#
…

2022-09-27

00:04dmmcf has quit [Quit: dmmcf]
00:42thelounge7571340 has quit [Remote host closed the connection]
01:00thelounge7571340 has joined #asahi-dev
01:06johey_ has quit [Ping timeout: 480 seconds]
01:08chip_x has quit [Remote host closed the connection]
01:15johey has joined #asahi-dev
01:49SSJGZ has joined #asahi-dev
01:59thelounge7571340 has quit [Remote host closed the connection]
02:09thelounge7571340 has joined #asahi-dev
02:27SSJGZ has quit [Ping timeout: 480 seconds]
03:09thelounge7571340 has quit [Read error: Connection reset by peer]
03:14thelounge7571340 has joined #asahi-dev
03:43kov has quit [Quit: Coyote finally caught me]
04:13thelounge7571340 has quit [Remote host closed the connection]
04:16thelounge7571340 has joined #asahi-dev
05:15thelounge7571340 has quit [Remote host closed the connection]
05:19thelounge7571340 has joined #asahi-dev
06:19thelounge7571340 has quit [Remote host closed the connection]
06:22thelounge7571340 has joined #asahi-dev
06:31chadmed has joined #asahi-dev
06:33zkrx has quit []
06:35zkrx has joined #asahi-dev
06:43SSJGZ has joined #asahi-dev
06:53zzywysm has joined #asahi-dev
06:55zzywysm_ has quit [Read error: Connection reset by peer]
07:02zkrx has quit []
07:04zkrx has joined #asahi-dev
07:16jannaumarcan: except for 2 downstream patch issues no problems with asahi-6.0-rc6-1 on t8103, t6002 and t8112
07:16jannautime for sending the pull request today
07:18jannaufixes are here https://github.com/AsahiLinux/linux/pull/54 - the cpufreq fix is hackish for now
07:20jannauI tried using a cpufreq instance per die but that did not work. Could be easily my fault. I did not look into why it failed as I was doing something else
07:21thelounge7571340 has quit [Remote host closed the connection]
07:26thelounge7571340 has joined #asahi-dev
07:36povik has joined #asahi-dev
08:18andersonwatts3562[m] has joined #asahi-dev
08:26thelounge7571340 has quit [Remote host closed the connection]
08:29thelounge7571340 has joined #asahi-dev
08:30amarioguy2 has joined #asahi-dev
08:34amarioguy has quit [Ping timeout: 480 seconds]
08:40MajorBiscuit has joined #asahi-dev
08:50andersonwatts3562[m]"I'll help 10individuals how to earn $30k in 72 hours from the crypto market. But you will pay me 10% commission when you receive your profit. if interested send me a direct message via WhatsApp by asking me HOW for more details on how to get started
08:50andersonwatts3562[m] +1 (559) 666‑3967
08:53andersonwatts3562[m] was kicked from #asahi-dev by ChanServ [You are not permitted on this channel]
09:10tobhe has quit [Ping timeout: 480 seconds]
09:29thelounge7571340 has quit [Remote host closed the connection]
09:31thelounge7571340 has joined #asahi-dev
09:52roxfan2 has joined #asahi-dev
09:57roxfan has quit [Ping timeout: 480 seconds]
09:59chipxxx has joined #asahi-dev
10:00chipxxx has quit [Remote host closed the connection]
10:00chipxxx has joined #asahi-dev
10:28tobhe has joined #asahi-dev
10:28amarioguy2 has quit [Quit: Leaving]
10:31thelounge7571340 has quit [Remote host closed the connection]
10:32marcanjannau: you can only have a single cpufreq instance, that's why it's done the way it is with all clusters in one device
10:33marcanit's stupid and honestly should not be bound in the DT like that, the DT should just have clusters and the cpufreq driver should pick them all up, but that's not how current cpufreq drivers do it...
10:33marcan(ping robher, any opinions about this? cases where the Linux model demands a single driver instance for what logically are completely independent pieces of hardware?)
10:34thelounge7571340 has joined #asahi-dev
10:35marcaneither way, the answer for the current DT model/driver is to put cpufreq outside of either die bus. could be in a third, shared/common bus.
