#asahi-dev · 2026-06-28 13:26
yukaI wonder why my uboot isn't working then#asahi-dev · 2026-06-28 13:25
yukaso the kernel turns it on later?#asahi-dev · 2026-06-28 13:25
chaos_princessjust don't turn it on#asahi-dev · 2026-06-28 13:25
chaos_princessit already does#asahi-dev · 2026-06-28 13:25
yukayes#asahi-dev · 2026-06-28 13:25
chaos_princessfor what, pmp?#asahi-dev · 2026-06-28 13:25
yukaI think m1n1 would have to provide a safe starting state#asahi-dev · 2026-06-28 13:25
yukaAh shit, that doesn't even work#asahi-dev · 2026-06-28 13:22
yukasimilar to atcphy#asahi-dev · 2026-06-28 13:21
yukaI guess this could be fixed in uboot by providing a no-op pmp driver just to bind to the compatible and pass through the dependency?#asahi-dev · 2026-06-28 13:21
yukafirmware is hardware in this case#asahi-dev · 2026-06-28 13:20
LarstiQwhere firmware is included with hardware or software? I imagine the border is sometimes hard to draw#asahi-dev · 2026-06-28 13:19
yukaI get why it's done this way, but the device tree structure for pmp does not seem quite right. As marcan said in some streams, device trees should describe the hardware and not how the software behaves#asahi-dev · 2026-06-28 13:18
yukadon't have a good idea to teach uboot to use ps_ans while keeping pmp_report_ans for linux#asahi-dev · 2026-06-28 13:18
yuka*ans#asahi-dev · 2026-06-28 13:18
yukaso I guess I put the ane in always-on for now#asahi-dev · 2026-06-28 13:17
yukadoesn't know how to enable power for the nvme#asahi-dev · 2026-06-28 13:17
yukauboot....#asahi-dev · 2026-06-28 13:16
yukaAnd it's broken lol#asahi-dev · 2026-06-28 13:16
yukafirst try to boot t6034 with pmp, wish me luck#asahi-dev · 2026-06-28 12:37
yukaafter boot they are turned off#asahi-dev · 2026-06-28 12:37
yukaAh, during boot#asahi-dev · 2026-06-28 12:14
yukahuh, the dispext[0-3] report entries are always on in macOS#asahi-dev · 2026-06-28 12:04
yukaI can just do that#asahi-dev · 2026-06-28 12:04
yukaAh, true. then it will have to override the power domain of ans back to the normal pmgr one#asahi-dev · 2026-06-28 12:04
Ami[m]chaos_princess: Oh I see I see, thanks for explaining :)#asahi-dev · 2026-06-28 12:04
jannauit doesn't because USE_PMP is not defined. we can sort this out when we decide to use PMP#asahi-dev · 2026-06-28 12:04
chaos_princessthe adt is completely different, as for example everything starts at 0x21{zeroes} instead of 0x20{zeroes}#asahi-dev · 2026-06-28 12:03
Ami[m]chaos_princess: Does it really have that much an affect on the software end? Sorry that seems naïve to ask#asahi-dev · 2026-06-28 12:03
jannauI have access to a m3 ultra but it's low priotrity since it's macos 15 and later#asahi-dev · 2026-06-28 12:03
yukajannau: the thing is, j575dap re-uses the same ANS pmp report entry#asahi-dev · 2026-06-28 12:02
chaos_princessand m3pro and m3max are materially different, while for m1/2, the pro was a chopped max#asahi-dev · 2026-06-28 12:02
chaos_princessbut m1/m2ultra had nvme on first die#asahi-dev · 2026-06-28 12:02
Ami[m]Ah I see I see#asahi-dev · 2026-06-28 12:02
chaos_princessit is 2 m3max glued together#asahi-dev · 2026-06-28 12:01
Ami[m]yuka: Sorry for interrupting again - but is M3 Ultra a different design to M3 Pro and similar? If so, why??? 😭#asahi-dev · 2026-06-28 12:01
jannauoverriding ans' power-domain from t6032.dtsi isn't much of an issue, it will need anyway to add all the devices from the second die. although the sram is probably used on both dies#asahi-dev · 2026-06-28 12:01
yukacomplete support is cool and everything but maybe there's more important things than fully supporting m3 ultra#asahi-dev · 2026-06-28 11:58
chaos_princessit is not even tested on m1/2 ultras#asahi-dev · 2026-06-28 11:58
chadmeddo we need to buy an m3 ultra studio #asahi-dev · 2026-06-28 11:57
yukaI'm trying to cover all of M3, and fixups for everything non-t6034 can come later#asahi-dev · 2026-06-28 11:56
yukawriting device tree nodes with lots of guessing and assumptions here#asahi-dev · 2026-06-28 11:56
yukaguess someone needs to test and potentially trace it eventually :-)#asahi-dev · 2026-06-28 11:56
yukaah you mean our code#asahi-dev · 2026-06-28 11:56
chaos_princessmuli-die#asahi-dev · 2026-06-28 11:55
chaos_princessand if any changes are needed#asahi-dev · 2026-06-28 11:55
yukawhy wouldn't it?#asahi-dev · 2026-06-28 11:55
chaos_princessalso, we have no idea if pmp even works on any ultras#asahi-dev · 2026-06-28 11:55
chadmeddamn it#asahi-dev · 2026-06-28 11:55
chadmedthink different