#asahi · 2023-01-14 01:45
amarioguyi can attest to that... i tried to get access to a SEM at my uni#asahi · 2023-01-14 01:38
amarioguythey have their own fabs it's less of a hit for them#asahi · 2023-01-14 01:38
amarioguychadmed: ofc i'm more knocking on the whole "SPR delayed" stuff more than high numbers of tapeouts#asahi · 2023-01-14 01:37
amarioguyfrom what i've heard SEMs take a lot of maintenance#asahi · 2023-01-14 01:36
amarioguychadmed: they can't even launch a server processor without high numbers of steppings...#asahi · 2023-01-14 01:34
amarioguymore of a node change indication#asahi · 2023-01-14 01:34
amarioguyusually "N5" and such doesn't track with actual feature size#asahi · 2023-01-14 01:33
amarioguy^#asahi · 2023-01-14 01:32
amarioguycould always unfuse one if you got the time and or cash :)#asahi · 2023-01-14 01:30
amarioguymarcan: just wait until TSS goes down :P#asahi · 2023-01-14 01:30
amarioguythen it'll lock to your account#asahi · 2023-01-14 01:30
amarioguyTheDcoder: if you enable Find My#asahi · 2023-01-14 01:28
amarioguythe authorized install stuff is just an HTTP sequence#asahi · 2023-01-14 01:28
amarioguyno#asahi · 2023-01-14 01:28
amarioguyTheDcoder: in practice apple combines all the authorizations into one ticket#asahi · 2023-01-14 01:27
amarioguySEP side... it's complicated#asahi · 2023-01-14 01:27
amarioguything is OTAs for AP side need to pin the nonce because you can't exactly request a ticket in iboot for those#asahi · 2023-01-14 01:26
amarioguymarcan: there's NVRAM vars you can set for AP side anyways :P#asahi · 2023-01-14 01:22
amarioguyTheDcoder: working on understanding the SEP init sequence btw if you're wondering how that's going#asahi · 2022-12-15 04:13
amarioguyspecs wise focus a lot more on disk space than CPU#asahi · 2022-12-15 04:13
amarioguy(tho in fairness if asahi's the concern anything goes, even a m1 mac mini works fine if desktops ever come in scope)#asahi · 2022-12-15 04:12
amarioguyif you want any sort of intense virt - definitely want a MBP (the real pros, not the pro badged touch bar one)#asahi · 2022-12-15 04:11
amarioguyeven an m2 one would be fine enough#asahi · 2022-12-15 04:11
amarioguyunixgeek: in terms of what to get - depends on what you want to do, a lot can be done with just the base m1 macbook air#asahi · 2022-10-15 01:20
amarioguy(no ARM platform that Windows is officially supported on lacks support for 4k pages)#asahi · 2022-10-15 01:20
amarioguySerebtty[m]: given that windows itself needs 4k pages and the apps running on Windows can safely assume 4k pages (for now anyways) yep the 4k patch is necessary#asahi · 2022-08-05 03:01
amarioguy(now *why* they do this is anyone's guess lol)#asahi · 2022-08-05 03:01
amarioguylucienmaxwell[m]: heads up that if you're ever on an Apple Silicon desktop, apple's current status quo is to not init the framebuffer until macOS boots so on desktops you need to hold it down until you see the power led go a bit dimmer#asahi · 2022-08-05 01:34
amarioguybacklight was discussed somewhat recently i think? (not sure how high it is on the priority list though)#asahi · 2022-08-05 01:31
amarioguylucienmaxwell[m]: this is the place lol - pretty much most major and minor discussion of asahi development happens here (and the other asahi IRC channels too)#asahi · 2022-08-04 14:24
amarioguymaz: ahh gotcha that makes sense#asahi · 2022-08-03 12:21
amarioguynow i'm not 100% sure if this is enough to imply that the hack used for m1/m1 pro is no longer necessary (or even if i tested for the right thing) but honestly the interrupt storm after setting ICH_MISR_EL2 didn't seem like something that should happen unless maintenance irqs are supported outright#asahi · 2022-08-03 12:19
amarioguy(furthermore having maintenance interrupts pending and enabling virtual interrupts results in an interrupt storm as there's no handling for maintenance irqs in m1n1)#asahi · 2022-08-03 12:19
amarioguy(specifically i found while doing stuff with m1n1 that ICH_MISR_EL2 is implemented, seems to be configurable and the register changes pretty much exactly as the spec says it should change)#asahi · 2022-08-03 12:17
amarioguymaz: i did some very very very initial testing, and i *think* (not 100% sure on this but it's pretty likely) that m2/t8112 actually supports the maintenance IRQs#asahi · 2022-06-27 13:47
amarioguyright right#asahi · 2022-06-27 13:47
amarioguyah my apologies#asahi · 2022-06-27 13:46
amarioguymade maz's life very annoying as i'm sure he can attest#asahi · 2022-06-27 13:46
amarioguySobek[m]: basically m1 had this odd bug where virtual SErrors sometimes fired on the host if the vGIC registers had invalid stuff written to them#asahi · 2022-06-27 13:45
amarioguybtw we should probs verify if the SEIS bug is fixed or not#asahi · 2022-06-27 13:44
amarioguy_jannau_: thanks!!#asahi · 2022-06-24 20:34
amarioguynicolas17: yea ofc marcan would need this stuff asap lol#asahi · 2022-06-24 20:20
amarioguywas also wondering if apple added other things to m2 via MSRs or anything like that (waiting on the m2 mba to release that's why i'm not able to probe myself lol)#asahi · 2022-06-24 20:19
amarioguyjannau: that was something i did want to account for yes#asahi · 2022-06-24 20:15
amarioguyj'ey: m2 mbp exists#asahi · 2022-06-24 20:13
amarioguyah so nested virt *is* on m2 now lol good to know#asahi · 2022-06-24 19:28
amarioguym2 msr dump when /s#asahi · 2022-06-07 13:29
amarioguyatm stuck on my desktop because reasons#asahi · 2022-06-07 13:29
amarioguyplanning to use that as my main dev machine#asahi · 2022-06-07 13:29
amarioguyComradeMisato: when my studio comes in :)