10:43amarioguy has joined #asahi-dev
10:47Soroush has quit []
10:47Soroush has joined #asahi-dev
10:51amarioguy has quit [Remote host closed the connection]
10:52amarioguy has joined #asahi-dev
10:54chadmed has quit [Remote host closed the connection]
11:23kov has joined #asahi-dev
11:33thelounge7571340 has quit [Remote host closed the connection]
11:42thelounge7571340 has joined #asahi-dev
12:19robhermarcan: Fix the OS? cpufreq has always been problematic though often that's because it's more a virtual device on top of other drivers.
12:42thelounge7571340 has quit [Remote host closed the connection]
12:46thelounge7571340 has joined #asahi-dev
13:09marcanrobher: yeah, and cpufreq-dt sort of does it right, but only works for devices with low-level clock/etc drivers (I tried that approach, was shot down and honestly wasn't a good match)
13:10marcanbut cpufreq-dt does probe magic behind the scenes
13:10marcanso the question is, how should this be represented in the DT? conceptually, the hardware is a cpufreq instance per CPU cluster, and they're independent
13:10marcanbut then Linux needs to probe them all as a single driver
13:11marcandriver/device instance
13:12marcanand we also need to describe the mapping between CPUs and cpufreq clusters
13:16marcanif we want to keep the "virtual device / physical devices" model, then there could be a top-level cpufreq instance (with no reg) with phandles to the individual cluster devices (which have a reg) and then just make the latter not probe by default, or probe as separate drivers but do nothing other than mmap IO and store it as driver data for the top-level device to reference?
13:17marcan(note: parent/child device model doesn't really work for the multi-die devices we have, assuming we don't want to pull this entirely outside of the die bus nodes, since the top-level cpufreq node has to span clusters on both dies)
13:20robhermarcan: The performance-domain binding can do the mapping.
13:21marcanoh, that one's news to me. neat, mediatek uses it.
13:22marcanok, so then that just leaves the question of how to break this up into multiple cluster nodes as it ought to be represented, instead of one magic single device
13:23marcancpufreq-mediatek-hw.yaml is basically what we do today, other than not using that generic binding (which I wasn't aware of)
13:23robhermarcan: For a virtual device, no need to put in DT. Just need the initcall to create platform device based on of_machine_is_compatible() and then it's probe can iterate over nodes by compatible.
13:24marcanof_machine_is_compatible, as in looking for apple stuff?
13:25marcanhuh, didn't know that was the recommended approach
13:25marcanso we'd just have our driver match on apple,arm-platform and then look for the cluster nodes (and bail if it finds none that it can handle, which is what would happen on as-yet unsupported platforms, or if it sees mixed compatibles since that'd be weird)
13:27marcanrobher: does that make sense? if so, I can do that I think, for v3 (I think it was v3 :))
13:27robhermarcan: Yeah, I think so.
13:28marcancool! the last pending thing we had in the prior submission was the cpu->perf domain binding, which was nonstandard, so I think fixing that plus breaking up clusters will solve all remaining issues.
13:45thelounge7571340 has quit [Remote host closed the connection]
13:50thelounge7571340 has joined #asahi-dev
14:01Gaspare has joined #asahi-dev
14:27sven[cpu4] [DCPTracer@/arm-io/dcpext] [mcdp29xx] [dcpav-service-epic0] < group 4 command 4
14:27svenuhhh...
14:27sveni guess the mcdp29xx endpoint in trace_dcp is mislabeled?
14:29marcansven: I think it's mcdp29xx in some devices, but also more generally the dcpav stuff?
14:30marcanas in, mcdp29xx is one implementation of dcpav
14:30svenah, could be
14:30marcanor something like that
14:30marcanthe inheritance hierarchy is a bit confusing
14:30sveni'm starting to get a better understanding of that
14:30svenbut the whole thing is very very weird in general
14:32gladiac has joined #asahi-dev
14:34sven[cpu2] [DCPTracer@/arm-io/dcpext] [mcdp29xx] [dcpdp-service-epic0] < getLocation
14:34sventhat thing seems to do a lot of crap actually
14:36svenso [dptx] has [dcpdp-controller-epic0] and [dcpav-controller-epic0]. [epicep2a] has [dcpdptx-port-epic0] and [dcpdptx-port-epic1]
14:37svenepicep2a is something like dcpdptx then i guess.
14:37svenwhere did these names originally come from btw?
14:39marcanI think I got them either from the advertisement messages or from some rtkit ep name thing? I can't remember...
