<vinifr>
I guess LRADC does not need a source clock, due to low frequency operation
bamvor has joined #linux-sunxi
jeremb has joined #linux-sunxi
akaizen has joined #linux-sunxi
akaizen_ has quit [Read error: Connection reset by peer]
vinifr has quit [Remote host closed the connection]
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
rz2k has joined #linux-sunxi
deasy has joined #linux-sunxi
hipboi has joined #linux-sunxi
Soru has quit [Quit: No Ping reply in 180 seconds.]
Soru has joined #linux-sunxi
deasy has quit [Quit: Quitte]
stekern has quit [Ping timeout: 264 seconds]
stekern has joined #linux-sunxi
deasy has joined #linux-sunxi
ZetaNeta has quit [Ping timeout: 268 seconds]
deasy has quit [Ping timeout: 245 seconds]
drachensun has quit [Ping timeout: 264 seconds]
drachensun has joined #linux-sunxi
deasy has joined #linux-sunxi
bamvor has quit [Quit: Leaving]
Colani1200 has joined #linux-sunxi
panda84kde has joined #linux-sunxi
atiti has joined #linux-sunxi
Colani1200 has quit [Read error: Connection reset by peer]
Colani1200 has joined #linux-sunxi
drachensun has quit [Ping timeout: 246 seconds]
rellla has joined #linux-sunxi
bsdfox\ has quit [Ping timeout: 264 seconds]
drachensun has joined #linux-sunxi
bsdfox\ has joined #linux-sunxi
Colani1200 has left #linux-sunxi [#linux-sunxi]
geecko has joined #linux-sunxi
rings_IIV has quit []
Ehsand has joined #linux-sunxi
Ehsand has quit [Remote host closed the connection]
Ehsand has joined #linux-sunxi
navlrac has quit [Ping timeout: 264 seconds]
foubarre has joined #linux-sunxi
<foubarre>
hello. Anyone knows the maximum resolution mali400 can display?
<foubarre>
or mali400MP. I would like to display 2160P movies; gpu can decode it but no info about displaying it.
bamvor has joined #linux-sunxi
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
libv has quit [Quit: leaving.]
bsdfox\ has quit [Ping timeout: 245 seconds]
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
bsdfox\ has joined #linux-sunxi
atsampson has joined #linux-sunxi
libv has joined #linux-sunxi
rah has joined #linux-sunxi
<rah>
your mailing list is on Google Groups
<rah>
ffs people
<rah>
Google Groups is the spawn of the devil
<rah>
don't you know this?
<jinzo>
:)
<libv>
rah: what we do is open anyway.
<libv>
rah: google groups is quite reliable and easy to set up and manage
<rz2k>
rah: there's long explanation by mnemoc somewhere around the ML about why we use google groups. tl;dr: no public mailing list service wanted to support us, and please, dont even start trying to move us to somewhere else.
<libv>
rah: there is simply no other service which easily provides just mailing lists
<rah>
libv: and also requires agreeing to Google's terms and services
<mnemoc>
you can subscribe by sending an email, no need to have a G account or accept any terms of services
<libv>
rah: it's not the first time that people have baselessly complained about this
<rah>
those terms and services can be summarised like so: I hereby give you permission to invade my privacy, up to and including sitting outside the window of my bedroom at night when my partner and I get busy, and rifling through my rubbish, and compiling data about my online behaviour, and reading my email, and selling information about me, etc, etc, etc
<rah>
mnemoc: are you sure about that?
<mnemoc>
yes
<rah>
mnemoc: how sure are you about that?
<mnemoc>
100%
<mnemoc>
Turl is subscribed that way for example
<libv>
rah: what search engine are you using?
<mnemoc>
so I can't set the moron mail admin :<
<libv>
rah: what phone do you own?
<mnemoc>
but he can post and receive mails
FR^2 has joined #linux-sunxi
<rah>
libv: I use Startpage for searching the web (sometimes DuckDuckGo)
<rah>
libv: I own a GTA04 phone
<libv>
ok, congrats for being consistent
<rah>
I'm not looking for congratulations
<rah>
I don't want congratulations
<rah>
I want to not have to use Google services
<libv>
rah: try things as mnemoc suggested, he is responsible for most things sunxi, he has heard all of this before
<rah>
I'm going to try to do that
<rah>
but only because I'm doing paid work
<rah>
if it turns out I have to agree to terms and conditions (which I'm pretty sure will be the case) then.. something
<libv>
rah: feel free to create a service which provides free mailing-lists for proper open source projects, no matter what type of project it is
<libv>
seems like an easy problem to solve, but it currently is not being done.
