#
✕ nick:sven
#asahi-dev · 2021-12-23 13:38
sventhat behaves on the T2
#asahi-dev · 2021-12-23 13:38
svenRedecorating[m]: could you install fio (apt-get install fio on ubuntu should work) and run fio --ioengine=sync --direct=0 --fsync_on_close=1 --randrepeat=0 --nrfiles=1 --name=randwrite --rw=randwrite --bs=4K --size=1G --end_fsync=1 --fallocate=none --overwrite=0 --numjobs=1 --sync=1 --directory=[SOME DIRECTORY ON THE NVME DRIVE]? i'm curious how
#asahi-dev · 2021-12-23 09:38
sventpw_rules: there’s a 4K series somewhere on LKMl that works fine but needs some more adjustments to be accepted upstream
#asahi-dev · 2021-12-22 08:55
sventhose gp?? keys could also use some more reverse engineering
#asahi-dev · 2021-12-22 08:53
sveni had to change the api a bit and it even parses that crash log now ;)
#asahi-dev · 2021-12-22 08:52
svenmarcan: just make sure to use rtkit from nvme/dev or that asahi-new-nvme branch once you tackle SMC.
#asahi-dev · 2021-12-21 20:36
sveni usually submit what I think is the best way. chances are the first submission has to be rewritten anyway ;)
#asahi-dev · 2021-12-20 19:21
svenyes, i didn't change it in nvme/dev
#asahi-dev · 2021-12-20 19:15
svenjannau: i think i have whatever was in the asahi branch in asahi-with-new-nvme
#asahi-dev · 2021-12-20 17:22
svenbut i dunno when it prints that message
#asahi-dev · 2021-12-20 17:22
svenbut 80ms shouldn't really be a long time :D
#asahi-dev · 2021-12-20 17:21
svenwithout deleting those it'll complain about an "unclean shutdown" and how it needs to "recover something" on the next boot
#asahi-dev · 2021-12-20 17:21
svenah, i've seen that before. it for some reason took ages to delete the submission and completion queues
#asahi-dev · 2021-12-20 17:11
svenouch.
#asahi-dev · 2021-12-20 15:04
svenah, nice, didn't see that one
#asahi-dev · 2021-12-20 15:03
svenokay, that sounds like the two things I fixed
#asahi-dev · 2021-12-20 14:41
svenmarcan: i'm on a train and can't test it right now but https://github.com/AsahiLinux/linux/tree/asahi-with-new-nvme should be the asahi branch without the old and with the new nvme driver
#asahi-dev · 2021-12-20 11:46
svenJamie[m]1: i think the pagetable should be at 0x103_60e1_c000 and 0x100_3555_4000
#asahi-dev · 2021-12-20 11:45
svenoh. wait. i'm just still confused
#asahi-dev · 2021-12-20 11:44
svenhrm. then my guess that the pagetable is at 0x10_3555_4000 doesn't work :-(
#asahi-dev · 2021-12-20 11:42
svenwhere did RAM start on the MAX again? 0x10_0000_0000?
#asahi-dev · 2021-12-20 11:37
svenactually, yes. i'm just confused :D
#asahi-dev · 2021-12-20 11:37
svenhow did i get that number :/
#asahi-dev · 2021-12-20 11:37
svener. no.
#asahi-dev · 2021-12-20 11:36
svenI think TTBRs are at 0x1400
#asahi-dev · 2021-12-20 11:29
svenand those also support 64 sids
#asahi-dev · 2021-12-20 11:28
svenmarcan: / Jamie[m]1: looks even different from the thunderbolt darts on the m1
#asahi-dev · 2021-12-20 10:21
svenworks for me :D
#asahi-dev · 2021-12-20 10:21
svenso what you’re saying is I should just never submit another version of that 4K patch for the good of the entire internet? :p
#asahi-dev · 2021-12-20 06:35
svenprobably
#asahi-dev · 2021-12-20 06:34
svenA few
#asahi-dev · 2021-12-20 06:34
svenyeah
#asahi-dev · 2021-12-20 06:33
svenit’s not arm’s SMMU
#asahi-dev · 2021-12-20 06:33
svenSMMU is this weird thing that looked like it had that pagetable cache
#asahi-dev · 2021-12-20 06:32
svenI was kinda hoping we could just ignore all this until we get to thunderbolt :(
#asahi-dev · 2021-12-20 06:31
sveneven the first TTBR does like 2 or 3 bits more
#asahi-dev · 2021-12-20 06:31
sven*support more
#asahi-dev · 2021-12-20 06:30
svenyeah, but it could already supoeot
#asahi-dev · 2021-12-20 06:30
svenit also looks like it can do one more stage for some reason?
#asahi-dev · 2021-12-20 06:30
svenyeah
#asahi-dev · 2021-12-20 06:29
svenI’m surprised it’s also used for avd though
#asahi-dev · 2021-12-20 06:28
svenit’s also used for thunderbolt
#asahi-dev · 2021-12-20 06:28
svenlook at strings of the t8110 driver
#asahi-dev · 2021-12-20 06:27
svenI think they actually did
#asahi-dev · 2021-12-20 05:59
svenanyway, nvme is simple enough. probably even simpler than rtkit
#asahi-dev · 2021-12-20 05:57
svenaw :(
#asahi-dev · 2021-12-20 05:41
svenor we just write our own kmutil and use that auxkc hack :>
#asahi-dev · 2021-12-19 20:00
svenprobably doesn’t matter in the end though
#asahi-dev · 2021-12-19 20:00
svenI’m still not 100% about the nvme compatibles. I think the ADT identifies it as ANS2 but ioreg calls it ans3
#asahi-dev · 2021-12-19 19:52
svennice, thanks!

plus ›