marcan changed the topic of #asahi to: Asahi Linux: porting Linux to Apple Silicon macs | General project discussion | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Topics: #asahi-dev #asahi-re #asahi-gpu #asahi-stream #asahi-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
torstenvl has joined #asahi
torstenvl has quit [Remote host closed the connection]
<marcan>
it should be configurable, but I'd say default to disarm since we can't assume the kernel has that driver built in / enabled
INTL has quit []
torstenvl has joined #asahi
yuyichao_ has joined #asahi
torstenv_ has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
torstenv_ has quit [Remote host closed the connection]
torstenvl has quit [Ping timeout: 480 seconds]
yuyichao has joined #asahi
yuyichao_ has quit [Ping timeout: 480 seconds]
hendry1 has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
hendry has quit [Ping timeout: 480 seconds]
<marcan>
going to stream a bit working on idevicerestore; I need to add Monterey support (I already have a PoC that relies on some manual steps; I just need to integrate it)
<marcan>
since my Max is coming in today and I want to make sure all that works fine on the new machines too
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
yuyichao has joined #asahi
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
bgb_ has joined #asahi
bgb has quit [Ping timeout: 480 seconds]
c10l has quit [Ping timeout: 480 seconds]
aquijoule_ has joined #asahi
richbridger has quit [Ping timeout: 480 seconds]
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
PhilippvK has joined #asahi
aquijoule_ is now known as richbridger
kov has quit [Quit: Coyote finally caught me]
torstenvl has joined #asahi
torstenvl has quit [Remote host closed the connection]
torstenvl has joined #asahi
torstenvl has quit [Remote host closed the connection]
torstenvl has joined #asahi
torstenvl has quit [Remote host closed the connection]
torstenvl has joined #asahi
torstenvl has quit [Ping timeout: 480 seconds]
marvin24_ has joined #asahi
marvin24 has quit [Ping timeout: 480 seconds]
robinp[m] has joined #asahi
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
robinp[m] has quit [Ping timeout: 480 seconds]
<Ariadne>
marcan has a cute doorbell (?)
<marcan>
welcome to Japan :)
<marcan>
it's even configurable!
<Ariadne>
maid cafes and configurable doorbells, amazing
<marcan>
*two* touchscreens; there's another unit in the bedroom :)
<marcan>
(and you can even use them as intercoms with each other)
<marcan>
it also lets me know when I receive a package in the shared lockers next to the mailboxes
<marcan>
fun fact: the brand is "aiPhone". Yes, really. This company had the trademark in Japan before Apple.
<marcan>
Apple uses the iPhone trademark under license from Aiphone K.K.
<marcan>
apparently apple paid about $1M for the license or so
<marcan>
then in 2015 apparently they started selling doorbells that link up with iPhones :-) (my apartment is too old for that)
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
torstenvl has joined #asahi
bgb has joined #asahi
bgb_ has quit [Ping timeout: 480 seconds]
bgb has quit [Quit: WeeChat 3.3]
chamomile has joined #asahi
torstenvl has quit [Remote host closed the connection]
linearcannon has joined #asahi
torstenvl has joined #asahi
torstenvl has quit [Ping timeout: 480 seconds]
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
sailorek1234 has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
yuyichao has joined #asahi
nkaretnikov has quit [Read error: Connection reset by peer]
nkaretnikov has joined #asahi
sjg1 has quit [Ping timeout: 480 seconds]
sjg1 has joined #asahi
torstenvl has joined #asahi
aleasto has joined #asahi
jabashque has quit [Read error: Connection reset by peer]
jabashque has joined #asahi
the_lanetly_052 has joined #asahi
the_lanetly_052 has quit [Remote host closed the connection]
the_lanetly_052 has joined #asahi
torstenvl has quit [Ping timeout: 480 seconds]
<tmlind>
fyi, plans to drop armv6 from omap2plus_defconfig by default and to make armv6 depend on !smp being discussed on the #armlinux, this to support vmap stack
<tmlind>
sorry wrong channel
<chadmed>
quite a few buildings here have aiphone systems, some of them are super cool!
