#asahi · 2021-07-12 13:29
kettenisI can hack something up ;)#asahi · 2021-07-12 13:28
kettenisneed to write a driver ;)#asahi · 2021-07-12 13:25
kettenisso I'll need to write a mailbox driver after all#asahi · 2021-07-12 13:23
kettenisbut maybe the rollover needs to be acked#asahi · 2021-07-12 13:23
kettenisi would expect the syslog to be a ring buffer of some sort#asahi · 2021-07-12 13:22
kettenisright#asahi · 2021-07-12 13:21
kettenissven, I'm not touching the mailbox under OpenBSD#asahi · 2021-07-12 13:20
kettenisso I wonder if it stops because it wants more memory...#asahi · 2021-07-12 13:18
kettenisat least I don't get any further interrupts#asahi · 2021-07-12 13:18
kettenisI'm investigating an nvme issue where the controller seems to stop completely#asahi · 2021-07-12 13:18
kettenisthat is something that remeains to be seen#asahi · 2021-07-12 13:02
kettenislinux could check whether the processor is running and skip initialization in this case#asahi · 2021-07-12 13:01
kettenisfor OpenBSD I'm not really interested in supporting a non-UEFI bootflow, so having it started in u-boot is just fine#asahi · 2021-07-12 12:53
kettenismust be possible as iBoot reads stuff from the NVMe disk but the coprocessor isn't running when u-boot takes over#asahi · 2021-07-12 12:51
kettenissven: not looked at that#asahi · 2021-06-18 18:33
kettenisas long as these spambots use all caps my brain will filter them out anyway#asahi · 2021-06-18 18:32
kettenisbut as you said, if it gets real ugly...#asahi · 2021-06-18 18:32
ketteniswhich is why marcan opted for +S#asahi · 2021-06-18 18:29
kettenisthe channel is already +S isn't it?#asahi · 2021-06-18 11:48
kettenisbut apparently they shut it down before handing control to the "kernel"#asahi · 2021-06-18 11:47
kettenisI expected the NVMe one to be running as well, since iBoot reads from the NVMe#asahi · 2021-06-18 11:47
kettenismarcan: given that you havr a display going, it would be logical for it to br running already ;)#asahi · 2021-06-18 11:43
kettenisthe SMC one is already running, but the NVMe one isn't#asahi · 2021-06-18 11:42
kettenismarcan: regarding the side-cores, the S#asahi · 2021-06-11 11:12
kettenisto add the necessary pci device nodes and aliases#asahi · 2021-06-11 11:11
kettenishttps://github.com/kettenis/u-boot/tree/apple-m1-m1n1-nvme#asahi · 2021-06-11 11:11
kettenispushed a change to#asahi · 2021-06-11 11:11
kettenisthat helps giving the ethernet interface the proper mac address#asahi · 2021-06-11 11:10
kettenishttps://github.com/AsahiLinux/m1n1/pull/76#asahi · 2021-06-10 14:39
kettenisbut it would be a viable stopgap measure while you guys figure out how to handle the mismatched page size issue in the IOMMU code#asahi · 2021-06-10 14:36
kettenistrue#asahi · 2021-06-10 14:35
kettenisso using "external-facing" there should already work#asahi · 2021-06-10 14:35
kettenissven: the DT binding for the apple pcie controller already includes a node for the bridge corresponding to the root port#asahi · 2021-06-10 14:30
kettenisso if the generic of pci code in linux can handle a suitable property on that node, we should be alright#asahi · 2021-06-10 14:29
ketteniswe need to have a place to store its mac-address#asahi · 2021-06-10 14:28
kettenissven: my plan is to add a node in the device tree for the wifi chip#asahi · 2021-06-10 11:28
kettenisstill works for me#asahi · 2021-06-10 10:08
ketteniswonder if the queing mechanism matches the "NVMe-over-Fabrics" standard#asahi · 2021-06-10 10:03
ketteniss/inear/linear/#asahi · 2021-06-10 10:02
kettenisthese apple "nvme" devices use some funky inear command queue instead of a ring#asahi · 2021-06-10 08:53
kettenisthe cheapest iMac doesn't have the ASM3142 chip I suppose?#asahi · 2021-06-09 19:10
kettenissven: works like a charm#asahi · 2021-06-09 19:01
kettenisso I think I end up running usb_iodev_init()#asahi · 2021-06-09 19:01
kettenissven: I'm loading u-boot over serial from a m1n1 chainloaded over serial from an older m1n1#asahi · 2021-06-09 18:57
kettenisand i verified that it works for u-boot#asahi · 2021-06-09 18:57
kettenisI think I'd like to use bypass mode for dwc3 in u-boot since it means I don't have to come up with a proper solution for some of the hacks I have now#asahi · 2021-06-08 20:29
kettenisbut a linux driver should probably be interrupt-driven#asahi · 2021-06-08 20:28
kettenisfor SMC and NVMe#asahi · 2021-06-08 20:28
kettenisI only implemented the states that are needed#asahi · 2021-06-08 20:27
kettenisnot really