#asahi · 2021-04-11 15:09
kettenis_don't think that is a sensible example to follow#asahi · 2021-04-11 15:01
kettenis_if you want the clock nodes to be children of that, it also needs a "simple-mfd" compatible#asahi · 2021-04-11 15:00
kettenis_anyway, to create a regmap you needs a "syscon" compatible node#asahi · 2021-04-11 14:58
kettenis_I never understood why#asahi · 2021-04-11 14:58
kettenis_sunxi went from a node-per-clock to a single clock node a few years ago#asahi · 2021-04-11 14:27
kettenis_in the "one node per clock" model you can probably avoid ioremapping the same page multiple times by using a regmap#asahi · 2021-04-11 14:19
kettenis_btw, just pushed a draft pull request for the PCIe initialization code for m1n1#asahi · 2021-04-11 14:18
kettenis_but the "one node per clock module" model seems to be preferred by the Linux maintainers these days#asahi · 2021-04-11 14:17
kettenis_the benfit of the "one node per clock" strategy is that you don't have to define clock IDs and can make the clock hierarchy explicit in the device tree instead of coding it in the clock driver#asahi · 2021-04-11 14:15
kettenis_assigned-clock-parents is to initially set the muxes when there are multiple parent clocks to choose from#asahi · 2021-04-10 20:59
kettenis_hmm, the llvm package is missing from the last OpenBSD arm64 package snapshot#asahi · 2021-04-10 20:49
kettenis_coding-style wise, is it something exitstng (in emacs) you're using?#asahi · 2021-04-10 20:43
kettenis_ah, yes, that helps#asahi · 2021-04-10 20:43
kettenis_a trick that is often used is to use a shadow framebuffer#asahi · 2021-04-10 20:42
kettenis_for framebuffer scrolling trying to avoid reading back the framebuffer contents is typically a good idea#asahi · 2021-04-10 20:36
kettenis_marcan: thanks#asahi · 2021-04-10 19:41
kettenis_I'll send a pull request after https://github.com/AsahiLinux/m1n1/pull/44 has been merged#asahi · 2021-04-10 19:19
kettenis_yay, I have a m1n1/u-boot combo that doesn't need the tunables#asahi · 2021-04-09 20:56
kettenis_wel heel erg summier#asahi · 2021-04-09 10:47
kettenis_vhe also made it into the arm64 tree it seems#asahi · 2021-04-08 11:51
kettenis_sven: heh#asahi · 2021-04-08 11:51
kettenis_maybe there should be a udelay(1) in the loop#asahi · 2021-04-08 11:50
kettenis_as the timeout is simply a loop count now#asahi · 2021-04-08 11:49
kettenis_poll32 may need a bit more thought though#asahi · 2021-04-08 11:49
kettenis_in particular poll32 and pmgr#asahi · 2021-04-08 11:48
kettenis_marcan: my current pcie code in m1n1 also depends on bits of https://github.com/AsahiLinux/m1n1/pull/36#asahi · 2021-04-07 20:52
kettenis_sven: thanks, I'll give it a go#asahi · 2021-04-07 20:08
kettenis_the first approach would make most sense in my context#asahi · 2021-04-07 20:05
kettenis_could just pull what you have now and I add the function I need as part of the pcie setup code#asahi · 2021-04-07 20:05
kettenis_not sure what marcan wants to do#asahi · 2021-04-07 20:03
kettenis_but the reg property is in the parent node#asahi · 2021-04-07 20:03
kettenis_it is based on a reg property#asahi · 2021-04-07 20:02
kettenis_so I'd just call adt_get_reg() on the parent node myself#asahi · 2021-04-07 20:02
kettenis_I need that base address later on anyway#asahi · 2021-04-07 20:01
kettenis_if I have one that takes a base address, I can use that#asahi · 2021-04-07 19:58
kettenis_sven: I'll probably need a tunables_apply_local() variant that takes an explicit base address#asahi · 2021-04-07 19:44
kettenis_version 5.30a-ea15 to be specific#asahi · 2021-04-07 19:44
kettenis_in other news, there is more evidence that the pcie root ports are based on synopsys designware crap#asahi · 2021-04-07 19:43
kettenis_not sure how linux handles DMA of network packets, but the problem probably already exists in that area when two packets are stored in a single page#asahi · 2021-04-07 19:41
kettenis_and may end up having to expose the same set of 4k pages multiple times#asahi · 2021-04-07 19:40
kettenis_it means that you're exposing more memory to the device than desirable#asahi · 2021-04-07 10:33
kettenis_arnd: yes adding a property like that would be easy#asahi · 2021-04-07 09:53
kettenis_that doesn't mean m1n1 can't be extended later to fully bring up pcie#asahi · 2021-04-07 09:47
kettenis_that means the device tree would contain enough information for linux to bring up pcie if booted directly from m1n1#asahi · 2021-04-07 09:45
kettenis_marcan: based on the discussion from yesterday I think my immediate goal would be to let m1n1 do some minimal RC port initialization and let u-boot take it from there#asahi · 2021-04-06 15:50
kettenis_anyway, back on topic, I'm going to play a bit more with having m1n1 doing some of the RC port initialization and having u-boot do the remainder#asahi · 2021-04-06 15:42
kettenis_and difficult to build a transmatter for those bands that doesn't spill over into the neighbouring spectrum#asahi · 2021-04-06 15:41
kettenis_difficult to build a broadband receiver without running into the ISM bands#asahi · 2021-04-06 15:31
kettenis_yeah, it might not fully meet the regulation for RFI, but as long as you stay away from our radio telescopes I ewon't complain#asahi · 2021-04-06 15:27
kettenis_the broadcom wifi needs a "nvram.txt" file with settings that are specefic to how the chip is integrated