thelounge7571340 has quit [Remote host closed the connection]
kevans91 has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
kov has quit [Quit: Coyote finally caught me]
<ElvishJerricco[m]>
I wonder if it would be reasonably possible to implement a firmware TPM in m1n1 using the SEP. Or is m1n1 just too far out of the picture by the time the OS has booted?
<chadmed[m]>
m1n1 does not leave any runtime artefacts for the OS to call into plus the SEP exists so why would we ever need an fTPM implementation in m1n1
<chadmed[m]>
once we have control of the SEP it can do pretty much everything a TPM can, and can do it better
<chadmed[m]>
especially given that it comes ready made with its own encrypted storage and private interface to it. relying on m1n1 to try and do the same thing in the easily accessed ESP is just asking for trouble
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
<ElvishJerricco[m]>
chadmed: Well there's a couple of reasons to want a TPM. 1) They're meant to be available early in the boot process so that the PCRs contain firmware measurements. And doesn't asahi used "reduced security" or something, with some self signed keys that get enrolled with the platform? Because that means we can trust m1n1 stage 1, but we don't know what got started in between then and Linux (of course a proper secure boot
<ElvishJerricco[m]>
implementation would be best, but a TPM can tell you a little more than that). 2) There's a LOT of existing userspace tools for using a proper TPM device. Like systemd has a variety of things having to do with it that enable e.g. passwordless disk/credential encryption.
<chadmed[m]>
the SEP will be able to do all of that eventually. and im going to hazard a guess and say (without having looked too much into the linux crypto device interface) that we can probably expose the SEP's services in such a way that existing userspace tooling can make use of them just fine
<chadmed[m]>
"reduced security" in Apple parlance just means that the "kernel" (m1n1) isnt signed by apple and iboot will not stop you from loading it. it doesnt mean that the system is *necessarily* less secure, just that apple isnt the guarantor of security
<ElvishJerricco[m]>
Right, I think "reduced security" means we're using self signed keys, right?
<ElvishJerricco[m]>
but yea we would need m1n1 and every boot stage up until Linux to support using the SEP if we can't provide a regular TPM device
<chadmed[m]>
which is fine because the two payload stages before the kernel are both m1n1
<ElvishJerricco[m]>
so that's m1n1, u-boot, and systemd-boot or grub that all need updates.
<ElvishJerricco[m]>
unless you just boot the kernel directly, but like boot loaders are good
<chadmed[m]>
the goal eventually is to have m1n1 stage 1 verify the integrity of m1n1 stage 2 which includes u-boot so once all that is settled we are safe up to that point
<ElvishJerricco[m]>
right but then u-boot has to verify the boot loader which has to verify the kernel
<ElvishJerricco[m]>
it's a general thing with the TPM. Every stage in the boot chain is "measured" into some of the PCRs in the TPM, including the platform firmware
<chadmed[m]>
yeah right well we will never be able to emulate that 1:1 because we dont have control over anything before m1n1
<ElvishJerricco[m]>
"measured" means that the stage is hashed and that hash is combined with the current state of the PCR, *that* is hashed, and stored in the PCR
<chadmed[m]>
but we do have implicit trust of it via the SEP and Apple signing it all
<chadmed[m]>
but i do see what youre saying about u-boot and beyond
<ElvishJerricco[m]>
Right but if the SEP implements some kind of measured boot of its own, then m1n1 would be measured into it and could continue the chain
<ElvishJerricco[m]>
it would mean each stage needing new code to measure the next one, but if the SEP has such a measurement feature that measures m1n1 then we should be good
<chadmed[m]>
well yes like i said peruse the apple docs
<chadmed[m]>
"reduced securitly" literally just means apple doesnt do the signing
<ElvishJerricco[m]>
(which is why I asked about m1n1 implementing a fTPM, so that the later stages could use existing code)
<chadmed[m]>
the SEP still signs the resultant MachO built from the raw m1n1 binary
<chadmed[m]>
its not self signed by any one developer, its signed by the machine and sepOS which has a trust direct from apple
<ElvishJerricco[m]>
ohh
<ElvishJerricco[m]>
that's... interesting
<chadmed[m]>
and to enable sepOS's signing, you need to assert physical presence AND authenticate with machine owner credentials
<chadmed[m]>
so not only is it signed by the hardware, you are also implicitly signing it as the user/owner of the machine
<ElvishJerricco[m]>
right
<chadmed[m]>
m1n1 will have payload checking etc eventually i believe, its one of the reasons the fat32 code was written in rust
<ElvishJerricco[m]>
The difference between secure boot and measured boot is significant though, and it's reasonable to want either one regardless of whether you want the other.
