#asahi-dev · 2022-02-02 14:50
mazyeah, I had some run-in with gmail, only solved by some heavy duty DKIM/SPF setup.#asahi-dev · 2022-02-02 14:47
maz(and that's saying something)#asahi-dev · 2022-02-02 14:47
mazI used to do that. and then I realised that they were worse than me.#asahi-dev · 2022-02-02 14:44
mazbut I'm everyday grateful I run my own email infrastructure.#asahi-dev · 2022-02-02 14:44
mazbut hey, they managed to get my key right, so I can still read email people send me.#asahi-dev · 2022-02-02 14:43
mazthat's the fun part. they have no clue.#asahi-dev · 2022-02-02 14:43
mazkeyservers*#asahi-dev · 2022-02-02 14:43
mazthey likely have imported the database from one of the keyuervers.#asahi-dev · 2022-02-02 14:42
mazyeah, proton has some interesting policies.#asahi-dev · 2022-01-28 16:12
mazis there a branch with an SMC driver somewhere? I may try and have a go at implementing PCIe hotplug during the weekend.#asahi-dev · 2022-01-27 09:02
mazsven: the best bet would be to pick the series and to push it forward. it is the lack of persistent engagement that ruined the effort so far.#asahi-dev · 2022-01-27 08:53
mazit's been going on for at least 5 years, with various patch series.#asahi-dev · 2022-01-27 08:52
mazthough this assumes idmap.#asahi-dev · 2022-01-27 08:51
mazsven: there was this: https://lore.kernel.org/linux-iommu/YTelDHx2REIIvV%2FN@orome.fritz.box/T/#asahi-dev · 2022-01-26 08:54
mazkettenis: yeah, this is a massive hack, and it may not even be reliable (nothing says that the CPU release addresses are constants).#asahi-dev · 2022-01-26 08:46
mazjannau: the only solution I have so far is to hack grub to provide a different DTB per kernel. i really hope we won't have to do that too often...#asahi-dev · 2022-01-24 20:13
mazhuh. odd. I wonder whether I changed something in the macOS config to get there.#asahi-dev · 2022-01-24 20:02
mazsven: doesn't it remember it was last on? I sometimes turn it off (and back on) from the PDU, and it does start OK.#asahi-dev · 2022-01-24 19:33
mazPDU FTW.#asahi-dev · 2022-01-24 18:43
mazbut hey, as long as it is only me, that's not a huge problem.#asahi-dev · 2022-01-24 18:42
mazkettenis: I actually do. apart for the lack of TB support, the mini is pretty much feature complete for me.#asahi-dev · 2022-01-24 18:34
mazyeah, that's pretty bad, but I don't see a good way to fix it.#asahi-dev · 2022-01-24 17:59
mazkettenis: yup, confirmed. revert the PMGR nodes, and we're back in business. gross.#asahi-dev · 2022-01-24 17:17
mazkettenis: guess so. I need to trace that a bit...#asahi-dev · 2022-01-24 16:57
mazerr... looks like 5.16 doesn't boot with the 5.17-rc1 DT.#asahi-dev · 2022-01-24 16:23
mazj`ey: yup, that one works fine.#asahi-dev · 2022-01-24 16:14
mazj`ey: hadn't noticed that one. let me try.#asahi-dev · 2022-01-24 16:13
mazI think your original patch still has value though.#asahi-dev · 2022-01-24 16:12
mazmarcan: I'm going to post a patch for this type-c Oops (checking for acpi_disabled in the offending functions)#asahi-dev · 2022-01-22 11:50
maz(feels like a poor-man's TFTP boot... ;-)#asahi-dev · 2022-01-22 11:50
mazjannau: yup, that's another possibility. I'll give it a shot.#asahi-dev · 2022-01-22 11:32
mazdefinitely a dev feature#asahi-dev · 2022-01-22 11:31
mazmarcan: yeah, that's really not something you'd want for 'production' stuff, but that would be useful to transiently test, let's say, a new u-boot version.#asahi-dev · 2022-01-22 11:27
mazof course, I could always get rids of USB altogether, as I don't have to upload huge payloads anymore (it was massively useful when I was uploading full kernels).#asahi-dev · 2022-01-22 11:25
mazI guess what I'd need would be to directly start the proxy, and only start the payload if no activity happens within x seconds.#asahi-dev · 2022-01-22 11:24
mazpovik: it is close to what I'd need, but not quite that. my m1 is permanently hooked to another box that captures its serial port and also uploads the chained m1n1 over USB. I think the sole enumeration would count as activity, and would drop directly into the proxy.#asahi-dev · 2022-01-22 11:07
mazmarcan: ah! that'd be fab.#asahi-dev · 2022-01-22 11:05
mazmarcan: one thing I have been thinking about is to be able to get m1n1 to wait a couple of seconds before starting the payload it is bundled with. this would allow sneaky chainloading from a m1n1 that was bundled with u-boot to a new m1n1 with more experimental feature.#asahi-dev · 2022-01-19 13:26
mazmarcan: I'll have a look once I'm done with the current $work deadlines. hopefully things will calm down next week (whether I hit the deadlines or not!).#asahi-dev · 2022-01-08 15:12
mazI have the ugly feeling that we should never be on this path for !ACPI systems, but that should at least kickstart the discussion with rjw.#asahi-dev · 2022-01-08 15:10
mazmarcan: yeah, the world is enabled in my base config... ;-)#asahi-dev · 2022-01-08 15:05
mazah, thanks.#asahi-dev · 2022-01-08 15:04
mazj`ey: got a pointer?#asahi-dev · 2022-01-08 15:03
mazsven: this is with next.#asahi-dev · 2022-01-08 15:03
mazsven: rings any bell? https://paste.debian.net/1226410/#asahi-dev · 2022-01-06 13:26
mazupdates his obsd VM on M1...#asahi-dev · 2022-01-04 13:33
mazseems reasonable. you probably don't want to let userspace mess with something that the kernel doesn't know about. unless you want to do some sort of PCI passthrough to userspace (and even then, you can have a stub driver).#asahi-dev · 2022-01-04 13:29
maznice one.#asahi-dev · 2022-01-04 13:28
mazcould also be firmware...