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-offtopic | Keep things on topic | Logs: https://alx.sh/l/asahi
<j`ey> krbtgt: yeah?
acelogic has joined #asahi
<j`ey> seems like the main issue with that is the DT. since it's a different format, and upstream might not want an 'ADT' parser
<jn__> different binary encoding, and also different bindings
<j`ey> and also that requires you to re-kmutil on kernel updates
<Shiz> yeah, the problem is not the binary encoding as much as the bindings
<Shiz> the format is trivial, significantly easier than FDTs
<Shiz> the semantics seem to differ a bit though, on top of being unstable as noted in the article
<Shiz> not sure if upstream wants to add generic names like `samsung,uart-1` to a very specific samsung uart driver
<jn__> even worse, i think i've seen "uart-1,samsung"
<jn__> and a typo (#address-cels)
acelogic has quit [Ping timeout: 246 seconds]
<Shiz> yeah, that's the one
dstzd has quit [Quit: ZNC - https://znc.in]
dstzd has joined #asahi
<yrlf> or some stuff like literal mount options for filesystem partitions
<yrlf> I've not seen a lot of linux device tree stuff, but those mount options also seem wierd
qyousef has quit [Ping timeout: 256 seconds]
qyousef has joined #asahi
qyousef1 has joined #asahi
qyousef has quit [Ping timeout: 272 seconds]
qyousef1 has quit [Client Quit]
qyousef has joined #asahi
qyousef has quit [Client Quit]
qyousef has joined #asahi
raster has quit [Quit: Gettin' stinky!]
Tokamak_ has joined #asahi
Tokamak has quit [Read error: Connection reset by peer]
tbodt has quit [Ping timeout: 256 seconds]
tbodt has joined #asahi
genevino has quit [Remote host closed the connection]
genevino has joined #asahi
_whitelogger has joined #asahi
_whitelogger has joined #asahi
Tokamak_ has quit [Ping timeout: 246 seconds]
Tokamak has joined #asahi
_whitelogger has joined #asahi
<brinly> yeah, android vendors used to do that
<opticron> and one of our employees got tired of our shit, bought <item> and requested source
<opticron> the CTO just about shat his pants
<brinly> and briefly that vendor would mail you a cd
<opticron> mostly because he was not ready for *any* source request across multiple product lines
<brinly> @opticron did he distribute it publicly?
<opticron> no and no
<brinly> lol
<opticron> at least, I don't think the employee did
<opticron> but eventually, out of date tarballs were posted
Tokamak_ has joined #asahi
<marcan> brinly: the GPL requires a source *offer*
<brinly> Ah, I didn’t think they needed to mention the offer.
<marcan> they do :)
<marcan> also, you know, actually mentioning the license etc is still a requirement
<marcan> so it needs an accompanying README at the *very* least
<brinly> I didn’t try download that bundle
<marcan> it's one binary file
<brinly> But I presume it has some user mode bits too?
<brinly> Just a kernel image?
<tbodt> I bet it has an initrd or something
Tokamak has quit [Ping timeout: 264 seconds]
<marcan> 14943571 0xE40553 Unix path: /home/skylark/nsa/corellium/psc/linux-5.4.14/net/mac80211/chan.c
<marcan> NSA👀
Tokamak_ has quit [Ping timeout: 256 seconds]
<maximus64> haha nice find
<marcan> it has an initramfs
<marcan> with busybox, so that's another GPL violation, and one that has in fact been enforced in court :)
<marcan> there is also this
<marcan> 2020-03-07 01:48:25 ..... 583163 583163 lib/firmware/brcm/brcmfmac43452-pcie.bin
<marcan> which, unless they magically got redistribution permission from broadcom.... again, copyright violation
<brinly> I’d be curious on the kernel compile time
<Shiz> who cares about em, we have our own things to worry about :p
<marcan> also: brcmfmac43452-pcie.bin
<marcan> er, sorry, same thing
<marcan> but yeah I assume that was ripped from macos? I can't find that file anywhere and it's not just a rename
<bkero> good ol' binwalk
<bkero> marcan: are the licenses present on the initramfs?
Tokamak has joined #asahi
<bkero> 14934845 0xE3E33D Neighborly text, "Neighbor" # haha classic
<marcan> bkero: not that I can see
<tbodt> at what offset is the initrd? I wanna try dumping it
<bkero> lol there's some mac address of some random card in there
<bkero> tbodt: binwalk -e should dump the whole thing
<Shiz> if you want to discuss this at all, might be useful to do it in -re as per the policy
<Shiz> license violations or not
krzywix has joined #asahi
<bkero> If this is being white-roomed none of us should have even looked at that, let alone discussed it.
