<azonenberg>
miek: hey, so what do you think of renaming the linux-gpib transport to just "gpib"?
<azonenberg>
we can document that it only works on linux
<azonenberg>
my thought being, we want scopesessions to be portable
<azonenberg>
so if in the future you get a windows machine and put it on the same network
<azonenberg>
and we build a windows gpib transport
<azonenberg>
we would want it to be able to open the same capture and reconnect
<azonenberg>
(I havent properly reviewed the code yet, this is just my first thought on seeing the PR)
<miek>
yeah, can do. i named it like that thinking there would be other gpib libraries but yeah, we can probably abstract that away
GenTooMan has quit [Ping timeout: 255 seconds]
GenTooMan has joined #scopehal
<azonenberg>
well ok let me rephrase
<azonenberg>
miek: it makes sense for the string driver name to be "gpib"
<azonenberg>
you can call the class what you want
<azonenberg>
in particular, it would be fine to have linux and windows driver classes using the same driver name
<azonenberg>
as long as you only ever register one or the other with an ifdef
<azonenberg>
the implication being that they're platform specific code, but at the abstract interface level are drop-in equivalents
<miek>
ah, right. yeah, that makes sense
<azonenberg>
Basically i just want to avoid the potential of adding gpib supoprt to windows later on and having files from one not work on the other to reconnect to the same instrument
<azonenberg>
in a mixed OS lab environment
<azonenberg>
anyway so if you fix that, i'll happily merge
<azonenberg>
mxshift: looks like the QSGMII data is 100mbit speed
<azonenberg>
I have a slicer from QSGMII to 4x SGMII working now, but right now i only have a decoder for 1000baseX. So it can decode SGMII in gig mode but not 10/100
<azonenberg>
that will be next up, shouldn't take long to implement
<mxshift>
Yeah, were using QSGMII as to bridge together two switches that are running SGMII out to 3 port switches that ultimately connect to stm32h753 RMII so it's all 100Mbps
<azonenberg>
Ok perfect. just double checking
<azonenberg>
also hmm i have a bug..
<azonenberg>
oh boy ok this is going to turn into a messy refactoring
<azonenberg>
(tl;dr we have an API that doesnt support multiple protocol streams per filter that has to get cleaned up)
<_whitenotifier-7>
[scopehal] azonenberg 538a6aa - Refactoring: GetColor() and GetText() now take size_t index instead of int, and also take a stream argument