<ElvishJerricco[m]>
for instance on my steam deck i use them both so that my drive is encrypted with a TPM-bound key, and that way if secure boot settings or my boot chain is tampered with, it suspiciously falls back to a disk password prompt because the TPM won't unlock the key.
<sven>
Pretty sure SEP doesn’t do measured boot
<sven>
Or, well, not in the way you want it
<sven>
it does put its own firmware, etc. into some PCR-like mechanism used to derive keys
<sven>
but usually iboot directly loads the kernel and there are no additional boot stages
<chadmed[m]>
yeah i cant see how we would implement measured boot in the way the PC does without some weird m1n1 runtime services that pretend to be a TPM which doesnt sound pleasant to me
<sven>
ftpm usually requires something like el3 to at least pretend to be secure afaik
<ElvishJerricco[m]>
well and in that case we would be relying on m1n1 to measure itself, since it sounds like the SEP doesn't expose its final measurement state, which is no good
<chadmed[m]>
is there a real problem with that though? like i said the SEP wont even let you boot m1n1 unless it matches the one you installed via 1TR
<chadmed[m]>
so you cant really tamper with the MachO or stage2 and just expect to get around it
<chadmed[m]>
(i mean you can mess with stage2 right now but thats because stage1 cant talk to the SEP yet)
<ElvishJerricco[m]>
that's true. Though in that case it begs the question: Why even encrypt the disk with LUKS, if the boot chain can't be tampered with?
<chadmed[m]>
or rather, doesnt do the checks
<sven>
that doesn’t have anything to do with SEP
<sven>
m1n1 will likely never bring up the sep
<chadmed[m]>
yeah i misspoke
<sven>
FDE will eventually happen with SEp support such that you can’t brute force passwords
<sven>
(or steal keys or whatever)
<ElvishJerricco[m]>
even if you're worried about the physical NAND being ripped off, doesn't the SEP encrypt all actual data before sending it to NAND?
<chadmed[m]>
Yes the disk is encrypted at rest
<chadmed[m]>
For macos/apfs volumes at least
<ElvishJerricco[m]>
so then encrypting the drive in Linux is just completely pointless as long as you trust your login program.
<ElvishJerricco[m]>
(and have a secure boot style chain)
<sven>
uhhh…
<chadmed[m]>
No?
<chadmed[m]>
You could still gain access to a different OS on the disk and exfiltrate data
<chadmed[m]>
Physical presence assertion etc is just to boot 1TR
<ElvishJerricco[m]>
wait it doesn't require user authentication? Just physical presence?
<chadmed[m]>
The disk encryption the SEP/ANE do is transparent once the machine is running
<chadmed[m]>
Filevault requires auth
<chadmed[m]>
Please just read the docs
<chadmed[m]>
Apple explain this all rather succinctly
<chadmed[m]>
Non-FV disks are transparently encrypted so that once youve logged in to the machine the data is no longer garbled. You could gain access to a macos partition and exfil data from an ext4 root if that ext4 root is not itself encrypted on top of whatever the machine “automatically” does
<chadmed[m]>
If you turn on filevault then you have authentication at the machine level before the SEP will let you use the disk
<ElvishJerricco[m]>
yea but you have to boot and login to macOS somehow, which would require the user password, unless there's a way to enable booting a different installation of macOS without owner authentication, and I thought a feature of 1TR was to prevent that.
