#asahi · 2021-04-17 14:38
arndright, it seems that the binding is not actually the problem here, it's more a question of how to convince the linux dw-pcie driver to handle it correctly#asahi · 2021-04-17 14:26
arndBut since it seems that Apple actually implemented a proper root complex with multiple ports, we wouldn't need to emulate one here, just adjust the Linux driver to properly deal with multiple ports#asahi · 2021-04-17 14:24
arndso I think the the mvebu pcie binding reflects the root complex that does not actually exist in hardware#asahi · 2021-04-17 14:23
arndIt might be helpful to look at what the older Marvell PCIe driver does, since that emulates a root complex with multiple ports on top of hardware that is somewhat similar to the separate dw-pcie ports#asahi · 2021-04-17 14:22
arndthat makes sense, since you can have nested buses. I think part of the problem is the the DT bindings for PCI largely go back to the pre-pcie days where a lot of things worked differently, and that the dw-pcie hardware in particular only normally implements a single port rather than a root complex#asahi · 2021-04-17 14:18
arndwhich implies using the dw-pcie driver plus some hardware glue, right?#asahi · 2021-04-17 14:17
arndright#asahi · 2021-04-17 14:12
arnda lot of the server chips actually use dw-pcie implementations, but all the phy handling is done in firmware before it gets handed off to the OS#asahi · 2021-04-17 14:11
arndmy hope was that this would all be part of what m1n1 can do before u-boot does the bus scan#asahi · 2021-04-17 14:10
arndIs the link speed set through standard pcie registers, or a hardware specific phy driver?#asahi · 2021-04-17 14:04
arndkettenis: are you sure we actually need these properties? Do you ever need to configure the link speed or trigger a reset using gpio?#asahi · 2021-04-12 13:36
arndsorear: HdkR is working on https://github.com/FEX-Emu/FEX for emulating x86 user space, you can find him on #armlinux. It requires x86 or arm64 hosts today, but I'm fairly sure porting to rv64 would be possible, given how similar it is to aarch64#asahi · 2021-04-12 11:57
arndhas anyone tried talking to uart2 at that baud rate?#asahi · 2021-04-12 11:56
arndI also see that /arm-io/uart2 has a "wlan-debug" child node with a "wlan.uart.baudrate = <0x2f8e7a>" property. I wonder if that is used for debugging the wlan firmware, or if there is a hidden debug mode that exposes a console over wlan#asahi · 2021-04-12 11:52
arndkettenis_: interestingly, the ethernet device in adt encodes the local-mac-address property in the same place that Linux would look for it, in /arm-io/apcie/pci-bridge2/lan-1gb/local-mac-address, but the wireless device has a device in /arm-io/apcie/pci-bridge0/wlan/ withouto such a property, and another device in /arm-io/wlan that has a few properties including local-mac-address but no registers#asahi · 2021-04-12 11:40
arndDoes Apple's bootloader replace that with the actual value before passing it to the OS?#asahi · 2021-04-12 11:40
arndkettenis_: it looks like arch/sparc still emulates the idprom from the m68k sun3 days, while everybody else uses the ibm-style local-mac-address property. ADT (in the link that tmlind cited) apparently contains a local-map-address property but it contains the "macaddr/wifiaddr,syscfg/WMac/6,zeroes/6" string instead of of actual address#asahi · 2021-04-12 09:30
arndkettenis_: for both ethernet and wireless, or just the former?#asahi · 2021-04-12 09:27
arndunless the ethernet driver also wants to read the mac address from the wireless device, in which case using the boot loader would help avoid an inter-dependency between two otherwise unrelated pci network drivers#asahi · 2021-04-12 09:26
arndin that case I think it is best to leave it in the driver and not try to access the PCI device from the boot loader just to pass the address through DT#asahi · 2021-04-12 09:26
arndkettenis_: my reading of the driver is that the mac address is stored in the nvram of the PCI device#asahi · 2021-04-12 09:19
arndagreed, the corellium patch doesn't look too bad as a start. Someone could probably just split it up into logical changes and add a changelog text after figuring out what the changes are in fact needed for. The maintainers might want some things to be handled differently, but most of it is rather straightforward#asahi · 2021-04-09 15:54