<chadmed>
nothing as complex as parcel delivery notifications though, couriers and auspost workers usually just dump the boxes in the foyer and we rely on human decency for them to not be stolen
<kettenis>
marcan: autostarting the watchdog timer has been the u-boot default since forever and this seems to be standard practice in the "embedded" world
<kettenis>
but there is a config option, so we could opt out for Apple Silicon systems if we think that is the right thing
X-Scale has joined #asahi
X-Scale` has quit [Ping timeout: 480 seconds]
<kettenis>
to be honest, I think it would be silly to build a linux kernel without the watchdog driver
[X-Scale] has joined #asahi
<kettenis>
at least as long as it is required to be able to reboot the machine
X-Scale has quit [Ping timeout: 480 seconds]
<sven>
even afterwards i don't really see a reason to skip it given how small the watchdog driver is
<maz>
marcan: I'm collecting PCIe fixes. I remember that you have a patch fixing some clock stuff, so if you want me to carry it, let me know.
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
gladiac is now known as Guest6113
gladiac has joined #asahi
<marcan>
kettenis: fair
Guest6113 has quit [Ping timeout: 480 seconds]
<marcan>
maz: I'll just fire them off
<mps>
in Developer-Quickstart I read in Step: 2 'Install macOS in the stub partition' part about "Reinstall macOS Big Sur"
<mps>
is this step necessary and why, I don't understand
gladiac is now known as Guest6116
gladiac has joined #asahi
<j_ey>
mps: if you use the new asahi installer, you dont need that
<mps>
j_ey: 'new asahi installer', what is it and where
<marcan>
you need to resize macos first, it won't do that for you
<opticron>
yeah...just seems like a waste if the ANE is the only thing that's bad, but then again they don't have a secondary market for those chips like Intel or AMD
<marcan>
just leave free unpartitioned space
<mps>
marcan: on machine connected to dest box with cable?
<marcan>
yes
<marcan>
on the dest box
<marcan>
the install is local, there is no remote install
<mps>
you mean I don't need usb cables and separate machine?
<marcan>
ideally do it from 1TR, but it will also work from macOS (for now; Apple can break that at any time)
<marcan>
you want a separate machine for development
<mps>
marcan: ok, nice
Guest6116 has quit [Ping timeout: 480 seconds]
<marcan>
you don't *need* it but I don't recommend spending 5 minutes to go through 1TR to flash a new m1n1 with a burned in kernel every time you want to try something, when using another machine makes it 10 seconds :p
<marcan>
(and you need another machine for any kind of interactive experimentation including the hypervisor)
<mps>
I tested usb connection with old chromebook over usb cable, it works fine, so I have usb 'console' ready
<marcan>
chromebook running linux?
<mps>
yes
<marcan>
should work as long as you have an aarch64 toolchain
<mps>
I'm running linux on chromebooks about 5-6 years now
chamomile has quit [Ping timeout: 480 seconds]
<mps>
I already built m1n1 on it
<marcan>
cool :)
<maz>
marcan: if you are confident that it does something useful and doesn't break existing stuff, I'll take it!
<mps>
thank you all for explantion and help, going to make archive first and then try about script
<maz>
marcan: I have a patch that allows the kernel to survive u-boot (as expected, we're missing the tickling of #PERST).
<marcan>
maz: I'm not confident that it does something useful, I just have a spidey sense that it might matter and shouldn't hurt (and macOS sets those bits too)
<marcan>
CGDIS sounds like clockgate disable which sounds like clearing those bits would enable some form of dynamic clock gating
<maz>
fire away!
<marcan>
fired :)
<maz>
thanks!
the_lanetly_052 has quit [Remote host closed the connection]
<maz>
marcan: btw, is there a way to get m1n1 to do the same thing as cpu_pstates.py by default? somehow, that doesn't play well with a chainloaded u-boot.
<balrog>
marcan: is it possible that the ane that appears is determined by firmware and calibration data, so it remaps the device to ane0 regardless of which one it really is?
