diff --git a/website/content/en/status/report-2024-01-2024-03/ten64-whle-honeycomb.adoc b/website/content/en/status/report-2024-01-2024-03/ten64-whle-honeycomb.adoc new file mode 100644 --- /dev/null +++ b/website/content/en/status/report-2024-01-2024-03/ten64-whle-honeycomb.adoc @@ -0,0 +1,37 @@ +=== Ten64, WHLE-LS1, and HoneyComb + +Links: + +link:https://wiki.freebsd.org/BjoernZeeb/[My wiki page with links to some status] URL: link:https://wiki.freebsd.org/BjoernZeeb/[] + + +Contact: Bjoern A. Zeeb + +Solid-Run's HoneyComb, Traverse Technologies's Ten64 and some versions of Conclusive Engineering's WHLE-LS1 all are NXP based platforms with the Data Path Acceleration Architecture Gen2 (DPAA2). + +Work has happened to support or improve support for peripherals on these boards. + +For DPAA2 I have local changes which will need review (or further discussion): +* Cleanup of memac (MDIO) code reducing bus attachment (ACPI and FDT specific) code into more common code. +* Cleanup of MC bus attachment code (again ACPI, FDT). +* For reasons of mii_fdt.c support on some PHYs on FDT-based platforms restructure MAC/MII code and mostly migrate it out of the network interface (NI). +* Improve Dmitry Salychev's (dsl) initial SFF/SFP code, prototyping a bus similar to MII for SFP with the hope that with more work it can grow into a larger, general FreeBSD framework and hooked it up to DPMAC. +* With this, minimal support (still fairly hacked up) for "managed" SFP+ mode (using the Ten64 terminology) is usable on FDT-based systems using DAC and fiber cables. +* Add more sysctl statistics to DPMAC and NI. +In short, I mostly cleaned up some of the mess I contributed to during the initial bringup. + +For the LS1088a based WHLE-LS1 systems changes include: +* device-tree file updates. +* Added support for the PCA9546 I2C Switch (committed). +* Added basic support for the PCAL6524 24-bit Fm+ I2C-bus/SMBus I/O expander. +* Added basic support for the PCA9633 4-bit Fm+ I2C-bus LED driver to drive the status LEDs. +* Added support to program the man:rgephy[4] LEDs (which needs to be validated). +* Started testing the eMMC with MMCCAM and GENERIC but had trouble (needs further investigation, seemed fine from firmware for updates). +* Tested one of three PCIe slots and USB fine. + +For the Ten64: +* Most of the basic lifting happened a while ago and it has generally been usable. +* Detecting the VSC8514 PHY as such went in end of last year. +* Used as the default platform to test the DPAA2 changes and SFP/SFP+ code. + +In addition Pierre-Luc Drouin has overhauled the Vybrid I2C support now attaching and working on both FDT- and ACPI-based systems (committed). + +Sponsor: Traverse Technologies (Ten64 hardware a while ago, support)