#asahi-dev · 2024-05-15 06:24
mazjannau: I have a similar hack in my tree.#asahi-dev · 2024-05-14 18:24
mazjannau: yup, does the job perfectly. no more whining! thanks again!#asahi-dev · 2024-05-14 18:18
mazjannau: thanks! let me try this sucker...#asahi-dev · 2024-05-14 18:17
mazj`ey: but yeah, this box is very dodgy. it was installed before any official support for it...#asahi-dev · 2024-05-14 18:15
mazj`ey: maybe. but it was OK with 6.8, and only the same patches with 6.9 showed the issue. so it has to be a change that is unrelated to the Asahi patches.#asahi-dev · 2024-05-14 18:14
mazremoving macsmc_power "fixes" the behaviour.#asahi-dev · 2024-05-14 18:14
mazjannau: another issue I forgot to mention is that on my M2 Air, I get a constant screaming of "[ 235.022683] power_supply macsmc-battery: driver reporting unsupported charge behaviour"#asahi-dev · 2024-05-14 15:47
mazj`ey: by then I will have missed the opportunity to get $work to pay for it! ;-)#asahi-dev · 2024-05-14 15:45
mazshame this thing is so locked down. would have made a hell of a "shelf computer"...#asahi-dev · 2024-05-14 15:43
mazjannau: is that looking at the DT?#asahi-dev · 2024-05-14 10:26
mazfor Apple, that's not really a problem: they already had AMX without SVE, so no SW expects SVE at all. but running non-Apple SW on that HW becomes more challenging.#asahi-dev · 2024-05-14 10:24
mazkettenis: yeah. it breaks a lot of expectations in userspace that you can exit streaming mode and still use SVE. but it you look at the architecture, there is no actual dependency and you are allowed to build one without the other.#asahi-dev · 2024-05-14 10:04
mazjannau: forget what I said. it reproduces the same way. I still had this silly dwc3 hack in my tree.#asahi-dev · 2024-05-14 08:23
mazno HV involved, obviously.#asahi-dev · 2024-05-14 08:23
mazjannau: this is odd. with your tree (and the same config), nothing catches fire. no USB3, but no SError either.#asahi-dev · 2024-05-14 08:04
mazjannau: ah, thanks for confirming. too bad it's not just my own bizarre setup.#asahi-dev · 2024-05-14 07:58
mazjannau: it could well be a merge problem on my side. let me try your rebase.#asahi-dev · 2024-05-14 07:52
maznicolas17: it seems to be lacking SVE though, which creates an interesting situation for luserspace.#asahi-dev · 2024-05-14 07:51
mazjannau: that's on an M2 pro mini.#asahi-dev · 2024-04-18 06:27
mazhaving recently experienced an unexpected lock-up on a CS while debugging remotely (which I suspect is from the USB stack), I've added a 5s watchdog reset. Patch pushed out.#asahi-dev · 2024-04-08 12:38
mazjust plugging a USB hub causes the SError.#asahi-dev · 2024-04-08 12:37
mazsven: no difference indeed#asahi-dev · 2024-04-08 07:19
mazsven: feel free to throw test patches in my direction, this is a test box that I can crash at will.#asahi-dev · 2024-04-05 17:09
mazXen doesn't require 32bit support, AFAICR#asahi-dev · 2024-04-05 17:09
mazE2H=1 only, no GIC.#asahi-dev · 2024-04-05 17:08
mazseems like a good option.#asahi-dev · 2024-04-05 17:07
maz(I see you've asked on the KVM channels as well, which is equally off-topic).#asahi-dev · 2024-04-05 17:07
mazI'm sure there is a channel dedicated to Xen where you'll be able to get some help.#asahi-dev · 2024-04-05 17:06
mazschroes: this channel is about Linux on Apple HW. Xen is majorly off-topic.#asahi-dev · 2024-04-05 17:03
mazschroes: unlikely, at least on an Apple platform. some of the things Xen relies on are missing from the CPU.#asahi-dev · 2024-04-01 12:38
mazsven: no worries, no hurry. that doesn't block me, as I don't have much hanging off these connectors anyway. Enjoy the rest of your vacation!#asahi-dev · 2024-04-01 11:42
mazsven: not sure you care at the moment, but USB-C support explodes on my M2 Pro with 6.9-rc2 (+ the asahi patches). I dumped a trace at https://paste.debian.net/1312676/ (but that's not overly useful given that it is a SError).#asahi-dev · 2024-03-21 17:34
mazwhen will Emacs be merged? ;-)#asahi-dev · 2024-03-10 04:51
mazj`ey: although 8.6 (and thus 9.1) *mandates* to have a 1GHz counter frequency, almost everybody ignored this requirement. #asahi-dev · 2024-02-13 17:19
mazdragging libc in something as low-level as m1n1 would be... interesting.#asahi-dev · 2024-02-13 16:27
mazj`ey: a better option is to keep building it, but to reclaim the section at boot time. I had patches for that, but it was something like a couple of pages, which wasn't worth the complexity.#asahi-dev · 2024-02-13 16:23
mazzzywysm: size doesn't matter in that case. I can compile a kernel in a couple of minutes. testing takes hours, multiplied by the number of architectural variants this needs to run on.#asahi-dev · 2024-02-13 16:21
mazthat's what maintenance means.#asahi-dev · 2024-02-13 16:21
mazno: the maintenance cost is that an every *patch* you apply, you test *all* the possible combinations.#asahi-dev · 2024-02-13 16:20
mazzzywysm: sure. but the maintenance cost is way too high, and completely impractical.#asahi-dev · 2024-02-13 16:19
mazI really don't want to have to ever split the two. specially when people use nVHE on 9.x HW.#asahi-dev · 2024-02-13 16:14
mazj`ey: by luck!#asahi-dev · 2024-01-18 16:48
mazfunny how these things go full circle...#asahi-dev · 2024-01-18 16:47
mazmeh. I had paged out that particular thread...#asahi-dev · 2024-01-18 11:39
mazthe AS bindings predate this work anyway...#asahi-dev · 2024-01-18 11:38
mazdo we have anything upstream on that front already?#asahi-dev · 2024-01-18 11:38
mazj`ey: it probably means that if there is a 'standard' to power-on wifi and co, we should probably implement it (or explain why we think they got it wrong).#asahi-dev · 2024-01-16 11:01
mazright. the only use case I could come up with was to preserve messages across a reboot of the console server (which is a vanishingly rare event).#asahi-dev · 2024-01-16 10:41
mazI don't really need this (I constantly log everything with conserver), but maybe it would fit other work flows?#asahi-dev · 2024-01-16 10:40
mazquestion to the couple of CS users here: would there be an appetite for a sort of "replay" function that would keep the last x KB worth of serial output and replay it on demand? some sort of HW-assisted scrollback, for the lack of a better description?