 with bosch sensors and ensure the end result is the same as master. How does that sound?

[dagar]

Thanks, sounds good.

[LorenzMeier]

Too bad, found this only now - @dagar do you want to rebase on top of my new PR or should I?

#8005

[dagar]

I'll rebase on top of #8005 once it's in.

[dagar]

Rebased on master.

[davids5]

Boot comparison MASTER - PR
`ls /dev' list is the same

1. Ordering is different - I suspect that was intentional/
   Just two notable differences:
   a. is a odd duplicate.
   b. is a loss of information. - The motivation for printing a non matching WHOAMI (hw ID) is: Was to catch chip changes in the wild.
   

[dagar]

The extra lis3mdl is optional external i2c that wasn't started before.

[davids5]

The diff to the laste is simple and looks Ok