14:41svenhrm, ok. i'll probably go ahead and rename a few at least in the tracer then so that they make more sense
14:41svenit wouldn't surprise me if the names we have were made up apple. fit to the entire mess the rest of DCP is
14:41sven*by apple
14:45* svenshould've cleaned up his code earlier
14:46gladiac has quit [Quit: k thx bye]
14:50thelounge7571340 has quit [Remote host closed the connection]
14:53thelounge7571340 has joined #asahi-dev
14:56chipxxx has quit [Read error: Connection reset by peer]
15:06marcansven: please feel free to clean up / hack on that as much as you want
15:08svenI still wonder how “I’ll just bring thunderbolt up” ended with me reversing this dumb DCP IPC crap :D
15:13kettenispoor sven
15:18kettenismarcan: I read something about switching the firmware archive from tar to cpio in the backlog?
15:18kettenisthat would be somewhat problematic for OpenBSD as we don't have cpio support on the installation media
15:20kettenisbtw, I did some investigation on the DIT feature (data independent timing) that Apple implements
15:21kettenisdid some benchmarking with "openssl speed" running a performance core and could not find any speed differences that are statistically significant
15:21marcankettenis: interesting, I thought cpio was pretty universal (and it's very simple)
15:23ketteniswell, our installation media are stripped down quite a bit
15:23kettenisfor some architectures we still provide 1.44M floppy images ;)
15:24marcanI don't think that will be necessary for ARM64 ;)
15:26kettenisno kidding
15:27kettenisif switching to cpio has real benefits to you guys, I'll look at what needs to be done to include cpio support
15:28ketteniscurrently the makefile logic that disables it isn't architecture-specific
15:30_jannau_it makes it very easy to provide firmware in the initramfs since linux supports concatenated cpio archives
15:32_jannau_and that makes it easy to update the firmware in the rotfs. the initramfs can update it before it switches to os root
15:35_jannau_I'm using that since a few days for development and it works in practice. that is of course independent of what we do with normal installed systems. I can always repackage the firmware
15:37_jannau_I use it as well to inject matching kernel modules
15:38marcancould just keep the firmware in RAM too, and never put it in the rootfs (which also helps avoid accidental re-distribution by users)
15:39marcankettenis: I suspect a polyglot ustar+cpio archive (that doesn't duplicate the data) can be done that would be accepted by both bsdtar and linux's cpio code, though I'm not sure if we want to push for such an abomination :D
15:40marcanI don't think it can be done fully compliant with both formats by spec, but probably with actual implementations
15:43kettenisprobably not worth it
15:44kettenisfirmware in RAM is an interesting idea, but would of course need driver support and probably DT binding changes
15:46_jannau_I assumed it ment copying the firmware into a tmpfs instead of the root fs
15:47marcanyeah, I meant moving the initramfs firmware into a tmpfs instead of the real rootfs
15:47kettenisnot sure how you'd make that work across OSes
15:47marcanthat would be a OS-specific thing
15:48marcanthe firmware is just a CPIO archive in the ESP; linux bootloaders would load it as an additional initramfs and the main initramfs would handle the copying to a tmpfs mount. for BSD, up to you what you do :)
15:48marcanfor the OS-independent portion the only change is the switch to cpio
15:50marcanand I would also have both versions for a while so as to not break your current installers of course, we have enough ESP space for that
15:52kettenisOpenBSD 7.2 is pretty much done, so yes, if we can keep the tar file until OpenBSD 7.3 is released that would be good
15:52marcanyeah, I can keep it as long as necessary, no problem
15:53thelounge7571340 has quit [Remote host closed the connection]
15:53marcanone nice thing is that cpio is mildly less annoying to parse than tar, so if u-boot needs to grow support for this to pull xHCI firmware (for full USB port support on those Macs that have asmedia controllers) it could do that, instead of having to extract that separately.
15:53mpsI agree with kettenis
15:53mpsthough alpine have cpio by default
15:53marcanof course that's preconditioned on you getting PCIe xHCI support in first ;)
15:54kettenisyeah, I need to get back to that ;)
15:55kettenishaving a proper dma mapping API in u-boot turns out to be rather painful :(
15:55thelounge7571340 has joined #asahi-dev
15:58marcankettenis: btw, for the cpio version, we'd use the /lib/firmware prefix since it gets loaded directly by the kernel (so filenames as lib/firmware/foo). would that be a significant issue for you?