arndsorear: it depends on whether marcan (or anyone really) will want to maintain a fork that contains a copy of all the in-flight patches. I'm generally happy with having the work distributed, despite having the centralized repository for the initial work#asahi · 2021-04-09 11:54
arndI pushed all the other branches but apparently missed this one#asahi · 2021-04-09 11:54
arndmarcan: fixed, thanks for pointing it out#asahi · 2021-04-09 08:49
arndok, I sent a link as /msg#asahi · 2021-04-09 08:47
arndmarcan: now would be good, or in the afternoon (in three hours from now or later)#asahi · 2021-04-09 08:43
arndmarcan: if you like, we can do a video chat for me to sign your gpg key and validating your fingerprint, that gets you to one of the three#asahi · 2021-04-09 08:42
arndmarcan: if you want to use git.kernel.org in the long run, don't bother setting up something else now. Using git.asahilinux.org permanently would be fine though#asahi · 2021-04-09 08:29
arndthe process for linux-next is just to email sfr, cc:lkml, asking him to add a git url#asahi · 2021-04-09 08:28
arndyes, https://korg.docs.kernel.org/accounts.html says you need three signatures to have a kernel.org account, though to get your tree into linux-next you can also host it elsewhere without that requirement#asahi · 2021-04-09 08:25
arndI think he generally builds a couple of configurations after each merge, something like x86 and powerpc allmodconfig as well as some arm/arm64/mips/s390 defconfigs#asahi · 2021-04-09 08:24
arndwhich is more thorough than what sfr does#asahi · 2021-04-09 08:23
arndright, and I think anything that gets pulled into linux-next, or that is hosted on git.kernel.org is automatically treated with the 0day build bot testing#asahi · 2021-04-09 08:20
arndonce you have your tree in linux-next, it will be somewhat easier as there is a lot more build testing on it before it gets into the soc tree#asahi · 2021-04-09 08:19
arndwe all try our best to avoid it, but in the end there is no way to prevent it completely, so I don't think anyone will blame you for the mistake#asahi · 2021-04-09 08:18
arndThe for-next branch is something that I don't mind rebasing, this only exists to make life for sfr easier#asahi · 2021-04-09 08:18
arndI have a single for-next branch that gets merged into linux-next, and this one contains (currently) my arm/dt, arm/soc, arm/defconfig, arm/drivers, arm/apple-m1, and arm/newsoc branches#asahi · 2021-04-09 08:17
arndmarcan: I try hard to avoid rebasing anything after it hits my tree, but I occasionally ask submitters to rebase their tree if a pull request contains a show-stopper that prevents me from merging#asahi · 2021-04-09 06:47
arndmarcan: in general, a Fixes: tag does help, but in this case there is no need as it will be applied right on top of the the series, so I think it makes no difference#asahi · 2021-04-09 05:13
arndmarcan: if that is the only patch that is still needed, send the fix to soc@k.o and I'll apply it on top of my branch#asahi · 2021-04-08 17:49
arndthere is really no ideal way when you have pulled in other dependencies, especially more than one. The infrastructure is trying to prevent submitters from sneaking in additional patches by faking a merge base, and we clearly want this, but it also gets in the way in this case#asahi · 2021-04-08 16:05
arndanyway, I'll make it work later when I get back#asahi · 2021-04-08 16:04
arndif you had passed -rc3 as the base for git request-pull, it would have worked, but then I would see all the commits of the tty tree as well#asahi · 2021-04-08 16:03
arndI think I can fix that up later on, just need to do another local merge to create a sensible git history#asahi · 2021-04-08 16:03
arndmarcan: I think I see the problem: you sent the merge commit as the base, but I don't have that merge in my tree locally#asahi · 2021-04-08 16:01
arndneed to make dinner, if you fix it up quickly, I can merge it later#asahi · 2021-04-08 16:00
arnddid you upload the wrong tag?#asahi · 2021-04-08 16:00
arndmarcan: git pull failed#asahi · 2021-04-08 15:59
arndhttps://www.irccloud.com/pastebin/3IwDxc8P/