<tbodt> kernel source will probably be posted tomorrow
<davidrysk[m]> Overall this is awfully sloppy of them.
<Shiz> well, i agree, nobody should have :)
<Shiz> this is just drama stuff
<winocm> good thing i still have two bottles of asahi left somewhere (though i hate the taste of beer)
<marcan> I'm not interesting in actually REing this, but yeah, I guess technically "binwalk" output should've gone in -re, as much as this ought to be GPL code
<Shiz> winocm: god i wish that was me
<bkero> mmmmbeer
<davidrysk[m]> I’d have expected Corellium of all companies to not play quite this fast and loose with copyright especially at this point but... idk
<marcan> in DM Chris told me "we won against apple, so that means all of our code is clean"
<clayfreeman> After catching up with all this... yeah, it's amateur hour for them to behave this way.
<marcan> that's... not how it works, but he seems to believe it is
<clayfreeman> I'm even surprised at qwerty - you don't often get second chances at getting licensing right; it's almost an art in perfection, which I bet is exactly why Asahi has the policies that are in place
<marcan> exactly
<marcan> *technically* Chris has just lost all rights to ever distribute Linux
<marcan> ... it's just that GPL copyright owners don't usually enforce that
<marcan> but the text of the license does indeed say there are no automatic second chances
<davidrysk[m]> That was one of the things that GPLv3 fixed. Also a court might not necessarily accept that argument, but then again a different court might accept it.
<MaxLeiter> Having been down similar roads, can I suggest just not talking about Corellium unless necessary? Will just breed negative attitudes
<clayfreeman> It's kinda hard to do that with them throwing shade, MaxLeiter
<marcan> davidrysk[m]: right, the goal of our policies is to ensure there is no *question* that a court would ever have to decide
<marcan> it's not just about license compliance, it's about making it *plain and clear* enough that no copyright was violated, even to non-lawyers
<marcan> because nobody has the money to ask lawyers and go to court :)
<marcan> so we need to go above and beyond what may be the minimum legal requirements
<davidrysk[m]> Licensing and copyright is not the kind of thing to play fast and loose with and I kinda fear that their legal troubles are not yet over (in other words I wholly expect to see an appeal, and fair use for their uses might not apply to the uses here, and other cans of worms that require lawyers)
<tbodt> my understanding of their lawsuit is it was partially dismissed and some of the tougher claims still stand
<_alice> I don't understand why they see marcan being careful as a personal attack on themselves
<_alice> careful wrt licensing and auditability
<tbodt> they still have to defend a DMCA claim for circumventing the boot chain "drm"
<davidrysk[m]> What I’m saying is that I expect Apple to appeal the dismissal to a higher court and that could go anywhere
<davidrysk[m]> There is that as well, separately
<davidrysk[m]> There is also the fact that fair use isn’t exactly well defined. Too many unknowns, too difficult to quantify the risk, better to avoid the risk by setting a strict policy (which is what marcan has done)
<marcan> bbl, lunchtime
<clayfreeman> I hope I can catch one of these work streams live - I've had a ton of fun watching the last two
sirn has quit [Ping timeout: 260 seconds]
sirn has joined #asahi
aratuk has joined #asahi
<aratuk> Marcan, you need theme music for your YT streams. Maybe the free music archive would be an appropriate source.
<JTL> he's already adressed *why* doesn't mux music into his streams
<JTL> aratuk: That being said, maybe a list of music he likes would be interesting? :p
aratuk has quit [Ping timeout: 246 seconds]
Tokamak has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<clayfreeman> yeah, I prefer no music
<clayfreeman> allows me to put on my own
<MaxLeiter> A web radio with an IRC bot to control is always fun
acelogic has joined #asahi
Necrosporus has quit [Killed (verne.freenode.net (Nickname regained by services))]
Necrosporus has joined #asahi
Swant has quit [Read error: Connection reset by peer]
genevino has quit [Ping timeout: 256 seconds]
Esmil has quit [Ping timeout: 256 seconds]
Esmil has joined #asahi
Swant has joined #asahi
scubasteve has quit [Ping timeout: 256 seconds]
genevino has joined #asahi
scubasteve has joined #asahi
CalumK has joined #asahi
_whitelogger has joined #asahi
<Necrosporus> So somebody made M1 run Linux before marcan did?
<Necrosporus> But that's just kernel and initrd with basic shell
<jix> also, given their latest released sources (which do not support m1 yet), likely in a state that would require significant work to be upstreamable, possibly more work overall than the approach marcan is taking
<jix> this is based on their linux on jailbroken iphone tech demo they've been working on for quite some time, so they did have a head start time wise, but since that in itself wasn't something that upstream would accept they didn't pay any attention to linux's upstream requirements
<Necrosporus> Is it different from Sandcastle?
