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)
<HdkR> Need to double check which one I got, since one one place cancelled my order
<HdkR> Looks like detachable
<robclark> HdkR: ok.. blame camera blobs.. but hopefully should get 64b soon on canary
<amstan> robclark: are they fused differently or something?
<amstan> why couldn't one load the stuff from trogdor onto strongbad?
<robclark> it would _probably_ work..
<robclark> they aren't fused differently
<robclark> but could be something refuses to start becaues of camera or something like that.. so YMMV
<robclark> not sure if you can just download an img to boot off usb or what would happen if you tried a trogdor recovery img on a strongbad
<HdkR> robclark: Weird that camera blobs block it
<clover[m]> In in San Antonio
<robclark> 32b blobs don't work in 64b camera thing.. and cros doesn't have a strong multiarch story
<HdkR> ah
<HdkR> It's okay, it won't need to have a multiarch story since Cortex is dropping 32-bit support :P
<robclark> hehe, we're just planning ahead :-P
GuildedThorn has joined #aarch64-laptops
<steev> clover[m]: on purpose?
<HdkR> robclark: Confirmed on my end, no venus in the VM on strongbad, sad times
<robclark> HdkR: hmm, venus/vk should be something else.. the main thing about 64b vs 32b crosvm is just 32b limits memory in guest
<HdkR> Seems to get 508G of VA space as expected
<robclark> so vk should be a matter of what capsets the host exposes to the guest and possibly issues w/ ancient mesa in guest
<HdkR> nah, I updated that
<HdkR> HEAD as of 30 minutes ago or something
<robclark> the amount of actual RAM mapped to guest would be limited w/ 32b crosvm since all memory in guest needs to be mapped by host vmm process
<robclark> could be some different args when launching crosvm are needed.. olv might have an idea
<HdkR> Only hit a few hundred megs physical memory used so far
<robclark> amount of RAM you can assign to guest should max out at ~3.5G w/ 32b crosvm... but that would be unrelated to what context types are exposed in guest
<HdkR> oh right, only 3.11G exposed to it atm
<HdkR> If the kernel enabled swap it would be less of a pain
<HdkR> Since then I could just slap some swap in to the VM
<robclark> ideally you only have zram swap in host (but ideally your vmm isn't 32b)
<HdkR> It's definitely having a hard time. Running Steam was capping out the memory
<HdkR> Oh interesting, pressure-vessel hates it. So no proton games
<HdkR> Not that you could run much of anything with the limited ram
<HdkR> Maybe I'll look at that more once the 32-bit memory issue is sorted
GuildedThorn has quit [Read error: Connection reset by peer]
<steev> 6.3 is gonna have the device links not linking
<HdkR> device links?
<steev> i don't understand what the deal is, or what causes it
<steev> around line 204
<bamse> steev: what do you have on your usb bus?
<steev> bamse: that was me trying to get the pinebook usb-c dock working
<steev> sometimes its finnicky, as you could see :)
<bamse> ohh, so you actually disconnect/connect all those times?
<steev> not every single one, but occasionally, yeah
<steev> sometimes it takes a bit to realize its okay
<steev> johan already knows about the issue(s) with the device links
hexdump0815 has joined #aarch64-laptops
hexdump01 has quit [Ping timeout: 480 seconds]
<steev> i have a monitor, and an nvme ssd plugged into one of the usb ports, and the ethernet port is plugged in
<steev> so
<steev> how bad is doing something like
<steev> qca_generate_nvm_name(hdev, config.fwname, sizeof(config.fwname), ver, bid == 0x8c ? 0xb8c : bid);
<steev> that would be very much nacked yeah?
<bamse> why do we need to maintain the 'b'?
<steev> that's what our nvm patch is named
<steev> without that, we get [ 7.407381] Bluetooth: hci0: qca_generate_nvm_name: nvm name is qca/hpnv21.8c
<steev> and there is no .8c
<steev> tim says the bid stuff is correct, and i'm saying if it's correct, then something is wrong with our devices
<bamse> bid is 0x8c but someone prefixed the filename with 'b'?
<bamse> didn't he also say that they do that in android?
<steev> he said that "b" firmware is android firmware
<bamse> and i presume that line isn't yours?
<bamse> i.e. it's generic...