<libv>
even i am using a mailinglist on a server of a former colleague, as the usual tosser at fd.o refused (like he did when we asked him for an ml for radeonhd 6 years ago)
<rah>
Google no longer allow subscription or unsubscription to their Groups services using email alone
<rah>
this is a change that they have no notified their users or group list administrators about
hramrach has quit [Remote host closed the connection]
<rz2k>
rah: why do you send to sunxi+subscribe@googlegroups.com when you should send to linux-sunxi+subscribe@googlegroups.com ?
<rah>
they now require you to subscribe to mailing lists through their web interface, which involves agreeing to the Google terms and conditions
<rah>
rz2k: because I suck
* rah
chomps on the humble pie
hramrach has joined #linux-sunxi
<rah>
oh for god's sake
<rah>
Google is so fucking evil
<rah>
I sent a subscribe request with an email address of rah+sunxi@settrans.net
<rah>
they sent the confimation email to rah@settrans.net
<rah>
which I replied to
<rah>
so now rah@settrans.net is subscribed instead of rah+sunxi@settrans.net
<rah>
do you fucking see why these people are wrong?
<rah>
you should not be involved with these people
<libv>
*sigh*
<rah>
if you really want a mailing list, I'll fucking set you up a mailing list
<rz2k>
oliv3r has his +list subscribed right
<FR^2>
I presume google is financed by nsa and that's a reason not to trust them ;)
<rah>
I already have a working mailman installation
<rah>
rz2k: well woopdi-fucking-do
<rah>
oliv3r is not me
<rah>
libv and mnemoc: do you want me to provide mailing list services to the linux-sunxi group?
_BJFreeman has joined #linux-sunxi
<rz2k>
no, we dont want, with your attitude it is going to die in even worse agony than it could die out of google evilness. seriously dude, sort your problems out and start being useful, instead of whining about how everything is bad.
<mripard>
rah: to be honest, I trust more google than I trust a ranting guy poping out of nowhere.
_BJFreeman is now known as BJfreeman
<rah>
rz2k: I'm not whining about how everything is bad
<rz2k>
yeah, sure...
<rz2k>
:)
<rah>
rz2k: there's a very specific and very serious badness that I'm ranting about
<oliv3r>
hello
<oliv3r>
you rang?
<libv>
rah: you seem to have been active on the powervr-devel mailing list
<libv>
rah: how is that project going?
<jukivili>
google groups is mangling utf-8 in patches.. not big issue, but for example chinese comments are transformed in to unreadable state (if patch moved allwinner code around files)
<rah>
libv: I am the list administrator for the powervr-devel mailing list
<libv>
ok
<rah>
libv: I wouldn't really call it a project as such
<rah>
libv: the mailing list was set up because there were emails going round with a CC list that kept growing
<rah>
libv: they were going round amongst people who had an interest
<mnemoc>
jukivili: can you give me a pull request url for those?
<rah>
libv: the idea that there's some "project" is a bit of a mischaracterisation
<libv>
rah: ok
<libv>
rah: well, my view is that people with an interest in powervr usually tend not to have worked with it before
<rah>
libv: you're welcome to your view
<jukivili>
mnemoc: for the patches that I sent today?
<oliv3r>
oh yeah, i use oliver+list@google-groups.com
<jukivili>
they are based on Hans' sunxi-3.4 repo
<mnemoc>
.oO
<oliv3r>
so if they send it to rah@settrans.net, then your mailserver is setup wrong and it 'fell through' check your header
<libv>
given that i have done a few things in that area, and have kept a few good folks from erring down the pvr path, i found the powervr project rather curious
<rah>
oliv3r: what does that even mean?