<marcan>
maz: it already does
<marcan>
just not to 100%
<marcan>
I think I set the p-cores to 2GHz or so
<maz>
marcan: right. that's the 100% I'm after! ;-)
<marcan>
just change the 7s to 15s in cpufreq.c :p
<marcan>
balrog: very unlikely
<marcan>
that would involve magically connecting it to another PMGR, among other things
gladiac has quit [Ping timeout: 480 seconds]
* maz
builds
<marcan>
e.g. the PMP and PMGR device lists still have ANE0 and ANE1
<marcan>
so they'd somehow transplant ane1 into ane0 in the devicetree, including those points, and end up with mismatched names...
<marcan>
it doesn't really make sense
<marcan>
and there's no way they have hardware to make ANE1 "show up" as ANE0 in addresspace, PMGR, etc
<maz>
marcan: works perfectly, thanks a lot. now only one step away from being able to permanently use u-boot!
[X-Scale] has quit [Ping timeout: 480 seconds]
<_jannau_>
if they could do that mapping it wouldn't make sense to list ane1 in the adt at all
X-Scale has joined #asahi
gladiac has joined #asahi
yuyichao has quit [Ping timeout: 480 seconds]
<marcan>
just sent out the simpledrm fixes and the tipd fixes too
<marcan>
everyone gets a fix!
<marcan>
maz: btw, let me know if you notice any magical speed improvements with m1n1 as of yesterday
<marcan>
since apparently we weren't enabling the L3/system cache... :-)
<maz>
marcan: Debian still installs in 2.23, so no changes so far. but I'm chainloading, so I don't know if I would see it.
<marcan>
it should
aidenfoxivey has quit [Quit: aidenfoxivey]
<marcan>
it does the init unconditionally (it's a no-op if it has already happened)
aidenfoxivey has joined #asahi
<marcan>
unless there's another bit somewhere to enable actually using the cache
<maz>
I mean, this is already 30% faster than the best KVM/arm64 host. anything faster would just be illegal.
chamomile has joined #asahi
<mps>
sorry for uneducated question: M1 linux use simpledrm?
<marcan>
define "m1 linux" :)
<marcan>
it'll use whatever you build
<mps>
ok, I mean linux kernel on M1
<marcan>
simplefb if you build that, simpledrm if you build that instead, and/or alyssa's DCP driver once that's merged
<mps>
aha, thanks
<marcan>
simpledrm is new, the fixes I just sent make it work like simplefb does (otherwise it regresses since you can only have one or the other)
<mps>
yes, simpledrm made some issue with newly upgraded kernel for alpine linux on some users machines (that is why I asked)
<marcan>
I expect the long term route to be simpledrm -> dcp
<marcan>
i.e. simpledrm for boot and dcp once that module is loaded
<mps>
I'm testing these days simpledrm on two cromebooks, one doesn't even booted
<marcan>
simpledrm doesn't even bind to /chosen framebuffers right now
<marcan>
so it won't work on ~any devicetree platform
<mps>
had to revert it
<marcan>
that's my first patch
<marcan>
you don't need to revert anything, just disable it and enable simplefb instead
yuyichao has joined #asahi
<mps>
well, I tried to remove simplefb (which was mistake for now)
leo60228 has quit [Ping timeout: 480 seconds]
sailorek1234 has quit []
leo60228 has joined #asahi
<marcan>
also lol, turns out the DRM guys made a major change to the format helpers a week ago... so now I'm going to have to rebase/rewrite half the series I just sent out
<marcan>
(at least it's short)
<FireFox317>
so i have a usb-c hub with 2 usb 3.0 ports, does this work with the current dwc3 bindings that we have?
<_jannau_>
it works with usb2 speeds if you have dwc3 bindings (mainline for M1 has none). hotplug at the hub should work too
<mps>
marcan: 'curl -L https://mrcn.st/alxsh | sh' failed when downloading one of big files with 404, restarted and it says it don't have anything to do
<FireFox317>
_jannau_, yeah okay that makes sense. thanks i'm gonna try that :)
<marcan>
that link works fine so I doubt it failed to download that file
<mps>
and I forgot which file name, but error was 404, so probably network problem
<marcan>
I just tested all the ipsw URLs and none of the m404
<marcan>
anyway, you can just delete any partitions it created and try again
<mps>
we are not on same continent I think
<marcan>
I'm pretty sure Apple's CDN is in most continents ;)
<mps>
ah, that is apple cdn
<mps>
nvm, I will try manual method as described in quickstart
<marcan>
this isn't the first time it fails due to a bad network for someone; I should add retries...
