Details

Reviewers

Group Reviewers

Commits

rGdb877a06ec41: vmm: Fix AMD-vi using wrong rid range
rGfc661f1903b4: vmm: Fix AMD-vi using wrong rid range
rGb5c74dfd6434: vmm: Fix AMD-vi using wrong rid range

Summary

The ACPI parsing code around rid range was wrong on assuming there is
only one pair of start/end device id range. Besides, ivhd_dev_parse()
never work as supposed. The start/end rid info was always zero.

Restructure the code to build dynamic-sized tables for each IOMMU softc
holding device entries. The device entries are enumerated to find a
suitable IOMMU unit. Operations on devices not governed (e.g. the IOMMU
unit itself) are no-op from now on. There are also a minor fix on wrong
%b formatting string usage.

Tested on my EPYC 7282.

Diff Detail

Repository

rG FreeBSD src repository

Lint

Lint Not Applicable

Unit

Tests Not Applicable

Event Timeline

khng created this revision.Jun 20 2021, 11:39 AM

Herald added a reviewer: bhyve. · View Herald TranscriptJun 20 2021, 11:39 AM

Herald added subscribers: bcran, rgrimes, imp. · View Herald Transcript

khng requested review of this revision.Jun 20 2021, 11:39 AM

Harbormaster completed remote builds in B39996: Diff 91133.Jun 20 2021, 11:39 AM

Oops..

Harbormaster completed remote builds in B39997: Diff 91134.Jun 20 2021, 11:41 AM

khng planned changes to this revision.Jun 20 2021, 12:21 PM

Handle ACPI_IVRS_TYPE_ALL as well.

Harbormaster completed remote builds in B40002: Diff 91141.Jun 20 2021, 5:01 PM

Fix KASSERT usage.

Harbormaster completed remote builds in B40410: Diff 91997.Jul 9 2021, 8:54 AM

afedorov added a subscriber: afedorov.Jul 9 2021, 8:30 PM

grehan added inline comments.Jul 13 2021, 3:16 AM

sys/amd64/vmm/amd/amdvi_hw.c
1218	Hmmm, rev 3.* has been out for a while now: might be time to update spec references (just an fyi, not necessary for this review).
sys/amd64/vmm/amd/ivrs_drv.c
190	Test (or assignment) inverted ?