<libv>
but understood, it is not really a project
<oliv3r>
rah: anyway, there are plans allready lead out to move to our own mailman, mnemoc needs some time to set it up on the new server, then we slowly transistion developer users over; 'users' may stay on google-groups as that's what they know etc
<rah>
2013-08-14 09:07:39 1V9X3L-0002b3-Vl <= rah+sunxi@settrans.net U=www-data P=local S=639 id=ba64cde25375c6107854ec6a3b3c5295@settrans.net from <rah+sunxi@settrans.net> for linux-sunxi+subscribe@googlegroups.com
<libv>
oliv3r: no split mls please :)
<rah>
can't see anything wrong there
<libv>
oliv3r: split mls only work at the height of a projects popularity
<rah>
oliv3r: that's good to know
<oliv3r>
libv: mls?
<oliv3r>
libv: oh split mailing lists
<oliv3r>
libv: yeah if we can somehow forward some of those messages to make the transistion easier for $users, 1 ml is better of course
<rah>
libv: the problem with not working on the pvr is that there are so many damned pvr chips out there
<oliv3r>
rah: so while your complaints are valid, they have allready beeing planned on being solved ;) give it time :p
<mnemoc>
A31 :'(
<rah>
A31 indeed
<jukivili>
mnemoc: on which tree should I be doing musb work? Hans already added 'musb: add host-mode sunxi support' to his tree.
<libv>
rah: imgtec is losing _a_ _lot_ of ground
<rah>
oliv3r: time? give me everything I want now dammit! I have a right to it and I demand it!
<libv>
rah: a31 and the second exynos 5 are the last sgx 5xx hickup
<mnemoc>
jukivili: it's fine. it's just that I'm kind of lost one what should I merge first
<rah>
libv: we'll see
<mnemoc>
jukivili: hard to catch up with the ML
<oliv3r>
rah: -> mnemoc :p
<jukivili>
mnemoc: ok
<oliv3r>
mnemoc: i have 600 mails on the ML from this week alone :(
<libv>
rah: trust me. sgx 5xx is done, and rogue is still a year out, and will be an absolute shambles
<rah>
in the mean time, my GTA04 and my two A31s all have a PVR chip :-/
<libv>
rah: i am not stopping you from doing the work
<libv>
if you _really_ want to, feel free.
<rah>
libv: glad to hear it
<oliv3r>
rah swap the a31s for a20's and join the cause
<rah>
when I get some spare time, I will work on it
<rah>
>_>
<rah>
oliv3r: not my choice unfortunately
<libv>
in the meantime, i will submit to the evil google overlord again, and post a video to youtube of a spinning smoothed cube.
<oliv3r>
libv: awe-some :D
deasy has quit [Remote host closed the connection]
<libv>
on the odroid, at 1024x768, 190fps on mesa/dri2, 495fps on limare/fb
<oliv3r>
oh snap
<libv>
but its clear where most of that performance is lost, so that might be regained easily once something stable and useful is out, and ssvb can poke at it as well
<oliv3r>
so that's mp4 though; so /3 for mp1?
<oliv3r>
roughly
<libv>
at least /3
<libv>
mind you, this is the nastiest solution
<libv>
i have to actually wait for the job to completely finish, and then i have to put in a usleep, as otherwise the copy_region will start too soon
<libv>
so the mali is sitting idle quite a lot of the time
<oliv3r>
what framerate are you looking at with blobs?
<libv>
oh, hrm...
<libv>
oliv3r: i am not going to test that atm, as that would mean having to replace most of the hacked-into-place mesa and driver
<libv>
but you are right, that is an important data point...
<libv>
i should run this on the sunxi, there i have two devices and two sd-cards, one with a hacked up mesa, one with the binaries
BJfreeman has quit [Quit: had a good time]
<oliv3r>
libv: i would expect it to run around 450 FPS :p
<libv>
no, far from it, it also sticks in a copy
<oliv3r>
libv: btw, we got extremly lucky on the A3, on our way out on 'black saturday' we had no stau at all, on our way back, there was a small accident between nuernberg and wuerzburg, took us maybe 20 or 30 minutes extra
<libv>
with limare and fb i just write as fast as i can, and flip buffers all the time (without looking at what the display engine is using)
<libv>
ah, lucky :)
<oliv3r>
yeah, my parents went on a thursday and had to crawl have way through germany ;)
<libv>
we are going to attempt to drive part of the A6 (to the sinsheim museum where a concorde and a russian equivalent can be visited) on saturday... i hope that between 8 and 10 it will not be too bad...