<Necrosporus> $ file linux.m1.macho
<Necrosporus> linux.m1.macho: Mach-O 64-bit arm64 filetype=12, flags:<|DYLDLINK>
<Necrosporus> I'm wondering if this thing can run on my arm64 SBC
<Necrosporus> probably not
<Necrosporus> Unless I pack the initrd of it with kernel from something like armbian
<Necrosporus> It has dropbear host keys
<Necrosporus> I'm wondering where did they come from
<Necrosporus> etc/issue says Project Sandcastle brought to you by Corellium team
* rwhitby wonders if all this talk about something that's not Asahi Linux should be in -offtopic ? ;-)
<marcan> Necrosporus: they've been working on this for Ax SoCs for over a year now, so of course porting that work to M1 is easy compared to doing it from scratch
<marcan> but nothing they've released so far is upstreamable, so at this point their work is not useful
<Necrosporus> So, does this binary run anyway?
<marcan> no idea, so far on first look there is a probable macOS copyright violation as well as the GPL issues, so I won't look into it further
<Necrosporus> How it copyright violation of macOS? Because of broadcom fw?
<marcan> yes
<Necrosporus> I guess it's mean to be run from iBoot directly?
<marcan> I think so
<Necrosporus> And is it illegal to try to run this thing?
<marcan> the macOS copyvio thing is a hypothesis based on the fact that I have not found that firmware anywhere else
<marcan> I don't know
<Necrosporus> Or only to distribute it?
<marcan> I am not a lawyer
<marcan> but at this point, I will wait for their upstream efforts and continue on with our project independently
<marcan> they've said they will upstream their stuff, and I eagerly await that effort :)
<marcan> until then, there is nothing for us to gain by following their announcements/releases, so I will continue with my original development approach
VinDuv has joined #asahi
<Necrosporus> It makes sense. But regarding iPhone linux, I'm wondering if it's going to grow to stage to be actually useable as real main OS rather than tech demo, and perhaps it could make iPhone as good as Nokia N9
<jix> Necrosporus: that's getting off topic here though
rwhitby has quit [Ping timeout: 264 seconds]
rwhitby has joined #asahi
<Necrosporus> ok
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
Calchan has quit [Ping timeout: 246 seconds]
Calchan has joined #asahi
dstzd has quit [Quit: ZNC - https://znc.in]
dstzd has joined #asahi
dstzd has quit [Client Quit]
dstzd has joined #asahi
dstzd has quit [Client Quit]
dstzd has joined #asahi
dstzd has quit [Client Quit]
dstzd has joined #asahi
acelogic has quit [Ping timeout: 246 seconds]
dstzd has quit [Read error: Connection reset by peer]
dstzd_ has joined #asahi
dstzd_ is now known as dstzd
dstzd_ has joined #asahi
dstzd has quit [Ping timeout: 264 seconds]
dstzd_ is now known as dstzd
julian[m]3 has joined #asahi
CalumK has joined #asahi
raster has joined #asahi
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
qyousef1 has joined #asahi
qyousef1 has quit [Client Quit]
qyousef has quit [Ping timeout: 256 seconds]
CalumK has joined #asahi
furkan has joined #asahi
qyousef has joined #asahi
qyousef has quit [Client Quit]
qyousef has joined #asahi
aimileus has quit [Quit: Connection closed for inactivity]
mogery has joined #asahi
frode_0xa has joined #asahi
frode_0xa has quit [Quit: leaving]
furkan has quit [Remote host closed the connection]
notafile has joined #asahi
<stemnic> Regarding the topic of GPL compliance I’m curious, currently experiencing what I have been told is malicious compliance where a hardware vendor which has shipped a embedded Linux distribution (Yocto) insists on shipping a CD instead of providing me with a download link with their modified Linux kernel and have used around 3 months now to comply
<stemnic> (I’m still waiting for it). Can this then be regarded as a violation since there is not a clear “offer” of their source code modification or is still valid even though I need to nag their support for months :P
<jn__> AFAIK offering the source code on a physical medium (and even collecting a reasonable fee for it) is valid for GPL compliance
<jn__> i don't know about timing, i.e. how fast a vendor has to respond
<opticron> I don't remember there being any time limit
<jn__> stemnic: regarding a written offer: how did you learn about the use of GPL code in their product?
