Re: ✘ Release blockers?

[Fred Wright]

On Mon, 23 Dec 2019, Gary E. Miller wrote:

> How do people feel today about a release?
>
> Everything I care about is good to go.
>
> About the only possibly blocking issue I can think of is the
> gpsfake issues in some of the pipelines.
>
> Is it:
>         a) fixed
>        b) almost fixed
>        c) can't be easily fixed
>
> Any other bloking issues?

It looks like the PRN and svid are swapped for SBAS satellites.  They 
ought to match this:

        https://en.wikipedia.org/wiki/Wide_Area_Augmentation_System

It's also not clear that it's correct to regard the NMEA numbers as 
"svids" when that term usually has a different meaning, but we should at 
least get the PRNs right.

It looks like the relevant code appears in multiple places, and even the 
magic number 87 isn't parameterized:

MacPro:gpsd fw$ grep -nw ' 87' *.c
driver_garmin.c:579:            session->gpsdata.skyview[j].svid = 
(short)sats->svid + 87;
driver_nmea0183.c:1179:                *ubx_svid = 87 + nmea_satnum;
driver_nmea0183.c:1279:        *ubx_svid = 87 + nmea_satnum;
driver_superstar2.c:197:            porn = (unsigned int)(getub(buf, off + 3) 
>> 1) + 87;
driver_tsip.c:76:            *svid = prn + 87;
driver_tsip.c:88:            *svid = prn + 87;
driver_ubx.c:126:            nmea_PRN = svId - 87;
monitor_superstar2.c:50:            porn = ((unsigned char)getub(buf, off + 3) 
>> 1) + 87;

I didn't want to fix it without finding out if there are further 
ramifications.

Also, running the 3.19 cgps against the latest daemon hangs without 
acquiring any data.

On Tue, 24 Dec 2019, Hal Murray wrote:

> > The in-progress test is crazy big:
> > -rw-r--r--  1 gdt  wheel  8.1G Dec 24 08:56 /tmp/gpsd-test-XXXXXXXXXXXXXX.B7OW
> > crJ0/test-2801.chk
>
> > and then diff gets unhappy wanting 8G of ram.
>
> That's the loop writing stuff that I encountered a few days ago.

If this is reproducible, it would be good to determine:

A) Is this a daemon bug or just a test bug?

B) Is it a new bug since 3.19?

If it's yes/yes, then it's almost certainly a release blocker.  If it's 
no/yes, then a probable release blocker.  If yes/no, then worth some 
investigation and maybe a fix if it's a reasonably focused fix.  If it's 
no/no, then leave it for later.

Fred Wright