<marcan>
well, if you had trouble downloading ~1GB using this installer, I think you're going to have more trouble downloading 15GB using the quickstart method :p
<mps>
idk, network usually works but not always, I'm on cable/docsis link
<FireFox317>
mps, yeah i think it also failed for me one time, and i tried again and it worked :D
<brentr123[m]>
Is there a way to easily switch back between Linux and MacOS yet? Even though not everything works, I wanna try for shits and giggles
<marcan>
brentr123[m]: dual-booting has worked from day 1
<marcan>
it's part of how the machines are designed
<brentr123[m]>
Ah and I have to go into 1tr for that? And enter some commands or do I go into a multi boot screen
<marcan>
(as long as you don't go around telling people to *replace* their macOS kernel with Linux like corellium did... but we never did that here :p)
<marcan>
no
<marcan>
you just hold down the power button and pick an OS
<marcan>
the 1TR thing is the nonsense corellium was pushing
<brentr123[m]>
Ah ok
<marcan>
this is why I've been telling people to do a whole second macOS install in the quickstart, so dual-boot works properly
<marcan>
(and why we have an installer now to make that unnecessary)
<_jannau_>
the initial recovery screen is a boot picker
<brentr123[m]>
There’s no risk to bricking with this right?
<marcan>
these machines are not brickable
<marcan>
(to my knowledge)
<marcan>
worst case you can do a DFU restore from another machine
<brentr123[m]>
oh i need to resize macos
<brentr123[m]>
hmmm
<brentr123[m]>
not sure how to do that
<marcan>
the quickstart guide has instructions for that
<marcan>
just stop after the resize step and use the installer after that
<brentr123[m]>
ah ok
<marcan>
the installer will eventually do this for you
<brentr123[m]>
sorry and uh... where is said quickstart guide:
<marcan>
stop after resizeContainer, then just boot into 1TR and run curl -L https://mrcn.st/alxsh | sh
<marcan>
you don't have to make it 200GB, but you need at least 2.5GB for the bootloader plus whatever space you want to have free to create linux partitions
<marcan>
so subtract that from your total space
<brentr123[m]>
at least it seems easier then windows disk management 😛
<FireFox317>
you can also decide to use a usb stick for the linux rootfs, then you only need the 2.5G for the m1n1 stuff
<marcan>
also make sure you have macos 12.0.1
<marcan>
mostly because I'm going to stop supporting anything older any time now
<maz>
marcan: shit, I need to remember my macos password...
<marcan>
heh :)
<chadmed>
does 12.0.1 fix the early fb problem
<marcan>
mine was "aaaa" until I enabled SSH the other day on stream and realized everyone saw my IPv6 pass by :p
<marcan>
the wonders of properly routed home networks
<marcan>
chadmed: no idea
<maz>
ah... with a bit of luck, this is the same 3 letter passwd that I use on everything else...
<brentr123[m]>
im just gonna wait until the installer does everything lol, too nervous
<_jannau_>
chadmed: no, 12.0.1 was the official monterey realease everything except 2021 macbook pros. i.e. it caused the problem for me
<FireFox317>
marcan, i guess for initial testing and stuff it doesnt matter if the m1n1 os still has 11.x firmware?
<_alice>
marcan: huh, you don't have non-ICMPv6 traffic blocked by default?
<marcan>
no, why would I?