<chadmed[m]>
yes, of 1TR
<chadmed[m]>
in a normal, non-FV boot, the disk is unlocked by the SEP/ANE and everything past that point should be assumed to be in the clear
<chadmed[m]>
if you enable FV, that process is guarded by authentication
<ElvishJerricco[m]>
gotcha
<chadmed[m]>
the boot menu is not guarded by 1TR, you can just hold the button down and boot a different install
<chadmed[m]>
hell you can pick a different install from userspace if youve already captured that
<chadmed[m]>
not that it matters becuase unless you have FV enabled then the entire disk is transparent at runtime
<chadmed[m]>
so yes if your goal is a full chain of trust to linux userspace/FDE you are going to need LUKS
<chadmed[m]>
which thankfully already works fine by itself
<ElvishJerricco[m]>
but if your macOS side has FV enabled, then the whole disk is inaccessible once you boot until you do some kind of authentication
<chadmed[m]>
only the APFS container with FV enabled
<ElvishJerricco[m]>
got it
<ElvishJerricco[m]>
ok that makes sense
<chadmed[m]>
we've had a tonne of similar questions lately, maybe a wiki page with a diagram and a plain english explanation is warranted now
<ElvishJerricco[m]>
but assuming external drive booting is enabled, anyone can show up with a macOS usb drive and steal any unencrypted partitions
<ElvishJerricco[m]>
yea I probably should have moved this conversation out of the #asahi-dev room once it went from pontificating about measured boot and turned into encryption troubleshooting
<chadmed[m]>
im actually not sure on how that works but i think you can only select external disks for booting from Startup Disk, which requires either presence via touchid or a password
<chadmed[m]>
so again useless if theyve already captured your username and password
<ElvishJerricco[m]>
well yea but I'm assuming I'm pretty pwned no matter what if they've managed to steal my password in one OS
<chadmed[m]>
like if you havent blessed them from userspace they wont show up in the boot picker
<chadmed[m]>
exactly, which is why you should still be using LUKS and FileVault if you care about that sort of thing. it will prevent them from exfiltrating data from other installs if they pwn one
<ElvishJerricco[m]>
yea, but it seems like it means they do need to pwn at least one of them to even see the disk in the first place.
<chadmed[m]>
yeah it describes the design perfectly. its a common term in aviation safety
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
<ElvishJerricco[m]>
oh it is worth noting that without measured boot, then we can't use the SEP for passwordless disk unlock without exposing the same issue of another booted OS being able to access the data. Not really a huge deal because I think people are generally ok with needing separate passwords for booting vs logging in. But I will say that's it's been quite nice to have passwordless booting on my steam deck. And when I boot another OS that's
<ElvishJerricco[m]>
only been signed with MS secure boot keys instead of my own, it is unable to access my data because of measured boot and PCR 7.
SSJ_GZ has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
<sven>
I want to eventually type my password at most once a day (unless I reboot more often) and use Touch ID for the rest fwiw
<sven>
gonna take a while to get there unless someone picks up SEP
opticron has quit [Remote host closed the connection]
opticron has joined #asahi-dev
<Jamie[m]>
anyone know what it means for an interrupt to be Active vs. to be Enabled (in the context of Cortex-M3)?
<Jamie[m]>
oh got it, active is currently being executed
<Jamie[m]>
(or has been pre-empted by a higher priority one, not yet complete in any case)
MajorBiscuit has joined #asahi-dev
Major_Biscuit has joined #asahi-dev
MajorBiscuit has quit [Ping timeout: 480 seconds]
m5zs7k has quit [Ping timeout: 480 seconds]
m5zs7k has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
goldsoultheory has joined #asahi-dev
pjakobsson has quit [Remote host closed the connection]
millenialhacker has quit [Ping timeout: 480 seconds]
compassion has quit [Quit: Ping timeout (120 seconds)]
compassion has joined #asahi-dev
millenialhacker has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
nicolas17 has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
goldsoultheory has joined #asahi-dev
goldsoultheory has quit []
goldsoultheory has joined #asahi-dev
<amarioguy>
i wouldn't mind being the person doing sep stuff but iirc i did discuss this a few months back the general consensus was "wait for APFS support in linux to be done"
<amarioguy>
(as SEP support would imply xARTs working and those are in the APFS containers iirc)
<amarioguy>
unless you want someone more seasoned to be doing sep stuff lol totally understand that but i do genuinely think i could help out here
<nicolas17>
did you manage to log SEP requests yet?