<libv>
my old man and my brother are driving over this afternoon, and driving back sunday (even though i seriously warned against that)
<oliv3r>
vacation time is starting to get to an end isn't it
<ssvb>
libv: xcb is also slower than old xlib when handling lots of requests, which don't need any responses (xcb is optimized for the roundtrips use case, but is slower on the bulk fire-and-forget workloads)
<oliv3r>
rah aren't those identical?
<oliv3r>
rah i think ifradead is the recent one
<ssvb>
libv: anyway, we can probably indeed hack something with dri2 so that it becomes faster
<rah>
oliv3r: that's odd because I'm subscribed to the arm.linux.org.uk mailing list but get no emails
<ssvb>
libv: may help to run more games with blobs, but maybe this is not relevant anymore when having a mesa driver
<oliv3r>
in the archives i see identical mails; but infradead is where i send my mails to
<libv>
ssvb: it is relevant, as i currently run off the binary compiler still
<hramrach>
no shaders :s
<libv>
ssvb: i still need to look into how emulation of the fixed function pipeline is done, and whether i can transplant this more directly onto essl without having to use mesa glsl compiler
<hramrach>
but at least there is source now
<libv>
but this stuff seems to go straight to gles2, good.
vinifr has joined #linux-sunxi
hipboi has quit [Read error: Connection reset by peer]
hipboi has joined #linux-sunxi
<rah>
is there an A31 tree or do patches for the A31 just go in the linux-sunxi tree?
<mnemoc>
rah: look at import/lichee-3,3/a31-dev in amery's github
<mnemoc>
that's reference only. the goal is to integrate sun6i support with the rest
<rah>
mnemoc: so the initial destination for patches should be the linux-sunxi tree?
<mnemoc>
there is a mainlinig effort too. but that's still very early
<rah>
I don't understand
<mnemoc>
allwinner has separated branches per soc
<rah>
saying "too" implies that there is some other effort
<rah>
but you've only mentioned a reference import
foubarre has quit [Remote host closed the connection]
<mnemoc>
A31 support has wo destinations sunxi-3.4 AND mainline
<mnemoc>
s/wo/two/
<rah>
I see
<mnemoc>
the core of A31 is alreay been integrated into mainline
<rah>
and sunxi-3.4 is a branch in the linux-sunxi repository
<mnemoc>
yes
<mnemoc>
stage/sunxi-3.4 for development
<oliv3r>
stage/sunxi-3.4 + sunxi-3.4
<oliv3r>
i'm too slow :(
hipboi has quit [Quit: Leaving]
<vinifr>
Can we merged linux-sunxi drivers to mainline?
vrga has joined #linux-sunxi
zumbi has quit [Quit: Lost terminal]
vrga has left #linux-sunxi [#linux-sunxi]
<mnemoc>
rah: so from the reference branch to stage/sunxi-3.4, unified, tested, improved, ....... and then DTied and mainlined
<mnemoc>
vinifr: the current state of most drivers doesn't allow them to be mainlined, yet. work is been done doing new code, but focused at the very core of the platform
<vinifr>
why drivers submitted to mainline are different from original, for example, i2c
<mnemoc>
larger drivers will be mainlineable after "some" refactoring, unification, testing and later DTification
<mnemoc>
vinifr: original code is crap, but without documentation they are the reference. once someone understand the unit well enough they can try to write a driver from scratchs
zumbi has joined #linux-sunxi
<mnemoc>
vinifr: or "discover" that allwinner actually is using a modified version of a driver that is already mainlined
<mnemoc>
vinifr: so we don't need to fix their junk but to extend the already mainlined code to support allwinner's variant of the IP
<vinifr>
alright
<mnemoc>
makes more sense now?
<vinifr>
yes, and I agree that many drivers have codes not very good
navlrac has joined #linux-sunxi
<vinifr>
mnemoc, and the device tree is a wonder. I might start using the same kernel on a13 and a10 :)
<mnemoc>
vinifr: we can start using the same kernel for a13 and a10 on 3.4 too thanks to soc-detection
<mnemoc>
drivers need to be modified to not #ifdef but if (sunxi_is_a20()) {
<vinifr>
linux-sunxi 3.4?