<marcan>
the whole point of IPv6 is to make stuff work as intended
<marcan>
firefox317: it'll start mattering with ASC drivers, especially DCP
<_alice>
isn't that the usual thing to do if you don't want random hosts to prod at your stuff with IPv6? (since you don't have NAT to obfuscate your network anymore)
<FireFox317>
marcan, yeah okay that makes sense
<marcan>
_alice: you shouldn't rely on obfuscation
<marcan>
if you don't want random hosts to prod at your stuff... don't have random stuff with open ports accepting connections from the internet
<marcan>
I think the only thing I have explicitly blocked at the router is my printer, because yes, the silly thing does IPv6 and doesn't block things properly
<_alice>
hmm, fair, I just thought it was better to block it from the firewall-level since you can't always trust non-Linux hosts and such to not listen on things randomly
<kettenis>
marcan: it seems the DRM folks spend 90% of their time reorganizing the codebase
<marcan>
heh
<kettenis>
and have been for the last decade
<marcan>
_alice: I don't really have non-linux hosts, except on wifi
<marcan>
any machine on my normal wifi is going to be a machine liable to be on any other random wifi and better not have dangerous open ports, v4 or v6
<marcan>
IoT stuff is on an isolated SSID with no internet access at all
WhyNotHugo has quit [Remote host closed the connection]
handlerug has quit [Remote host closed the connection]
aidenfoxivey has quit [Quit: aidenfoxivey]
aidenfoxivey has joined #asahi
<_alice>
in my case, it's mostly just not having the energy to audit that every device is adequately secure in terms of random listening ports ><
WhyNotHugo has joined #asahi
handlerug has joined #asahi
<_alice>
although, it doesn't help that macOS still uses password authentication for SSH by default (and changes to sshd_config are reverted on OS updates due to SSV)
<sven>
yeah, I have everything incoming blocked in my router just because I don’t want to spend my energy auditing random devices like my printer
<sven>
marcan: don’t forget the arm
aidenfoxivey has quit [Quit: aidenfoxivey]
<sven>
pagetable fix
aidenfoxivey has joined #asahi
<mps>
ok, I'm upgrading macos from 11.4 to 12.0.1
<mps>
and will try installer again
<marcan>
sven: yeah
<marcan>
_alice: it's not SSV, it's some other nonsense that causes those reverts
<marcan>
something with the updater
<marcan>
it just decides to clobber it
<_alice>
hmm, maybe it's the behaviour with the read-only system volume then (before they did APFS sealing of the root volume)
<_alice>
I noticed it starting to clobber my changes when they introduced that
<marcan>
it's not the sealing, the sealing stops you from modifying / outright
<marcan>
/etc is in the data volume (well, it's a symlink to /private/etc which is)
<marcan>
SSV/read-only SV stuff doesn't behave like that; it stops you from modifying it outright (and reverts changes to the snapshot on every boot)
<_alice>
oh, right, firmlinks were introduced in catalina
<marcan>
/etc getting clobbered on updates is just some updater stupidity
<marcan>
(which may have been introduced at the same time)
<_alice>
iirc Read Only SV just mounts the root filesystem as read-only, and has firmlinks (which are basically bi-directional symlinks) that 'overlay' the /private/etc and similar directories on top of /, from the data volume (/System/Volumes/Data)... I remember you used to be able to run commands in recovery or w/e to simply mount the root filesystem as read-write and have your changes persist... but Sealed/Signed SV is replacing the / mount with a
<_alice>
signed snapshot and enforcing that on boot
<marcan>
oh, apparently they have a .d now?
<marcan>
yes, and none of that has anything to do with /etc due to the firmlink :p
<marcan>
/etc/ssh/sshd_config.d is apparently a thing now, maybe if you drop a config file there it won't get deleted/reverted on updates
<_alice>
hmm, maybe this is just the updater blindly over-writing the root filesystem to make it consistent with the contents of / after the update, and the unintended consequence is that it clobbers changes
<_alice>
oh, that's neat.
torstenvl has joined #asahi
<_alice>
now if they could add a way to make changes to /etc/pam.d/sudo for Touch ID, and /etc/auto_master for NFS, then I'd have all of my use cases covered ><
<marcan>
well, I just did that and am updating to 12.1 beta3, let's see
torstenvl has quit [Remote host closed the connection]
<mps>
after upgrade to 12.0.1 axlsh script worked, but when trying to boot from linux disk boot process fails, repeating few times reboot and then goes to macos asking to reinstall it
<FireFox317>
sometimes i get a 'usb: tps6598x_powerup failed for /arm-io/i2c0/hpmBusManager/hpm0' 'usb: failed to init hpm0' error message in m1n1, and then m1n1 is not recognized by my linux laptop. is this a known bug?
