#asahi-dev · 2026-08-20 12:06
svenmakes sense though, everything that's pure usb4 will work#asahi-dev · 2026-08-20 11:58
sveni guess that makes sense, just didn't expect that#asahi-dev · 2026-08-20 11:58
svenwait, this actually works through a usb4 hub?#asahi-dev · 2026-08-20 11:58
svenhahahaha :D#asahi-dev · 2026-08-20 09:00
svenah, no, that was a *different* quirk and it's actually unconditional now#asahi-dev · 2026-08-20 08:57
sveniirc it stuffs the channel into the upper bits so it's either firmware or board specific i think#asahi-dev · 2026-08-20 08:56
svencould be. the nice thing is that the quirk won't break anything. i think my v1 even had the &0xff unconditionally #asahi-dev · 2026-08-20 08:52
svennice#asahi-dev · 2026-08-20 08:52
svenso with that the coex issues should be gone then?#asahi-dev · 2026-08-20 08:52
svenhah#asahi-dev · 2026-08-20 08:51
sveni guess coex is a different thing then#asahi-dev · 2026-08-20 08:50
svenah, that was for device scanning and still didn't go anywhere upstream :(#asahi-dev · 2026-08-20 08:49
svendid anyone ever make any progress on that coex thing? there was some vendor specific command that had to be sent but I vaugely remember that that just went nowhere with upstream#asahi-dev · 2026-08-20 08:44
svenstill amazing we just missed an entire controller for that fix and no one noticed so far#asahi-dev · 2026-08-20 08:38
svensomeone should just implement usb4net in m1n1 so that we can load a kernel with a full rootfs in <1s!#asahi-dev · 2026-08-20 06:00
sventhe pages from that page_pool can just stay mapped#asahi-dev · 2026-08-20 06:00
svensomething like drivers/net/ethernet/freescale/fec_main.c, commit 95698ff6177b should work for usb4net as well I think#asahi-dev · 2026-08-20 05:54
svengonna prototype something this weekend#asahi-dev · 2026-08-20 05:53
sveni've been reading some network code and i think i can make usb4net re-use dma mappings in the rx path as well and still retain zero copy#asahi-dev · 2026-08-19 20:56
svenand then we get to play another game of spot the difference that’s gonna end up in something dumb like the second dart stream 🙃#asahi-dev · 2026-08-19 20:55
svenand then Intel/Linux where that thing above happens from the Linux side as well #asahi-dev · 2026-08-19 20:55
svenyes, from the macos side #asahi-dev · 2026-08-19 20:30
svenhow fast is this on non-apple machines again? :>#asahi-dev · 2026-08-19 20:29
sven28gbps even#asahi-dev · 2026-08-19 20:29
svenwow, 27.5gbps during peaks#asahi-dev · 2026-08-19 20:26
svenotherwise we should've seen traffic in one direction before we found that secondary stream#asahi-dev · 2026-08-19 20:26
svenchaos_princess: i think both rx and tx are on stream 2 and just the control ring is on stream 1#asahi-dev · 2026-08-19 20:21
svenhah, nice!#asahi-dev · 2026-08-19 20:00
svenah :/#asahi-dev · 2026-08-19 19:58
sven*remap#asahi-dev · 2026-08-19 19:58
sveni guess stream map internally then doesn't adjust the stream id but just mirrors the configuration (and invalidations) :(#asahi-dev · 2026-08-19 19:57
sveni wonder if we can make the RX path re-use buffers that stay mapped the entire time as well#asahi-dev · 2026-08-19 18:47
svennice, thanks for confirming!#asahi-dev · 2026-08-19 18:28
sveni think we need to add a capable function that returns true for IOMMU_CAP_DEFERRED_FLUSH#asahi-dev · 2026-08-19 18:24
sventhat sysfs write just bypasses that check :D#asahi-dev · 2026-08-19 18:24
svenlast i checked that won't work because we claim we can't do dma-fq#asahi-dev · 2026-08-19 18:19
sveni think all we need to do it set some bit in capabilities somewhere since all we can do is "invalidate everything" anyway#asahi-dev · 2026-08-19 18:19
sveni guess we should add support for dma-fq to dart anyway#asahi-dev · 2026-08-19 18:09
svenat least for the cpu pagetables you don't need to invalidate anything after adding new mappings, only when deleting old ones#asahi-dev · 2026-08-19 18:09
svennice!#asahi-dev · 2026-08-19 17:01
svenbut in theory datastatus/status should come from that EUDO message so this should hopefully be fine#asahi-dev · 2026-08-19 17:01
svenxnu takes it directly from datastatus/status but i can't pass those through to anything outside tipd#asahi-dev · 2026-08-19 16:58
svenchaos_princess: if you have some time, can you try with your cables again as well? i've completely rewritten the cable vse nonsense and it should match whatever xnu does now#asahi-dev · 2026-08-19 16:51
sven"good"#asahi-dev · 2026-08-19 16:16
sveni'm not too concerned if it works because pipehandler is just weird, but i thought we got rid of most of these warnings by now#asahi-dev · 2026-08-19 16:15
svenodd#asahi-dev · 2026-08-19 16:13
svenmaybe we don't need the lock dance on disconnect? we're about to reset the entire thing anyway#asahi-dev · 2026-08-19 16:13
svenyeah, i hit that as well#asahi-dev · 2026-08-19 16:11
svenhm, i see that on tbt disconnect here#asahi-dev · 2026-08-19 16:08
svenbut either, happy it's fixed :D