AMD-Vi (IOMMU)support for bhyve
AbandonedPublic
Actions

Authored by rgrimes on Mar 7 2017, 9:08 PM.

Details

Reviewers

grehan
anish

Summary

This add support for the AMD-Vi feature into bhyve

Test Plan

Can anyone come up with a good way to automate testing of this?

Diff Detail

Lint

Lint Skipped

Unit

Tests Skipped

Event Timeline

rgrimes updated this revision to Diff 26068.Mar 7 2017, 9:08 PM

rgrimes retitled this revision from to AMD-Vi (IOMMU)support for bhyve.

rgrimes updated this object.

rgrimes edited the test plan for this revision. (Show Details)

rgrimes added reviewers: grehan, anish.

rgrimes added a subscriber: editor_callfortesting.org.

Thanks for doing the work on adding that feature. I have a few textual suggestions for comments and user visible error messages.

amd64/vmm/amd/amdvi_hw.c
281	s/retsrating/restarting/
396	s/address/addresses/
521	s/Retry few/Retry a few/
583	s/readble/readable/
762	s/capability/capabilities/
1063	s/so can/so we can/
1085	s/hw.vmm.amdvi_enable 1/hw.vmm.amdvi_enable=1/
1373	s/suppress rest/suppress the rest/
amd64/vmm/amd/ivrs_drv.c
127	s/in system/in the system/
390	s/minum/minimum/
413	s/attachs/attach/
448	s/couldn't initialised/couldn't be initialised/

Thanks a lot for testing, review and code review. My new ssh key is in place, will get back to work.

amd64/vmm/amd/amdvi_hw.c
169	s/abent/absent
1386	Better XXX: Page table valid[TV] bit must be set even if host page tables are not enabled.
amd64/vmm/amd/amdvi_priv.h
28	This will need alignment fixes. I can send a patch or we can fix it later.
amd64/vmm/amd/ivrs_drv.c
140	We will be adding code to find IVMD devices, that's why commented out.

Can we close this one since we have https://reviews.freebsd.org/D10049?

Fine by me. I didn't check, but were all the comments in this applied to the new revision ?

I think I have taken care of all comments in https://reviews.freebsd.org/D10049

amd64/vmm/amd/ivrs_drv.c
127	Fixed in https://reviews.freebsd.org/D10049
140	Removed for now in https://reviews.freebsd.org/D10049
390	Fixed in https://reviews.freebsd.org/D10049
413	Fixed in https://reviews.freebsd.org/D10049
448	Fixed in https://reviews.freebsd.org/D10049

I was going to close this but I see several things from bcr that are not marked as fixed by anish in D10049, can you please review all of bcr's nits and make sure we have those fixed in D10049 before I abandon this?