<galibert[m]>
I'm curious to know what's behind the "that" :-)
<whitequark[cis]>
there was a snafu in the yosys release process (for which i am partly to blame) and the fix for it wasn't merged in time for the release
<whitequark[cis]>
galibert: i don't really want to handle the analyzer any time soon. the code is old and very bad (there are tests failling for no reason that are just disabled as a result) and doesn't perform well
<galibert[m]>
ok, it has been useful lately as a quick&dirty LA when I don't feel like connecting the saleae. I agree that it doesn't perform fell though
ari has quit [Ping timeout: 260 seconds]
<galibert[m]>
overflowing after 30ms is ridiculous
<whitequark[cis]>
the problem with building an LA is that the capture side is easy, it's the display side that's a problem
<galibert[m]>
s/fell/well/
<whitequark[cis]>
gtkwave is awful, pulseview is very marginally less awful or in some cases more, what do you do?
<galibert[m]>
You mean we should document applets when they are converted if they weren't before?
<galibert[m]>
Or just add them to the list in the doc if it's missing?
<whitequark[cis]>
v1 applets can't be added to sphinx docs
<whitequark[cis]>
v2 applets can be. so when you convert an applet to v2, you should add it to the manual
<galibert[m]>
makes sense, wouldn't have guessed though :-)
<whitequark[cis]>
for sure, that's why i'm saying that it was my responsibility to point it out in review or beforehand
<galibert[m]>
no problem, I'll know for the next ones
<whitequark[cis]>
thanks
<_whitenotifier-4>
[glasgow] github-merge-queue[bot] created branch gh-readonly-queue/main/pr-930-321f0a118ab646f53b1a35165b2806e404d0df9b - https://github.com/GlasgowEmbedded/glasgow
<whitequark[cis]>
anyway, i think something else worth pointing out is that while i usually do some housekeeping work on V2 applets to bring them closer in line to what i'd expect from a new applet today (and to serve as an example), i have no expectation of you doing the same work; for several reasons main of which is that there's no guidance on writing an applet so it's not like you would know what i expect
<whitequark[cis]>
(also, i'm in the process of figuring out what the guidance should be)
<galibert[m]>
yeah, I get that. It's obvious you're not yet 100% sure where things should go exactly, which is perfectly normal at this point. But the conversions can become tedious, so advancing them may help you working on those parts
<whitequark[cis]>
the difficult part of conversions for me is something you would probably find unexpected
<whitequark[cis]>
which is "standing up and going no more than 15m to where the device is and then going back to the laptop"
<galibert[m]>
I wouldn't have expected the 15m
<whitequark[cis]>
well sometimes it's upstairs
<whitequark[cis]>
usually it's more like 2m
<galibert[m]>
My glasgow is less than 20cm from the keyboard :-)
<whitequark[cis]>
i have 5 on my desk right now
<whitequark[cis]>
but the DUTs are not
<galibert[m]>
I have port A connected to the internals of a mu100b, and port b to a i2c 128x64 screen I haven't yet managed to turn on (but which does reply in i2c)