<mnemoc>
and we can have single-kernel with magic
<mnemoc>
yes, stage 3.4 currently has soc-detection
<mnemoc>
works for sun[4567]i
<mnemoc>
magicly, and they are all armv7 ... so single bin is possible
<mnemoc>
pending work is to refactor script.bin handling into a generic platform_device generator
<mnemoc>
so the drivers start to work as civilized drivers
<mnemoc>
even when using script.bin
<vinifr>
I realized that peripherals are also similar
<vinifr>
to a13,a10,a20
<mnemoc>
there are more things identical in sun[34567]i than different
<mnemoc>
that's why unification is critical
<wigyori>
vinifr: not sure about the need for a clock on lradc if it's needed
<wigyori>
prob not
<wigyori>
haven't had a chance to actually run your code and check it out but by the looks of it it's much further ahead, so let's use yours instead of mine
<vinifr>
wigyori, there are things that should be reviewed before testing
<vinifr>
one is the channel selection and other, source clock
<vinifr>
i am using 'ADC_CHAN_SEL(chan->channel)', but I'm not sure if chan->channel is the channel that user want to read
<vinifr>
line 200
eebrah|away is now known as eebrah
tinti has joined #linux-sunxi
jacq|2 has joined #linux-sunxi
Ehsand has quit [Read error: Connection reset by peer]
Ehsand has joined #linux-sunxi
zerodamage has joined #linux-sunxi
Ehsand has quit [Quit: Leaving.]
<zerodamage>
hi, anyone tried tslib with sun4i_ts
<zerodamage>
?
<oliv3r>
touschreen? or transportstream?
<zerodamage>
touchscreen
<zerodamage>
kernel module is called sun4i_Ts
<oliv3r>
that's for resistive touchscreens however
<zerodamage>
yes
<oliv3r>
while possible i think very few people actually use that ;)
<zerodamage>
well i do :)
<zerodamage>
or rather, im trying to
<zerodamage>
evtest shows touch event fine, but tslib will not register anything
<zerodamage>
actually, i dont care if its tslib or something else, as long as the events get to X11
<ssvb>
mnemoc: would it be a good idea to make sunxi-3.4 the default branch on github?
<mnemoc>
ssvb: i don't know the impact that would have
<mnemoc>
as i don't know if android people can use that branch yet
<ssvb>
ah, those mythical android people.... :)
<mnemoc>
*g*
<mnemoc>
done
\\Mr_C\\ has quit []
techn_ has quit [Read error: Connection reset by peer]
panda84kde has quit [Quit: Konversation terminated!]
deasy has joined #linux-sunxi
<jeremb>
mnemoc: what about updating sunxi-bsp to refer to sunxi-3.4 by default ?
<mnemoc>
can't do that atm.... poke techn :)
<mnemoc>
u-boot integration also needs to be rebumped
<mnemoc>
will try to move the builder and sunxi vservers to the new hardware this sunday, after getting the new data plan activated. but for now, I'm limited to andchat + znc and connectbot :|
<jeremb>
mnemoc: ok no hurry
<mnemoc>
having the build server relocated I'll be able to catch up with the "hwpacks", pulls and merges
<jeremb>
great
navlrac has quit [Ping timeout: 250 seconds]
ZetaNeta has joined #linux-sunxi
zerodamage has quit [Quit: Page closed]
Funkypenguin has quit [Read error: Connection reset by peer]
Funkypenguin has joined #linux-sunxi
<theOzzieRat>
spi
atiti has quit [Ping timeout: 245 seconds]
vinifr has quit [Quit: Saindo]
bsdfox has joined #linux-sunxi
bsdfox has joined #linux-sunxi
techn_ has joined #linux-sunxi
wingrime has joined #linux-sunxi
<wingrime>
oliv3r:
<wingrime>
ping
<RaYmAn>
Usage: csis-irk.exe [options]
<RaYmAn>
Options: --cli
<RaYmAn>
uhm
<RaYmAn>
wtf
<RaYmAn>
that was so not supposed to go here =P
* RaYmAn
pets irssi for warning before pasting the remaining million lines
zerodamage has joined #linux-sunxi
<zerodamage>
anyone with sun4i_ts knowledge?
