00:12 <set_> Man...I pulled out the CAN TX/RX and I was ready. ha.

00:12 <set_> Prewired of course.

00:33 <set_> Um...I think I need 40,000 photos of cats and dogs to train a neural synapse.

00:33 <set_> Damn.

00:33 <set_> They are thinkin' of millions of photos. Where would one collect so many photos of one thing or actually two things that are similar but ultimately different?

00:34 <set_> People online started charging.

00:34 <set_> Oh well, no time now. Time to listen to AI tell me what I think.

00:38 <set_> How can a chip be 512-bit? Wait...did I miss somethin'?

00:38 <set_> I thought 128-bit was a big thing for the future.

00:41 <set_> oh. Vector inst. SIMD.

00:41 <set_> <<< knows nothing now

01:11 <set_> So...it seems that the vector math is 0 to 1 in an array.

01:11 <set_> i.e. 0.112 and so on.

02:35 <Guest15> @zmatt how did you create that git graph? Been looking for how to do that on the command line.

02:54 <Guest15> @zmatt I'm guessing all the changes are hidden in this commit: d783299 Update jacinto6evm target to Mesa3D based frontend UM libraries.

02:58 <Guest15> and with no visibility into what actually happened there, I think. I'm guessing that any of the branches prior to that commit will work in a similar manner to the 1.14.x. I'm guessing the only branch worth testing is remotes/origin/release/ti-img-sgx/thud/2019.04+

03:00 <Guest15> @zmatt while I'm thinking about it, could the changes I've discovered be packaged up? Specifically the Qt environment variables (into /etc/profiles.d/qt_env.sh) along with that kms config file put somewhere else?

05:59 <zmatt> Guest15: that doesn't seem appropriate since that configuration is specific to your situation, you don't generally want to force that particular qt configuration let alone the video mode

05:59 <zmatt> a write-up would be the best place

06:00 <zmatt> the best longer-term solution would be for someone to figure out why qt5's autoconfiguration doesn't work correctly here (and how to fix it)

06:01 <zmatt> Guest15: btw, have you considered setting a name that's not "Guest15" ? :)

06:21 <zmatt> Guest15: and I created that graph with: git log --oneline --graph --date-order --remotes='ti/*1.17.*'

06:21 <zmatt> (replace "ti" by the name of the remote, i.e. "origin" by default)