roxfan has quit [Remote host closed the connection]
pg12_ has joined #asahi
roxfan has joined #asahi
pg12 has quit [Read error: Connection reset by peer]
<sven>
marcan: just pushed a small change for the t6000 dart changes (of_device_get_match_data instead of open-coding it). if that still works on the pro/max i'll submit that series as well
<jannau>
sven: refreshed dart/t6000 works on t6001 with 'compatible = "apple,t6000-dart";'
___nick___ has joined #asahi
Palaver has joined #asahi
Palaver has quit [Remote host closed the connection]
palmer_ has joined #asahi
palmer_ has quit [Remote host closed the connection]
palmer_ has joined #asahi
palmer_ is now known as remlap
remlap has quit [Remote host closed the connection]
bpalmer4[m] has left #asahi [#asahi]
palmer_ has joined #asahi
palmer_ has quit [Remote host closed the connection]
palmer_ has joined #asahi
palmer_ has quit [Remote host closed the connection]
chamomile has joined #asahi
<sven>
awesome, thanks!
povik has quit [Ping timeout: 480 seconds]
aidenfoxivey has quit [Ping timeout: 480 seconds]
aleasto has quit [Remote host closed the connection]
___nick___ has quit [Ping timeout: 480 seconds]
<FireFox317>
jannau, its a m1 macbook air. will see if i get a repro tomorrow
<radnic>
So, stupid question, how do I select the default boot volume back to the original one?
<jannau>
radnic: permanently? select options in recovery and then "Startup disk" from the apple menu
<radnic>
Ok, solved, thanks.
<radnic>
I am a bit confused aobut what I did when I booted m1n1, so maybe someone can dispel the confusion in my mind. When I ran curl buildmachine.lan:8000/m1n1.macho > m1n1.macho I got the file compiled on the build machine. And then I told it "kmutil configure-boot -c m1n1.macho -C -v /Volumes/Linux"
<radnic>
I.e that means boot this kernel( m1n1, right?)
<j_ey>
for that Volume, yes
<radnic>
BUt I did not give it a path for m1ni, so how does it know how to get to it?
<j_ey>
'-c m1n1.macho'
<radnic>
But that is not a full path.
<radnic>
Ah, it just takes the current directory.
<radnic>
It seems after I ran bputil -n the m1n1 is simply gone. Or I can't find it.
<radnic>
Hmm, now I tried to set bootpoloty back to m1n1 but it says code unknown bootpolicy error: 17
<jannau>
you have to change the default startup disk back to asahi volume
<radnic>
yes, worked now. Thanks.
<radnic>
Interesting that it removes the m1n1.macho if you ever boot in MacOs.
<radnic>
Sooo, now I get the I2C hpm0 error terribly consistently. My GF is totally annoyed by the mac booting sound by this point.
<j_ey>
radnic: you should have turned it off when you were in macOS!
<radnic>
I did now.
<radnic>
Sooo, 10/10 times I get that 12c hpm0 error. Do you think this is valueable somehow?
<jannau>
which hardware? if a laptop is a charger connected? what happens if connect or disconnect the charger?
<radnic>
Air 13. Charger connected. Same happens with nothing connected.
<radnic>
Same happens with all connected.
kettenis has quit [Read error: Connection reset by peer]
kettenis has joined #asahi
minecrell has quit [Quit: Ping timeout (120 seconds)]
minecrell has joined #asahi
<radnic>
what is the "adt" ?
<j_ey>
radnic: apple device tree
artemist has joined #asahi
kettenis_ has joined #asahi
kettenis has quit [Read error: Connection reset by peer]
<radnic>
So, when I run kmutil does that copy the .macho in a different location where it boots from?
<kettenis_>
yes
<radnic>
I hoped that I could just copy the new one on top of the old one so I could avoid running 2 commands.
<radnic>
Is there a simple way to update the m1n1 without thw whone tiersome boot cycle/curl/update?
<kettenis_>
no, that would break the security model
<kettenis_>
this is why m1n1 support chainloading itself