15:58Gaspare has quit [Quit: Gaspare]
15:59marcan(could also go with something generic like vendorfw/ and put a symlink in the initramfs; just something other than the current tarbomb)
16:00kettenisan absolute path would be painful I think
16:01marcannot absolute, just adding at least one extra directory level
16:01kettenisdon't think that would be a problem
16:01marcanit's implicitly absolute as loaded by Linux
16:02marcanI'm leaning towards (/)vendorfw/, since then we could symlink /lib/firmware/vendor to /vendorfw and add that to the kernel load list as I discussed (and it's nice and neutral)
16:02mpsmarcan: what if distro/installation doesn't use initramfs
16:02marcanand different distros might want to use /lib/firmware or /lib/firmware/vendor or something else and this gives them that choice
16:03marcanmps: shrug? then do it the same way you do it now with the tar
16:03mpsmarcan: ok, tar or cpio doesn't matter
16:03marcanthe advantage of the initramfs is you don't have to play games with driver load order
16:05mpsyes, I know pro and cons of using initramfs
16:05marcanif you don't have an initramfs then you absolutely need to ensure that e.g. xhci-pci is not built in, which therefore implies that your distro can't support USB keyboards (or USB anything) until the rootfs loads at least udev, and obviously can't support root-on-USB either with the same kernel
16:05marcanif that's a problem for you, get an initramfs ;)
16:06mpsmarcan: I have it working without initramfs fine and on usb stick also
16:06svenoh god, it does hdcp shit. i hope i can just disable that somehow
16:06marcanno you don't, not with asmedia ports
16:06marcanand it never will
16:06marcanbecause that needs firmware
16:06mpsah, didn't know about asmedia
16:07MajorBiscuit has quit [Ping timeout: 480 seconds]
16:07mpsthanks for info
16:08marcan(and if you don't provide firmware before the driver loads, it will fail to probe, and those machines will be stuck without USB on those ports)
16:08marcanso barring some horrible hackery with /sys/foo/bind, if you want asmedia support and usb boot support, you need an initramfs
16:09mpsfor now it works, will see later to add initramfs to usb media image
16:10mps(experience from chromebooks where initramfs is inconvenient)
16:22svenenough DCP for today. let's see if I can manage to clean up my dcp_altmode.py tomorrow
16:22svenlooks like I actually confused some of the endpoints in there and that i'm missing at least one call which may or may not do anything
16:26marcantoo late, merged :p
16:27svenoh, not in the merge request :D
16:27sventhat's actually the cleaned up version
16:27svenjust need to clean up the .py that tries to drive these endpoints now
16:36jluthra has quit [Remote host closed the connection]
16:36jluthra has joined #asahi-dev
16:44thelounge7571340 has quit [Ping timeout: 480 seconds]
17:08Gaspare has joined #asahi-dev
17:49thelounge7571340 has joined #asahi-dev
18:10FellFromTheSky[m] has joined #asahi-dev
18:20thelounge7571340 has quit [Remote host closed the connection]
18:40Gaspare has quit [Quit: Gaspare]
18:43Gaspare has joined #asahi-dev
18:51thelounge7571340 has joined #asahi-dev
18:53Gaspare has quit [Quit: Gaspare]
18:55Gaspare has joined #asahi-dev
19:08roxfan2 is now known as roxfan
19:22thelounge7571340 has quit [Remote host closed the connection]
19:42cfpsws^ has joined #asahi-dev
19:53thelounge7571340 has joined #asahi-dev
19:58Gaspare has quit [Quit: Gaspare]
20:04thelounge7571340 has quit [Remote host closed the connection]
20:09Gaspare has joined #asahi-dev
20:40espo has quit [Read error: Connection reset by peer]
20:57thelounge7571340 has joined #asahi-dev
20:58thelounge7571340 has quit [Remote host closed the connection]
21:34goldsoultheory has joined #asahi-dev
21:40thelounge7571340 has joined #asahi-dev
21:43Gaspare has quit [Quit: Gaspare]
21:50thelounge7571340 has quit [Remote host closed the connection]
21:57SSJGZ has quit [Ping timeout: 480 seconds]
22:08nela_ has joined #asahi-dev
22:12nela has quit [Ping timeout: 480 seconds]
22:12nela_ is now known as nela
22:24goldsoultheory has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
22:45thelounge7571340 has joined #asahi-dev
22:48thelounge7571340 has quit [Remote host closed the connection]
23:47thelounge7571340 has joined #asahi-dev
23:48thelounge7571340 has quit [Remote host closed the connection]
…