<rz2k>
zerodamage: no one. if you need help with it - read the sources
<rz2k>
it is fairly limited driver, by the way.
<rz2k>
for example, it should support two point multitouch, but that doesnt actually work
<rz2k>
:p
<wingrime>
rellla: ping
<zerodamage>
i know, actually the sun4i_ts part i think works, but i cannot get it to the X server
<zerodamage>
tslib says its not a touchscreen even after patching from buildroot git
<techn_>
zerodamage: sun4i_ts bug.. it should give correct flags for userspace
<hramrach>
some time ago you said you were looking for linux developers or something like that
<wingrime>
oliv3r: some guy used some decompiller / disassebler
<drachensun>
ah yes interested?
<wingrime>
oliv3r: and try rewrite libve
<hramrach>
interested in asknig more details at the very least
<oliv3r>
wingrime: ahh ok status?
<drachensun>
ok, I sent a pm
<wingrime>
oliv3r: The core part (libve.c) and major part of VP8 decoding is ready (but not actualy tested).
<wingrime>
oliv3r: he find many usefull info accoring some bit-fields
<wingrime>
oliv3r: but this is direct rewrite
<wingrime>
oliv3r: it can't be used
<oliv3r>
yeah but info can be extracted for docu
<oliv3r>
:)
<wingrime>
oliv3r: better ask him onboard
<wingrime>
oliv3r: it looks like he have mistery hex-rays arm decompiller
<oliv3r>
well any new information we can document is good
<wingrime>
oliv3r:he can give much help with VP8
Colani has joined #linux-sunxi
Colani has quit [Client Quit]
<wingrime>
oliv3r: linux-sunxi going too fast for me now
<oliv3r>
wingrime: you mean IRC? or ML
<oliv3r>
wingrime: i still have to back-read tons from the last week
<drachensun>
lol both if you ask me
<oliv3r>
lots of development == good ;)
<drachensun>
I've got to just search for key terms these days and ignore a lot of it
<drachensun>
yup
<drachensun>
its not really a complaint
Colani1200 has quit [Ping timeout: 256 seconds]
<oliv3r>
but i have to work on 2 unrelated things right now
<oliv3r>
first, i have to get my tvheadend setup working with our new dish (dvb-t -> dvb-s)
<oliv3r>
and i have to build an at-tiny85 IR receiver for our ne TV box
<wingrime>
oliv3r: thats guy posted in ML
<wingrime>
oliv3r: I also have some unrelated works
<wingrime>
oliv3r: I reading some books about Machine Learning
<oliv3r>
and vacation time == time with family :p
<wingrime>
oliv3r: and probability theory
<oliv3r>
wingrime: 3D printing/CNC stuff?
<wingrime>
oliv3r: actualy no,
<wingrime>
oliv3r: wtf how CNC can be related with Machine Learning?
<oliv3r>
i saw some of your video here you built some machine! :p
<wingrime>
oliv3r: actualy machine learning can also be named - Data mining
<oliv3r>
ahh
jacq|2 has quit [Remote host closed the connection]
<wingrime>
oliv3r: for example, genetic algos
<wingrime>
oliv3r: also, good example, any search engine
<wingrime>
oliv3r: spam filtring
<oliv3r>
ah yeah
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
notmart has quit [Quit: notmart terminated!]
BJfreeman has quit [Quit: had a good time]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
sanka has joined #linux-sunxi
boycottg00gle has joined #linux-sunxi
ZetaNeta has quit [Ping timeout: 245 seconds]
<boycottg00gle>
hey! just got an olimex a20 and booted debian :) unfortunately colors on hdmi output are broken, and my monitor doesn't like the resolution? how do i change that? do have to go through that fex stuff or is there a way to change the mode at runtime? (ssh works just fine)
ZetaNeta has joined #linux-sunxi
<techn_>
boycottg00gle: use community kernel
<techn_>
stage/sunxi-3.4
ynezz has joined #linux-sunxi
<ynezz>
Hi! I've gathered some manuals for A13 around the interweb and can't find anything about SDRAM controller in any of them
ZetaNeta has quit [Ping timeout: 246 seconds]
<boycottg00gle>
techn_: i think that already is a community kernel
bsdfox has quit [Read error: Connection reset by peer]