#asahi-dev · 2021-01-31 05:40
davidrysk[m]ldhacker: M1 uses aarch64... I'd also recommend any good computer architecture books#asahi-dev · 2021-01-31 05:39
davidrysk[m]night modwizcode#asahi-dev · 2021-01-31 05:39
davidrysk[m]marcan: any plans to do another stream sometime?#asahi-dev · 2021-01-31 05:27
davidrysk[m]a lot of it probably comes up when you're doing more complex stuff#asahi-dev · 2021-01-31 05:27
davidrysk[m]oops :)#asahi-dev · 2021-01-31 05:13
davidrysk[m]ever dealt with boring dull appliance manuals? :P#asahi-dev · 2021-01-31 05:13
davidrysk[m]I've seen someone say that ARM documentation "reads like stereo instructions"#asahi-dev · 2021-01-31 05:09
davidrysk[m]what about G5 and all of its idiosyncrasies?#asahi-dev · 2021-01-31 05:07
davidrysk[m]how much of it is ARM and how much of it is Apple's version of it?#asahi-dev · 2021-01-29 17:26
davidrysk[m]is there any aarch64 apple chip?#asahi-dev · 2021-01-29 17:26
davidrysk[m]yeah I didn't expect it#asahi-dev · 2021-01-29 17:02
davidrysk[m]As for difference between performance and efficiency: there are some differences but not everything is mapped out yet#asahi-dev · 2021-01-29 17:01
davidrysk[m]I don't think upstream qemu has support for the older Apple-custom cores though? I might be wrong#asahi-dev · 2021-01-29 17:01
davidrysk[m]So the M1 probably does need a new core type (though it probably can share with the A14 if there is an A14)#asahi-dev · 2021-01-29 17:00
davidrysk[m]There are plenty of different HID registers and whatnot in the M1.#asahi-dev · 2021-01-29 16:14
davidrysk[m]yeah#asahi-dev · 2021-01-29 16:14
davidrysk[m]I don't think anything available supports virtualization of macOS/aarch64#asahi-dev · 2021-01-29 16:14
davidrysk[m]yeah for Intel macOS#asahi-dev · 2021-01-29 16:13
davidrysk[m]the EULA allows virtualization of macOS on Apple hardware :)#asahi-dev · 2021-01-29 16:13
davidrysk[m]I believe it would also be legal EULA-wise (though IANAL etc)#asahi-dev · 2021-01-29 16:12
davidrysk[m]since the ability to virtualize macOS/ARM would be useful#asahi-dev · 2021-01-29 16:12
davidrysk[m]It would be nice if whatever infrastructure is used to load m1n1 can be extended to load xnu/macOS too#asahi-dev · 2021-01-29 16:11
davidrysk[m]well, my concern about direct mach-o loading in qemu is that upstream might not like it. I dunno.#asahi-dev · 2021-01-29 15:52
davidrysk[m](or they're just running openfirmware)#asahi-dev · 2021-01-29 15:51
davidrysk[m]I think that's the approach that they're taking for booting PPC Macs#asahi-dev · 2021-01-29 15:51
davidrysk[m]modwizcode: why not write a fake/stubbed iBoot for qemu?#asahi-dev · 2021-01-27 21:43
davidrysk[m]never_released: I see.#asahi-dev · 2021-01-27 17:20
davidrysk[m]marcan: boo apple, that's disappointing#asahi-dev · 2021-01-27 17:14
davidrysk[m]I can see why shared-context may be appealing to security researchers, but that is not whom Asahi Linux is for, and that's not what upstream caters for either#asahi-dev · 2021-01-27 17:01
davidrysk[m]marcan: nested virtualization is a base feature of ARMv8.3#asahi-dev · 2021-01-24 19:28
davidrysk[m]https://worthdoingbadly.com/xnuqemu3/#asahi-dev · 2021-01-24 19:00
davidrysk[m]very similar to the rules for kernel upstreaming#asahi-dev · 2021-01-24 19:00
davidrysk[m]upstreaming to qemu: https://wiki.qemu.org/Contribute/SubmitAPatch#asahi-dev · 2021-01-22 15:03
davidrysk[m]Apple says not to use it even though xnu says they don't use it#asahi-dev · 2021-01-22 15:03
davidrysk[m]x18 is defined as the platform register by the AAPCS though, so you shouldn't be using it as a scratch register#asahi-dev · 2021-01-22 14:31
davidrysk[m]if that vendor stops making/selling said cable, we're out of luck#asahi-dev · 2021-01-22 14:31
davidrysk[m]marcan: if anything I'm concerned that there is only one vendor of either required cable#asahi-dev · 2021-01-22 14:25
davidrysk[m]opticron: I'm aware of the memory model differences, but they don't appear to lead to ABI differences on Windows aarch64#asahi-dev · 2021-01-22 14:00
davidrysk[m]windows arm64 abi is basically the same as the standard aarch64 abi, which is a nice departure from what MS usually does#asahi-dev · 2021-01-22 14:00
davidrysk[m]I only know of lp64 and ilp32, and llp64 for windows but that doesn't use a different ABI than lp64#asahi-dev · 2021-01-20 22:47
davidrysk[m]it seems that apple doesn't use interrupts though, their driver is polled-mode#asahi-dev · 2021-01-20 22:47
davidrysk[m](in xnu)#asahi-dev · 2021-01-20 22:47
davidrysk[m](for the uart)#asahi-dev · 2021-01-20 22:47
davidrysk[m]lmao the DMA stuff for the core is `#if 0`'d out#asahi-dev · 2021-01-20 22:45
davidrysk[m]aah...#asahi-dev · 2021-01-20 17:58
davidrysk[m]alright :)#asahi-dev · 2021-01-20 17:52
davidrysk[m]marcan: when are you going to sleep? it's nearly 3am#asahi-dev · 2021-01-20 16:37
davidrysk[m]marcan: cutoff at 2am?#asahi-dev · 2021-01-20 13:47
davidrysk[m]I can talk here#asahi-dev · 2021-01-20 13:47
davidrysk[m]that's me