00:45 <chadmed> j_ey: marcan's desktop colours are just the default breeze dark theme that comes boxed with plasma (at least, it does on gentoo)

00:46 <chadmed> you can make konsole follow that too, but pretty sure he's just got white on black for that

06:15 <rkjnsn[m]> Another random question: What kind of cryptography acceleration does the M1 provide? E.g., are the ARM cryptographic extensions implemented, or can acceleration only be used via the SEP?

06:16 <rkjnsn[m]> Is it safe to assume the accelerator blocks are designed to be able to keep up with the new, faster SSD storage?

06:45 <sven> no idea if it has the arm crypto extensions but it's safe to assume that whatever acceleration it has is enough to keep up

06:55 <sorear> it may be hard for anyone outside of apple's export control division to give you a complete list of cryptographic accelerators in the M1 - they

06:55 <sorear> 're quite small these days and don't _need_ to be centralized

06:56 <sven> there's at least an aes engine for the main SoC and another aes engine (that we can't and dont' need to access) in the secure enclave

06:56 <sven> but i don't know how fast those are. the one inside the secure enclave might be deliberately slow to make brute forcing harder

06:57 <sven> but yeah.. there might be many more of these things we don't know about

06:57 <sven> i know that ANS/nvme can also do aes. i don't know if that's hardware accelerated though

07:01 <sorear> i'm imagining that there's separate datapath cryptography in all 8 cores, plus PAC (which is likely distinct since it is connected to integer registers and not NEON), then there's the armv8 microcontroller cores they use which might also have PAC and/or simd cryptography. then we've got the ANS cryptography and whoever is responsible for HDCP

09:13 <rkjnsn[m]> Thanks! I'm specifically interested in disk encryption for the internal SSD under Linux, and whether it will be achievable without noticeable performance loss.

09:17 <dottedmag> Well, given how fast M1 is, it's probably achievable without noticeable performance loss even without acceleration.

09:23 <arnd> rkjnsn[m]: 'lscpu' shows "Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 asimddp sha512 asimdfhm dit uscat ilrcpc flagm sb dcpodp flagm2 frint", so both the CRYPTO_AES_ARM64_CE ("aes") and CRYPTO_AES_ARM64_NEON_BLK ("asimd") variants of the kernel's AES code should work, and it will pick the fastest one at runtime

09:27 <rkjnsn[m]> Nice. Thanks!

15:01 <nsklaus_> https://www.phoronix.com/scan.php?page=news_item&px=Apple-PCIe-For-Linux-5.16

15:01 <nsklaus_> :)

15:02 <nsklaus_> it's nice to see phoronix reporting on the progresses done by asahi team

15:20 <opticron> I'm looking forward to the bringup on the new MBP next week

15:20 <opticron> because that's almost exactly the model I'm looking to buy

16:12 <marcan> nsklaus_: and he even managed to not say "but gpu will take a long time" this time!

16:57 <kettenis> with these new models shipping soon, isn't there a mac os release that has the device trees already?

17:03 <nsklaus_> marcan: :)

18:30 <marcan> kettenis: there is, 12.0 RC

18:30 <marcan> https://updates.cdn-apple.com/2021FCSFall/fullrestores/071-78663/75110021-A2D3-44BB-B90F-3AC39F7A4F00/UniversalMac_12.0.1_21A558_Restore.ipsw

18:30 <marcan> t6000/t6001

18:31 <marcan> Firmware/agx/armfw_g13g.im4p Firmware/agx/armfw_g13x.im4p

18:31 <marcan> g13x, eh

18:31 <marcan> <ZipInfo filename='kernelcache.release.mac13g' compress_type=deflate filemode='-rw-r--r--' file_size=24181531 compress_size=24142240>

18:31 <marcan> <ZipInfo filename='kernelcache.release.mac13j' compress_type=deflate filemode='-rw-r--r--' file_size=24271373 compress_size=24233117>

18:31 <marcan> so they're *still* doing separate kernels for each SoC gen, lol

18:32 <marcan> linux at least got *that* figured out

18:32 <marcan> and it looks like the device models are j314c and j314s / j316c and j316s

18:32 <marcan> Firmware/ave/AppleAVE2FW_H13.im4p Firmware/ave/AppleAVE2FW_H13C.im4p Firmware/ave/AppleAVE2FW_H13S.im4p

18:35 <j_ey> marcan: wait, whats the mac13g and 13j then?

18:36 <dottedmag> marcan: It's not like there is a thousand of Apple boards to build kernels for, so for them it's probably a very low priority

18:36 <j_ey> dottedmag: cuts down on kernel builds though

18:36 <j_ey> testing etc

18:37 <marcan> also they have to install *all* kernels on every system to make the installs cross-bootable

18:37 <marcan> which eventually kinda bloats preboot containers

18:37 <marcan> j_ey: 13g is m1, 13j is m1 pro/max

18:38 <marcan> j for "jade" I assume

18:38 <marcan> no idea about the g

22:01 <brentr123[m]> Anyone know how to enable verbose boot? I tried Command +C upon restart and it isn’t working

22:09 <nobodynada> cmd+v, not cmd+c

22:09 <brentr123[m]> Ah