ChanServ changed the topic of #aarch64-laptops to: Linux support for AArch64 Laptops (Asus NovaGo TP370QL - HP Envy x2 - Lenovo Mixx 630 - Lenovo Yoga C630)
derzahl has quit [Ping timeout: 480 seconds]
derzahl has joined #aarch64-laptops
iivanov has joined #aarch64-laptops
<steev> bamse: i was wrong - the flex5g *does* still reboot, even with suspend disabled, however it takes a log longer for the issue to happen
<steev> robclark: turns out i'm a derp. i didn't enable the freedreno vulkan driver in my build (woo)
<robclark> oh, yeah, that'd do it
<steev> yay! WARNING: tu is not a conformant Vulkan implementation, testing use only.
<steev> robclark: so if i wanna do testing things... how to i tell it not to use llvmpipe at all? vulkaninfo shows both tu and llvmpipe implementations
<robclark> the app should enumerate all the "devices" physical and sw, and pick which one to use.. in theory it should prefer the physical device
<steev> ah, i was trying to play with that vkrunner thing and... it fail bad
<steev> to be fair though, i have absolutely no idea what i'm doing and just followed the readme
<robclark> man vkrunner needs some arguments to tell it which device to pick? idk.. I've not done too much on the vk side of things.. others in #freedreno would know better
<bamse> steev: i've come to the conclusion that it doesn't seem to be suspend related at all...just something bad happening related to wifi
<steev> bamse: oh, that might make sense
<bamse> steev: because i had a couple of hangs while i was using my laptop, and then i switched to usb ethernet dongle...and i apparently have 15 days uptime now
<steev> this is ath11k right? not ath10?
<bamse> no it's ath10k
<steev> ah
<bamse> so it's almost identical to what we have in c630
<bamse> sdm845 being wcn3990 and sm8150/sc8180x being bundled with a wcn3998
<steev> there may be some stuff chilling in net-dev or kvalo's trees that isn't in yet for 3998 then
<bamse> i don't remember what sc7180 uses...other than that i don't think we should expect anything there
<bamse> all the cool kids have ath11k-hardware now
<steev> to be fair... ath10k is kinda garbage
<steev> the firmware lies about a lot of things
<bamse> it's polished garbage, if we compare with what it was in the beginning of c630
<bamse> or sdm845...
<steev> tru
<bamse> i spent sooo much time getting wifi working on db845c
<steev> but, i also have a usb-c->usb-mini-b for external wifi cards
<steev> i keep an rtl8187 device around because even if it only hits the older stuff, it doesn't require any firmware. useful on my big endian arm64 box where i haven't figure out why it won't load firmware
<robclark> bamse: sc7180 looks like ath10k + wcn3990 (assuming I'm understanding dts and dmesg properly)
<steev> ayy, ./build/src/vkrunner --device-id 1 compiled-examples/*.shader_test = pass
<steev> although i dunno why it's 1 there, when vulkaninfo says it's 0
<steev> bamse: i just tested the "new" board-2.bin file that kvalo has in his ath10k-firmware repo... it doesn't work, doesn't seem to have the flex 5g's chip id in it
<steev> "failed to fetch board data for bus=snoc,qmi-board-id=ff,qmi-chip-id=30224
<bamse> steev: i think the problem is that the board-id is supposed to be programmed in the hardware
<bamse> steev: and ff means it isn't...and when i mentioned that i don't have a single board with !ff kvalo was unhappy
<robclark> lenovo screwed up provisioning?
<steev> wouldn't surprise me
<steev> i think all of mine are ff as well
<steev> well, all 2, i sold one of my c630 to gwolf :)
* gwolf feels summoned
<gwolf> yes! :-) And am mostly done setting it up. In some days, I predict I'll start robbing time from this channel's dwellers (-:
<bamse> robclark: yes, same as every other qualcomm customer
<bamse> gwolf: welcome to the family :)
<HdkR> and now I've got the Resident Evil quote in my head
<HdkR> Too much game testing
<robclark> bamse: presumably we got this right on the chromebooks? I'm not sure how to check..
<gwolf> bamse: thanks!
<gwolf> Say... So far, the main important thing on my mind wrt the c630 is there lack of support for video output via the USB-C port. I understand it should be internally exposed as a (pluggable) DisplayPort. Do you have any pointers for that?
<gwolf> (either for "it works, move some knobs" or "it can follow the model used in xxx xxx")
<bamse> gwolf: you can get usb-c/altmode/displayport out of one of the usb-c ports...but last time i looked at it we were missing a few different pieces
<bamse> gwolf: now i have it working on the lenovo flex 5g, so if nothing else my understanding is in a better place and i hope to find some time, some day, to figure out the exact details on c630 as well
<steev> bamse: does it require the TYPEC stuff? because we don't enable those in our defconfig
<steev> i mean, your dump does
<bamse> steev: yes it does
<gwolf> That is great news indeed! Better than I expected even
<bamse> steev: we need to extend the battery driver to cover the usb-related parts of the EC interface and then wire things up with the displayport and usb phy
<bamse> steev: and then there's something hanging off i2c under the GPU node in acpi, which worries me
<steev> in b4 the c630 actually has dual gpu
<bamse> steev: there's a box in the reference schematics which might correlate to that though...
<steev> are the reference schematics available?
<bamse> that would make things too easy
<steev> maybe that... thing... is why i2c doesn't reset properly sometimes?
<bamse> i have to double check, but i think the answer is no
<bamse> it's more likely that there's something going on surroundig the EC
<steev> that would be cool to finally get the battery into mainline
<bamse> yes
<bamse> which reminds me that i need to finish up my request to get permission to release the battery driver for the flex 5g as well
<steev> oh pls do
<steev> it would be nice to see how much battery is left
<steev> fwiw, i'll have a bunch of free time in december and january since i'm taking dec 13-jan 28 off
<steev> i just don't know how my internet access will be
<bamse> steev: until then we should have battery monitor and you will have 5g :)
<steev> oh, right, i should pay my verizon bill :P
<bamse> that sounds like a good plan, regardless :)
<steev> i don't think verizon service will be available in europe :P
<gwolf> Buy a SIM at the airport, I think it is a simpler option!
<bamse> steev: sure it will, might add some roaming charges though :)
<steev> i'm sure they would love that too
iivanov has quit [Remote host closed the connection]
<bamse> i still miss my at&t subscription i had when i worked at sony; had the unlimited data option (really unlimited) and the international data option...and for some reason they just did a logical addition between the two
<robclark> due to lack of (conference or otherwise) travel these days, I've not had much opportunity to take advantage of it.. but I'm a fan of the international roaming situation with fi
<bamse> right, haven't even had a use for the lte/5g support in the laptop :(
<steev> i thought about bringing mine with me when i went up to austin, but then i sat in the herman miller store most of the day while they sold to my coworker and had him sit in all the damn chairs to see the difference
<robclark> you coulda been sitting there surfing the interwebs ;-)
<steev> i dunno, my phone barely got a signal
<bamse> steev: as usual, if you just enable some debug prints the wifi bug goes away ;)
FizzBuzz has joined #aarch64-laptops
<steev> oh, interesting