<steev> well... i wrote that to make it load our nvm patch
<steev> i haven't pushed it anywhere, i'm just playing with his u16 stuff
<bamse> so it's conditional on this bt-chipset?
<steev> that... i don't know. i have no idea what other devices out there might have a wcn6855 in them to test what nvm patch(es) they would ask for
<steev> i dunno if the windows drivers do some weird shit to make it load the patches or what
<steev> but i also know... we can ignore it and just let it whine that hpnv21.8c doesn't exist and can't load and bluetooth works
<bamse> but is there a problem with just renaming the file?
<steev> nope, we can do that, or symlink it
<bamse> sounds easy enough ;)
<steev> i actually meant to try some of the upstream nvm patches to see if they load just fine or freak out
<bamse> i don't know what the difference is, perhaps it's something worth paying attention to...
<steev> steev@wintermute:/lib/firmware/qca$ sudo ln -s hpnv21.301 hpnv21.8c
<steev> [ 768.749444] Bluetooth: hci0: QCA Downloading qca/hpnv21.8c
<steev> [ 768.979958] Bluetooth: hci0: QCA setup on UART is completed
<steev> doesn't pitch a fit (but also, i just noticed... the files have execute permissions on them, they really shouldn't
<steev> also can't seem to actually connect
<steev> yeah, with the 302 or 301 that are in linux-firmware, if those are renamed to 8c, bluetooth won't connect or scan properly
<steev> renamed/symlinked
<bamse> that seems fine, 8c != 301...or perhaps i'm missing what you're trying to say
<steev> you aren't
<steev> i was just checking to see if symlinking those would work
<steev> since ours isn't submitted to linux-firmware :)
<bamse> instead of leaving it searching for hpnv21.8c and then someone having to rename the file from windows?
<steev> yeah basically
<steev> if one of those worked, could have just said that, instead of having to grab it outta windows (which i have, but others probably don't)
<bamse> the question to ask is, is there an actual difference between hpnv21.b8c and hpnv21.8c...
<steev> according to tim... "b" is android and !b is linux... wtf that means... i have no idea
<steev> since android is linux...
<bamse> different projects within qualcomm
<bamse> but no, it doesn't answer your question
<steev> i do notice that my speaker connects a lot faster with the nvm patch
<steev> with the nvm patch applied, it even sees my oura ring
<steev> maybe this is specific to the h2go speaker i'm using, but it always mutes the audio after connecting, which is a bit annoying
<bamse> that's the nvm from windows?
<steev> yeah
<bamse> nice
<steev> it's the hpnv21.b8c file in qcbtfmuart_hsp8280.inf_arm64_69bd85311531b34a
<bamse> maybe i should get me one of those as well then ;)
<steev> would be the combined patch (the link is wrong, this is v5 with the nvm stuff from tim's v2 melded in)
<clover[m]> steev: yeah on purpose! Just left Brasserie Mon Chou Chou
<steev> oh that's one of those places i need a picture menu to order because i can't speak what is on it
<steev> that is roughly-ish down the street though, assuming it's at the pearl
iivanov has joined #aarch64-laptops
matthias_bgg has quit [Ping timeout: 480 seconds]
matthias_bgg has joined #aarch64-laptops
shoragan has quit [Quit: quit]
shoragan has joined #aarch64-laptops
derzahl has quit [Remote host closed the connection]
minecrell has quit [Read error: Connection timed out]
minecrell has joined #aarch64-laptops
jhovold has quit [Quit: WeeChat 3.7.1]
jhovold has joined #aarch64-laptops
hightower2 has quit [Ping timeout: 480 seconds]
hexdump0815 has quit [Ping timeout: 480 seconds]
hexdump0815 has joined #aarch64-laptops
hightower2 has joined #aarch64-laptops
user_ has joined #aarch64-laptops
<user_> hi
matthias_bgg has quit [Quit: Leaving]
user_ has quit [Quit: user_]
<amstan> \o
<steev> bamse: with that pinebook usb-c dock.. it seems like, things are far more stable if i plug it in *before* plugging the usb nvme drive to it
iivanov has quit [Ping timeout: 480 seconds]
<amstan> steev: power issue perhaps?
<steev> Possibly, the thinkpad
<steev> Sorry, the thinkpad is plugged in though, it’s not like it was on battery, but who knows