<nicolas17>
(er, that's plural 'you')
<sven>
reverse engineering the sep doesn’t require Linux apfs support
<sven>
there’s an initial tracer inside m1n1 that i wrote months ago
<sven>
and some newer version I should push
<nicolas17>
last I heard, there were some tight timeouts and trying to intercept SEP mailbox traffic with m1n1 easily led to errors
<nicolas17>
but maybe you got past that since then
<sven>
you can probably get away with just finding the xART blocks inside the partition and manipulating them directly
<sven>
that file has a constant nice
<sven>
*size
<sven>
That won’t do for a real driver(tm) ofc but it should be fine for reverse engineering
<sven>
nicolas17: dunno, it’s always worked for me
<nicolas17>
huh, interesting
<sven>
it’ll just spam a lot if you trace everything
<amarioguy>
alright, i'll comb over the tracer once you push the newer one
<sven>
but there’s an allowlist for that
<amarioguy>
and see if i can get it to be "fully featured" in a sense
<nicolas17>
also
<nicolas17>
I have an M2 MBP now
<sven>
my first step was going to be that secure key store and then FileVault and then Touch ID
<nicolas17>
so I guess I can get nerdsniped into helping with asahi now
<sven>
but then I started shaving yaks with nvme and somehow that led to thunderbolt then
<nicolas17>
has anyone looked into touchbar yet? the fact that I don't even get static Fn keys is really annoying so maybe that's what I'll help with first :p
<sven>
amarioguy: remind this evening, I’ll push it after $work
<sven>
Or grep through the -re backlog, I think I pasted it at some point
<sven>
iirc it had all function names for the first level key store ipc
<sven>
but there was an second layer ipc protocol as well
<amarioguy>
sven: will do
<nicolas17>
is there a wiki page of who's working on what? so that potential contributors can know what features got 0 attention so far, or else who to talk to that already looked into it?
<sven>
not really, there’s a wiki page with small tasks somewhere
<sven>
otherwise it’s easier to just ask here
<sven>
we’re all too lazy to maintain wiki pages ;)
<nicolas17>
maybe I could make such a wiki page no no dammit nicolas stop volunteering yourself to things
<amarioguy>
btw whitespace fixes (tm) headed to mailing list today so that's fun
<_jannau_>
nicolas17: I think someone started looking ages ago (i.e. m1 macbook pro). touch input appears to be similar to the trackpad. I'm not sure anyone has started looking into the display controller for the touchbar
<sven>
amarioguy: fwiw, I always look at the final patch in vim, that usually catches white space issues
<nicolas17>
I'm still at Akademy, all I did so far was basic macOS setup and basic asahi install, I didn't even get to try compile speed yet :D
<sven>
otherwise you can hilight lines in vs code and see space vs tab but that’s annoying
<amarioguy>
ofc lol i caught them on my side with nano
<nicolas17>
oh someone on the church of nano
<amarioguy>
i use it because it's simple really lol
<amarioguy>
i know how vim can make workflow really good and i should get to learning it at some point but i'm more comfortable with GUI based text editors for now, but when i need a command line editor it's for things GUI editors don't catch
<amarioguy>
hence nano
<amarioguy>
ultimately i just use what works for me :)
<chadmed[m]>
ive been trying to maintain the wiki but theres too many things in flight so i just add stuff ot feature support when i see stuff get merged
liszt has quit [Read error: No route to host]
jonmasters has quit [Read error: No route to host]
linxz has quit [Read error: No route to host]
jonmasters has joined #asahi-dev
Manouchehri has quit [Read error: No route to host]
msteffen has quit [Read error: No route to host]
nathanchance has quit [Read error: No route to host]
cptcobalt has quit [Write error: connection closed]
snek has quit [Read error: No route to host]
Chainsaw has quit [Write error: connection closed]
msteffen has joined #asahi-dev
nathanchance has joined #asahi-dev
arnd has quit [Read error: No route to host]
Chainsaw has joined #asahi-dev
snek has joined #asahi-dev
steev has quit [Read error: No route to host]
liszt has joined #asahi-dev
kov has joined #asahi-dev
cptcobalt has joined #asahi-dev
linxz has joined #asahi-dev
arnd has joined #asahi-dev
steev has joined #asahi-dev
Manouchehri has joined #asahi-dev
<_jannau_>
nicolas17: tracing the rouch-bar is a little annoying since apple uses an coprocessor (sio) to emulate DMA. best starting point is the keyboard tracer but I stopped looking into it once it was good enough for the keyboard/trackpad
<nicolas17>
hm
<nicolas17>
I have very little experience with hardware, maybe I should let someone else do that and take care of the userspace side >.>
<nicolas17>
anyway, late for lunch ->
nicolas17 has quit [Quit: lunch]
<sven>
well, there's only one way to get experience ;)
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
<millenialhacker>
I learned a lot with ISP tracer, so I confirm even without previous experience you can achieve something if are you willing to sell your soul to demon, excuse... to spend enough time on it :)
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
nicolas17 has joined #asahi-dev
___nick___ has joined #asahi-dev
nicolas17 has quit [Ping timeout: 480 seconds]
kensan has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
___nick___ has quit []
___nick___ has joined #asahi-dev
___nick___ has quit []
nicolas17 has joined #asahi-dev
___nick___ has joined #asahi-dev
nicolas17 has quit [Ping timeout: 480 seconds]
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
derzahl has joined #asahi-dev
nirya has joined #asahi-dev
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
nicolas17 has joined #asahi-dev
goldsoultheory has joined #asahi-dev
<ChaosPrincess>
how does one get the clock address from clock-gates property in the adt?
