i have tried builds for F2A85M since version 4.8 dating back to mid 2018,
AMD-A8 5500 works properly with IOMMU enabled, still some stabilty problems (full system freezes).
AMD A8-6600K shows the described behavior
I also tried that manual microcode update in the past, didnt change anything there, i could retry it for the current build.
Serial output gives no information about malfunctioning IOMMU or something, the strugle begins as without IOMMU the system is notable to initialize add-on videocards or the iGPU.
Am 17.04.19 um 10:50 schrieb Mike Banon:
On Tue, Apr 16, 2019 at 11:23 PM Kinky Nekoboi kinky_nekoboi@nekoboi.moe wrote:
Amd-Vi loopback errors at linux-boot, several peripherie is not working.
Surely it needs only minor changes... is there anybody working at this port at the moment?
Hi there Nekoboy (sorry if I'm repeating my older questions, I forgot). Please tell:
- IOMMU - did it ever work on F2A85M board? If you're unsure, could
you try building the older coreboot just to check it? 2) Have you installed the unofficial microcode patches? If not, please run this ./ucode.sh script at your coreboot directory: http://dangerousprototypes.com/docs/Lenovo_G505S_hacking#AMD_microcode_updat... to conveniently and securely patch your cloned coreboot with these updated microcodes, then try running Linux and tell what microcode version do you see (confirmation of successful microcode install) and check if IOMMU got fixed by it.
Step "2)" should be tried both for the latest coreboot and some older version. Yes, a bit time consuming, but hopefully you could test it well. By the way, I got A88XM-E which is not merged to coreboot yet but is based on F2A85M port, however I have a lot of parallel projects and didn't have time to even test if my new (actually used) board is working, so you'll have to rely on your own for a while.
Best regards, Mike Banon _______________________________________________ coreboot mailing list -- coreboot@coreboot.org To unsubscribe send an email to coreboot-leave@coreboot.org