Skip to contentSkip to conversation

some kvm-kmod issues

PATCH In review Zachary Amsden 2 participants 3 messages May 26, 2010 KVM

lore
Share & tools
lore
#0 lore raw

cover some kvm-kmod issues

I found some issues with kvm-kmod from recently pulled tree on F13 (2.6.33) kernel. I am not 100% sure the version numbers are right in all of these fixes, but at least it compiles and seems to run VMs again.

The FPU change and the iommu change are, well, rather scary. They could use some eyes and suggestions.

Zach

#1 lore raw
Quoted 5 lines · Zachary Amsden

I found some issues with kvm-kmod from recently pulled tree on F13 (2.6.33) kernel. I am not 100% sure the version numbers are right in all of these fixes, but at least it compiles and seems to run VMs again.

The FPU change and the iommu change are, well, rather scary. They could use some eyes and suggestions.

Zach

Well, this apparently doesn't work. The module builds, load, and runs non-paging mode fine, but things appear to go haywire just around when paging gets activated. Any recent changes in the module (large page, TLB contract) that might be relying on addition kernel support?

Or just too long since I pulled mainline. I'll try a bsect and see where I get.

Zach

#2 lore raw
Quoted 15 lines · Zachary Amsden
Quoted 8 lines · Zachary Amsden

I found some issues with kvm-kmod from recently pulled tree on F13 (2.6.33) kernel. I am not 100% sure the version numbers are right in all of these fixes, but at least it compiles and seems to run VMs again.

The FPU change and the iommu change are, well, rather scary. They could use some eyes and suggestions.

Zach

Well, this apparently doesn't work. The module builds, load, and runs non-paging mode fine, but things appear to go haywire just around when paging gets activated. Any recent changes in the module (large page, TLB contract) that might be relying on addition kernel support?

Quoted 4 lines

Or just too long since I pulled mainline. I'll try a bsect and see where I get.

Zach

Thanks a lot for the patches! Just yesterday, I finally started hacking on this as well. I have some FPU wrapper running (see next branch) that seems to run fine - but is in some parts also more complicated as required (and has a build problem on very old kernels).

Will comment on individual patches.

Jan