<sven>
there’s a pmgr dump tool somewhere in m1n1
<sven>
the clock-gates entries refer to an array inside the pmgr node that uses two or so other arrays to encode the address
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
gladiac has joined #asahi-dev
<ChaosPrincess>
so, if adt refers to id 37, that node has name = u'FPWM1' in adt dump then the clock i want is ps_fpwm1 ?
<sven>
sounds about right
<sven>
that thing is more than a clock as m arcan figured out fwiw
<sven>
apple sometimes also uses the same node index inside power-gates
<j`ey>
blazra[m]: seems like ChaosPrincess wants to look at pwm now, are you going to take the code forward or should I send it to them?
<sven>
the ADT is just very misleading at times :(
<bluetail>
dang. I wanted to contribute to the asahi installer script. Worked most of the day and now I'm tired. How can you guys dual wield life+work+duties and asahi on top? I cannot even do free time activities. Gotta hit the sack now. Nighties
<nicolas17>
"simple, I have no life"
<ChaosPrincess>
bluetail: the trick is that sleep is overrated :P
<millenialhacker>
+1 to bluetail question.
<sven>
sleep is for the weak ofc!
<sven>
seriously though, having a relaxed job that never requires overtime and also doesn't have overhead like perf/promotion season that friends seem to complain about right now helps ;)
nicolas17 has quit [Ping timeout: 480 seconds]
<millenialhacker>
perf/promotion is a peace-of-mind killer right now for me
<millenialhacker>
I really get estressed on Oct/April because of it
<sven>
the tradeoff ofc is that i don't make a crazy FANG-like salary *shrug*
<j`ey>
sven: that's why you only have 2 m1s!
<sven>
:D
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<blazra[m]>
<j`ey> "blazra: seems like ChaosPrincess..." <- I have done some little changes like filling in the pwm_get_state() function, adding the bit defines and I think you had the enable bit wrong? I can share it later when I will get to my PC. However, sure, let ChaosPrincess have it - I haven't got much time recently. Sorry.
<millenialhacker>
well... I work for a FAANG and can only afford one m1 (I'm sole provider for a big family) xD
<j`ey>
weird if the enable bit was wrong, it was working for me, but maybe i mucked something up when squashing the code
<blazra[m]>
I have the 14" MBP. Is there a chance it is different?
<j`ey>
shouldnt be, I'll look when you send it late
<j`ey>
r
<blazra[m]>
I think you had bit(1) as enable and for me it's bit(0). But maybe I am wrong...
goldsoultheory has joined #asahi-dev
gladiac has quit [Quit: k thx bye]
genhack[m] has joined #asahi-dev
derzahl has quit [Ping timeout: 480 seconds]
Major_Biscuit has quit [Ping timeout: 480 seconds]
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
goldsoultheory has joined #asahi-dev
derzahl has joined #asahi-dev
goldsoultheory has quit []
goldsoultheory has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
goldsoultheory has joined #asahi-dev
goldsoultheory has quit []
goldsoultheory has joined #asahi-dev
derzahl has quit [Read error: No route to host]
derzahl has joined #asahi-dev
derzahlah has joined #asahi-dev
derzahl has quit [Read error: Connection reset by peer]
<jannau>
the latest ventura dev beta prints syslog to the display during boot
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
goldsoultheory has joined #asahi-dev
goldsoultheory has quit []
goldsoultheory has joined #asahi-dev
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
derzahlah has quit [Ping timeout: 480 seconds]
derzahlah has joined #asahi-dev
___nick___ has quit [Ping timeout: 480 seconds]
illenialhackerm has joined #asahi-dev
millenialhacker has quit [Ping timeout: 480 seconds]
derzahlah has quit [Ping timeout: 480 seconds]
thelounge7571340 has joined #asahi-dev
thelounge7571340 has quit [Remote host closed the connection]
jluthra has quit [Remote host closed the connection]
jluthra has joined #asahi-dev
derzahlah has joined #asahi-dev
goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]