<jn__> (in any case, perhaps the Software Freedom Conservancy can probably give more accurate advice than i can)
<stemnic> jn__: opened the product up, saw some uart ports and when connecting to it. From the output I could see that it had a uboot bootloader and used yocto
<jn__> stemnic: ok, that does sound like no source code offer, which AFAIK is non-compliant
<jn__> (assuming the product didn't ship with the source code itself, either)
<jn__> errr, s/perhaps // in my comment above
<stemnic> jn__: Nop, not any mentions on their website either that I could fine. I'm surprised though if this is not compliant, that would mean that many hardware vendors are doing the same since i have not experienced many who openly give out their modified GPLv2 source code
<stemnic> ¯\_(ツ)_/¯ maybe it's just the fact that many play "lose" with licencing
<dottedmag> I have bought a Bosch oven and learned that it runs Linux reading the list of software licenses / copyrights enclosed, but that's probably for #asahi-offtopic
<jn__> i've heard often enough that people had inquire rather persistently before they got gpl compliance out of a vendor
<jn__> this whole discussion is probably better done in -offtopic
<stemnic> Yeah your right, can move it over there
marvin24 has quit [Ping timeout: 240 seconds]
marvin24 has joined #asahi
superk has joined #asahi
dontwasteusernam has joined #asahi
stblassitude has joined #asahi
jjanzic has joined #asahi
dontwasteusernam has left #asahi ["User left"]
Bublik has quit [Ping timeout: 246 seconds]
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
mogeryy has joined #asahi
Bublik has joined #asahi
Calchan has quit [Ping timeout: 256 seconds]
Esmil has quit [Ping timeout: 256 seconds]
Esmil has joined #asahi
mogery has quit [Ping timeout: 256 seconds]
scubasteve has quit [Ping timeout: 256 seconds]
Calchan has joined #asahi
mogeryy is now known as mogery
scubasteve has joined #asahi
aratuk has joined #asahi
aratuk has quit [Remote host closed the connection]
aratuk has joined #asahi
aratuk has quit [Remote host closed the connection]
Tokamak has joined #asahi
Tokamak has quit [Ping timeout: 246 seconds]
CalumK has joined #asahi
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
modwizcode has joined #asahi
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
xarkes has joined #asahi
ch has joined #asahi
q3k|m has quit [Ping timeout: 264 seconds]
q3k|m has joined #asahi
raster has quit [Quit: Gettin' stinky!]
shaunren has joined #asahi
shaunren has quit [Quit: Quit]
Baughn has quit [Quit: ZNC 1.6.2+deb1 - http://znc.in]
BaughnLogBot has joined #asahi
crafteck[m] has joined #asahi
CalumK has joined #asahi
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<flying_sausages> yo what's this thing corellium is claiming to have achieved?
modwizcode_ has joined #asahi
<flying_sausages> did they just make their own VM layer or something
<tbodt> no it is linux running on m1
<tbodt> it's a port of their previous project sandcastle tech demo
<flying_sausages> is there a summary of what is referred to as the drama?
<flying_sausages> civil and all
modwizcode has quit [Ping timeout: 256 seconds]
<flying_sausages> just trying to make heads and tails
<j`ey> just some heated messages back and forth, nothing worth looking over
modwizcode_ is now known as modwizcode
<flying_sausages> fair enough
<j`ey> they have a fork, linux-sandcastle, that was for running linux on iphones, so it was some kinda port of that to m1
<flying_sausages> mm ic
CalumK has joined #asahi
CalumK has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
CalumK has joined #asahi
CalumK has quit [Client Quit]
raster has joined #asahi
acelogic has joined #asahi
ncason has joined #asahi
merbanan has quit [Ping timeout: 240 seconds]
merbanan has joined #asahi
mogery has quit [Quit: Leaving]
ni291187 has joined #asahi
ni291187 has left #asahi [#asahi]
ewlsh[m] has joined #asahi
raster has quit [Quit: Gettin' stinky!]
irl25519 has joined #asahi
irl25519 has quit [Quit: irl25519]
irl25519 has joined #asahi
TheJollyRoger has quit [*.net *.split]
DarthCloud has quit [*.net *.split]
DarthCloud has joined #asahi
TheJollyRoger has joined #asahi
irl25519 has quit [Quit: irl25519]
raster has joined #asahi
yrlf has quit [Quit: The Lounge - https://thelounge.chat]
yrlf has joined #asahi
yrlf2 has joined #asahi
Esmil has quit [Ping timeout: 256 seconds]
Esmil has joined #asahi
raster has quit [Ping timeout: 256 seconds]
scubasteve has quit [Ping timeout: 256 seconds]
yrlf has quit [Read error: Connection reset by peer]
raster has joined #asahi
scubasteve has joined #asahi
yrlf2 is now known as yrlf
raster has quit [Quit: Gettin' stinky!]