#
✕ nick:kettenis1
#asahi · 2021-03-29 13:16
kettenis10x00b0: Capability 0x11: Extended Message Signalled Interrupts (MSI-X)
#asahi · 2021-03-29 13:14
kettenis1U-Boot doesn't touch the MSI-related registers
#asahi · 2021-03-29 13:12
kettenis1so I think the generic driver won't be sufficient
#asahi · 2021-03-29 13:12
kettenis1anyway, I think the OS needs to program the registers that control how you split up the MSIs
#asahi · 2021-03-29 13:11
kettenis1think even
#asahi · 2021-03-29 13:11
kettenis1I thing the Linux name for the Ethernet driver is tigon3
#asahi · 2021-03-29 13:10
kettenis1That is ethernet, which is Broadcom BCM57762
#asahi · 2021-03-29 13:10
kettenis10x0000: Vendor ID: 14e4, Product ID: 1682
#asahi · 2021-03-29 13:10
kettenis1That is XHCI, which is Fresco Logic FL1100
#asahi · 2021-03-29 13:09
kettenis10x0000: Vendor ID: 1b73, Product ID: 1100
#asahi · 2021-03-29 13:09
kettenis1That is WiFi, which is Broadcom BCM4378
#asahi · 2021-03-29 13:09
kettenis10x0000: Vendor ID: 14e4, Product ID: 4425
#asahi · 2021-03-29 13:08
kettenis1oh, wifi doesn't support msi-x
#asahi · 2021-03-29 13:06
kettenis1I don't know if Linux makes use of multiple vectors for these devices though
#asahi · 2021-03-29 13:05
kettenis1I beleive all three devices support MSI-X
#asahi · 2021-03-29 13:04
kettenis1so yes ;)
#asahi · 2021-03-29 13:03
kettenis1WiFi+BT is on one of the ports, Ethernet is on its own port and XHCI is on itw own port as well
#asahi · 2021-03-29 13:03
kettenis1The one that is used for onboard stuff has 3 ports
#asahi · 2021-03-29 13:02
kettenis1One is used for the onboard stuff and two are used for Thunderbolt
#asahi · 2021-03-29 13:02
kettenis1There are three PCIe bridges
#asahi · 2021-03-29 13:01
kettenis1so yes, the splitting is enforced by the hardware, but the way the splitting is done is programmable
#asahi · 2021-03-29 13:00
kettenis1For OpenBSD my aplpcie(4) driver simply shares the 32 MSIs among all 3 ports
#asahi · 2021-03-29 13:00
kettenis1Apple (and Corellium) give each port 8 MSIs and leave 8 MSIs on the table
#asahi · 2021-03-29 12:59
kettenis1there are registers in the PCIe bridge block that can be used to distribute the MSIs over the ports
#asahi · 2021-03-29 12:58
kettenis1Each PCIe bridge has multiple (3?) ports
#asahi · 2021-03-29 12:58
kettenis1there are 32 MSIs for each PCIe bridge
#asahi · 2021-03-29 11:19
kettenis1with some adjustments to follow the binding in marcan's linux upstream patch series
#asahi · 2021-03-29 11:18
kettenis1what I'm currently using is mostly a straight copy from corellium
#asahi · 2021-03-29 11:17
kettenis1we need to have (somewhat) stable devicetree bindings before I can upstream that though
#asahi · 2021-03-29 11:14
kettenis1https://github.com/kettenis/u-boot/tree/apple-m1-m1n1
#asahi · 2021-03-29 10:58
kettenis1maz: I'd be interested to hear howyou think the MSI support should be reflected in the device tree
#asahi · 2021-03-29 09:47
kettenis1maz: any success yet with getting PCIe to work in Linux?