00:37pmoreau: karolherbst: Mmh, looks like I either messed up something when building LLVM, or I need to patch Mesa to work with the latest: https://hastebin.com/azifuninak.rb
00:48karolherbst: seems like you messed up
00:49karolherbst: why are you building llvm anyway?
00:49karolherbst: best thing I end up doing was to use system wide llvm and just use the targeted branches of the translator
10:59pmoreau: I should probably do that. I might even have the Arch package for it installed already. --"
11:20karolherbst: pmoreau: problem in the past simply was having the translator built, but that's not a real concern right now. We might miss a few features, but afaik it never mattered so far
11:20karolherbst: and cherry-picking certain things always works
17:56karolherbst: HdkR: uff.. something is up with the motherboard... probably
17:57karolherbst: I got an OS installed though, but now it refuses to even POST
20:42njsg: swapped to another CRT (to see if I could get slightly better image quality), with this one the screen blinks when I run "xrandr" :-D
20:43imirkin_: that's expected
20:43imirkin_: not ideal, but expected.
21:23imirkin_: Lyude: wouldn't it be better to respect the "max bpc" thing instead of just sticking a limit of 8?
21:24Lyude: imirkin_: yes, but i'd rather get this fixed in stable first
21:24Lyude: also not even intel has that implemented with mst yet :P
21:24imirkin_: at some point i'm going to get back to trying to get high-bpc going
21:24Lyude: i've got a pretty decent idea of how we can do that in the future though, luckily the way I designed the atomic mst helpers will come in handy :)
21:24imirkin_: (with HDMI - no hw for DP)
21:25Lyude: imirkin_: I probably can talk to kem about working on the dp bits
21:25Lyude: ..and just in general, playing around the crc stuff earlier (will be getting back to this soon once I get some bugs out of the way) gave me a much more solid understanding of how all of this evo stuff works then I had before :)
21:26imirkin_: i think i almost sorta understand it too now
21:26imirkin_: now the error messages aren't random gobbledygook
21:26imirkin_: they're orgnanized gobbledygook :)
21:29HdkR: karolherbst: Interesting. Stuck on memory training or something?
21:43karolherbst: HdkR: no clue.. I bet it's due to the 3600, but the package of the mb is saying ryzen 3000 ready :D
21:45HdkR: karolherbst: Did you update the bios via flashback?
21:47HdkR: Originally the board didn't ship with support for Zen2, so it may not have shipped with a bios new enough
21:56HdkR: Or did it boot and install the OS then never come back? :P
22:08karolherbst: HdkR: the last one
22:08karolherbst: I had the same issue before though, but I thought it's something else
22:08karolherbst: the power button also refuses to do a force shutdown unless it gets to the POST
23:18karolherbst: HdkR: there was some dirty between the CPU pins...
23:19karolherbst: same color as the PINs itself
23:19karolherbst: now it boots...
23:21karolherbst: or.. well the BIOS did a reset as well...
23:22karolherbst: ahh, BIOS version F40
23:22karolherbst: this is the one with full zen 2 support
23:23HdkR: Glad they ship with it now
23:23karolherbst: there is a F43d though
23:23karolherbst: mhh, let's check if that's important enough
23:23karolherbst: "Improve Destiny 2 gaming compatibility" the heck :D
23:24karolherbst: AGESA 126.96.36.199 ABBA and some PCIe Gen4 disabling stuff...
23:24karolherbst: who cares
23:24HdkR: Oh yea, it works around a bug in RDRAND
23:24imirkin_: people with a Gen4 PCIe board, presumably
23:24karolherbst: well.. I have a gen4 pcie board now... as it seems
23:24imirkin_: scary that a bios update affects gaming compatibility though
23:25HdkR: It's the issue where RDRAND can fail constantly if you spin on it too fast
23:25karolherbst: but the board spec only say something about gen3
23:25karolherbst: but the bios menu has gen4 stuff
23:26HdkR: Oh yea, because it is an older board, it shouldn't have PCIe4
23:26karolherbst: that's the bug :D
23:26karolherbst: "PCIe Gen4 disabled when using a 3rd Gen Ryzen (Matisse) CPU due to this new AGES"
23:26karolherbst: oh well, let's update then
23:26karolherbst: let's see if the bios supports updating
23:26karolherbst: if not.. I just don't care
23:27HdkR: I can definitely understand AMD not wanting to ship PCIe 4 on older boards that were never originally verified to work with the higher signaling requirements
23:27karolherbst: the CPU can't do it
23:28karolherbst: or can they?
23:28HdkR: Hm? It's Zen2, it has PCIe4
23:29karolherbst: ohh, nice
23:29HdkR: The board was originally made for the Zen/Zen+ stuff which only had PCIe3 though
23:29HdkR: and AMD decided to not allow that combination for PCIe4
23:31HdkR: I believe the new lowend Zen2 specific chipset also may not have PCIe4 through the chipset?
23:31HdkR: Would need to check again, I don't remember
23:32HdkR: Theoretically that should be a pretty good lowend rig regardless :P
23:38karolherbst: now I need new thermal paste...
23:39HdkR: hah, oops. I'll add that to the list for next time as well :D
23:40karolherbst: and I need like 5x 2m ethernet cables
23:40karolherbst: ehhh.. "temp1: +16.8°C (crit = +20.8°C)"
23:40karolherbst: "Warning: AMD Processor - this hardware has not undergone upstream testing. Please consult http://wiki.centos.org/FAQ for more information" :D
23:41karolherbst: I guess that's what I get for installing centos 8
23:42karolherbst: wondering if I should go with a different distribution
23:42karolherbst: I won't use debian
23:42karolherbst: and I'd like to use something serverish which I don't have to deal with
23:47njsg: HdkR: Bug? Hey, -1 was chosen by fair dice roll!
23:48njsg: (Was it xkcd that did this joke before? I don't remember.)