#
✕ nick:kettenis
#asahi · 2022-02-19 13:04
kettenisthat doesn't make sense to me; shouldn't that be just SZ_128M?
#asahi · 2022-02-19 13:03
kettenisstatus |= env_set_hex("kernel_comp_size", base + SZ_128M);
#asahi · 2022-02-19 13:02
kettenisI might pick that up if it gets submitted upstream
#asahi · 2022-02-19 11:49
kettenisfor that reason I didn't even bother adding the t600x DTs
#asahi · 2022-02-19 11:48
kettenisnote that now that the device trees are in sync, there is no downside of using the kernel DTs from the asahi kernel branch instead of the ones from u-boot
#asahi · 2022-02-19 11:44
kettenisno
#asahi · 2022-02-19 11:44
kettenisthat now has the DTs fully sunched with the asahi kernel branch
#asahi · 2022-02-19 11:43
kettenishttps://github.com/kettenis/u-boot/commits/asahi
#asahi · 2022-02-17 14:03
kettenisj'ey: a piano is out of tune by definition!
#asahi · 2022-02-14 21:36
kettenisMPIDR_EL1
#asahi · 2022-02-14 21:35
kettenisthe cluster ID is encoded in the MPIDR register
#asahi · 2022-02-14 21:33
kettenisI wouldn't be surprised if the mapping of CPU IDs onto PMGR registers is messed up for the modelsthat have some of the cores disabled
#asahi · 2022-02-14 21:31
kettenisnote that the cpufreq code is an old diff that is effectively abandoned because the maintainer of the relevant subsystem wan to see a different approach
#asahi · 2022-02-14 18:56
kettenisdisplay is probably the biggest consumer
#asahi · 2022-02-11 20:57
kettenisI thought he was asking about u-boot
#asahi · 2022-02-11 20:57
kettenisthe same way as for the Linux kernel I belive
#asahi · 2022-02-11 20:56
kettenisyou'll need to set the CROSS_COMPILE environment variable
#asahi · 2022-02-11 20:08
kettenisthere was a commit missing on my branch
#asahi · 2022-02-11 18:09
kettenisbut I'll declare apple-m1-m1n1-nvme obsolete as soon as we have an asahilinux branch with smc support
#asahi · 2022-02-11 18:08
kettenisno
#asahi · 2022-02-11 18:08
kettenisthere is no new pcie patch yet
#asahi · 2022-02-11 18:07
kettenisit did and and I rebased my t6000 branch on master and added jannau's patches to it
#asahi · 2022-02-11 13:41
kettenisI just noticed the pull request on the mailing list
#asahi · 2022-02-11 13:39
kettenis(well, I assume it is v3)
#asahi · 2022-02-11 13:39
kettenismarex sent a pull request with v3 included
#asahi · 2022-02-11 12:32
kettenistpw_rules: looks like your usb keyboard fix diff is on its way to master as well
#asahi · 2022-02-11 12:19
kettenisthat is dependent on things landing on the asahi branch of the kernel repository
#asahi · 2022-02-11 11:42
kettenisar: the development doesn't have to wait until after 2022.04, but I'm fairly certain it will miss 2022.04
#asahi · 2022-02-11 09:47
kettenisbut if not, the asahi-installer can just use a handful of patches on top of the official U-Boot release
#asahi · 2022-02-11 09:46
kettenism1 pro/max support might still make 2022.04 since it only touches board-specific files
#asahi · 2022-02-11 09:45
kettenisI'd like to put back pcie support at some point (type-A ports on the mini), but that will have to wait until after U-Boot 2022.04 has been released
#asahi · 2022-02-11 09:44
kettenisyes
#asahi · 2022-02-11 09:44
kettenisand there is the usb3 role switching issue (but we can work around that in the device trees)
#asahi · 2022-02-11 09:43
kettenistpw_rules's usb keyboard thing would also be really helpful
#asahi · 2022-02-11 09:42
kettenisjannau's dynamic mem layout would also be helpful for m1
#asahi · 2022-02-11 09:36
kettenisyay, next batch of u-boot patches is in
#asahi · 2022-02-10 18:14
kettenisbut that's not an option for a generic distro kernel
#asahi · 2022-02-10 18:12
kettenisaxboe: ok, then Linux maybe needs that property on the individual device nodes
#asahi · 2022-02-10 17:53
kettenisnot entirely sure how this works in Linux-land, but adding a dma-coherent property on the "soc" node (right next to "nonposted-mmio") might be enough
#asahi · 2022-02-10 17:50
kettenisas far as I can tell DMA is cache-coherent on these machines
#asahi · 2022-02-10 17:49
kettenisregarding dcache flushing: I think the device tree needs a "dma-coherent" property somewhere
#asahi · 2022-02-10 16:49
kettenismaz: I'd prefer to avoid that ;)
#asahi · 2022-02-10 16:46
kettenisthe keyboard support in u-boot isn't create, but if you type carefully it gets the job done
#asahi · 2022-02-10 16:45
kettenisphew
#asahi · 2022-02-10 16:39
kettenisit should, but your device tree might be broken
#asahi · 2022-02-09 12:24
kettenismaz: the challenge there seems to be in adding properties that are handled by common code instead of the individual drivers
#asahi · 2022-02-09 12:10
kettenisyup
#asahi · 2022-02-09 12:08
kettenisunfortunately that idea is (still) somewhat controversial
#asahi · 2022-02-09 12:07
kettenisideally the DTs would be maintained and reviewed independently from the linux kernel
#asahi · 2022-02-09 12:05
kettenis(and I keep spelling definitely wrong)

plus ›