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
brett_ has quit [Quit: Connection closed for inactivity]
odmir has quit [Remote host closed the connection]
marvin24 has quit [Ping timeout: 245 seconds]
marvin24 has joined #asahi
<Jaehyuk[m]>
Is it possible to execute AMX instruction on current version of asahi linux m1?
<Jaehyuk[m]>
cause I continuousl enocunters illegal instruction fault even though I start and end with the special instruction of amx
<Jaehyuk[m]>
Also, Is there any reference apple device tree blob of M1 mac mini?
<Jaehyuk[m]>
I'd like to explore what is in the ADT but doesn't know how to retrieve it.
<jannau>
AMX is disabled by default. it would need OS support to save and restore register contents on task switch. support has not been written yet
<jannau>
not sure there is a reference ADT but the ADT can be retrieved from m1ni with proxyclient. see how adt_base and adt_size are used in chainload.py
<pipcet[m]>
I don't think a reference ADT is a good idea. Legally, it's probably copyrighted. It also contains private information like WiFi keys.
VinDuv has joined #asahi
<Jaehyuk[m]>
jannau: Aha, I see, I guess that AMX could be supported by changing some bits in the MSR registers but would you happen to know which register controls the AMX?
<Jaehyuk[m]>
I cannot find them on the documents describing apple registers
<Jaehyuk[m]>
pipcet: Ah, I didn't know that it contains private information. Thanks for letting me know that
<sven>
someone should write a dump_dt.py for proxyclient. ba.devtree contains the virtual base addr for the ADT
<sven>
let's see if i can find that magic invocation to just dump it :D
<sven>
it might need some bit in that hacr to enable reading/writing to some msrs
<sven>
and then some bit in those msrs to enable it
<sven>
actualy... just ignore me
* sven
just woke up
<Jaehyuk[m]>
@sven haha thaks a lot. BTW, I heard that M1 supports Intel's memory oredring and that can be enabled by chaning some bits in the rosetta. Would you happen to know which register manage it also?
<sven>
coffee first before i say more stupid things :-)
<sven>
it should be called TSO
<Jaehyuk[m]>
:P
<Jaehyuk[m]>
Ah I just found it on the wiki doc
<Jaehyuk[m]>
SYS_APL_HID9_EL1
<Jaehyuk[m]>
thanks :D!
<sven>
:)
VinDuv has quit [Quit: Leaving.]
jeffmiw_ has joined #asahi
vimal has joined #asahi
jeffmiw_ has quit [Ping timeout: 240 seconds]
austriancoder has joined #asahi
austriancoder has quit [Changing host]
vimal has quit [Quit: Leaving]
vimal has joined #asahi
vimal has quit [Remote host closed the connection]
raster has joined #asahi
vimal has joined #asahi
maknho_ has quit [Ping timeout: 252 seconds]
maknho_ has joined #asahi
doof has quit [Ping timeout: 258 seconds]
austriancoder has quit [Ping timeout: 245 seconds]
arnd has quit [Read error: Connection reset by peer]
WhyNotHugo has quit [Read error: Connection reset by peer]
Guest45273 has quit [Read error: Connection reset by peer]
<pipcet[m]>
marcan: that sounds like a good strategy to me!
<pipcet[m]>
kettenis: I think it might be a good idea to use t8103 (or a non-code name) if that's the only hardware we tested our driver on. We might be relying on unspec'd behavior that differs between those devices.
<pipcet[m]>
(not likely to matter if apple won't release specs, though)
<kettenis1>
that is why both "t8103" and "t8101" are present
<kettenis1>
personally I don't care about the older SoCs
<pipcet[m]>
so use t8103 in the driver, have both in the DT?
<sven>
i think AIC uses "apple,t8103-aic", "apple,aic"
jeffmiw has quit [Remote host closed the connection]
choozy has joined #asahi
maknho__ has joined #asahi
maknho_ has quit [Ping timeout: 245 seconds]
odmir_ has joined #asahi
odmir_ has quit [Remote host closed the connection]
odmir_ has joined #asahi
odmir has quit [Ping timeout: 240 seconds]
odmir_ has quit [Remote host closed the connection]
odmir has joined #asahi
KindTwo has joined #asahi
KindOne has quit [Ping timeout: 240 seconds]
KindTwo is now known as KindOne
odmir has quit [Remote host closed the connection]
odmir has quit [Remote host closed the connection]
odmir_ has joined #asahi
klaus has quit [Ping timeout: 240 seconds]
TheJollyRoger has quit [Remote host closed the connection]
TheJollyRoger has joined #asahi
mrkajetanp has joined #asahi
<pipcet[m]>
looks like iBoot doesn't obey large alignments. My kernel isn't reliably aligned to 2 MiB, so it fails to boot, but it's just a few bits that are missing, so by sheer luck it will boot once in a while...