RFC: out of sync shadow
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest13 patches 28 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/13 | x86/mm: get_user_pages_fast_atomicarch/ include/ | +43 | 3 replies | |
| 2/13 | KVM: MMU: switch to get_user_pages_fastarch/ arch/ arch/ +2 more | +27 | 5 replies | |
| 3/13 | KVM: MMU: gfn_to_page_atomicinclude/ virt/ | +25 | 0 replies | |
| 4/13 | KVM: MMU: switch prefetch_page to gfn_to_page_atomicarch/ | +21 | 0 replies | |
| 5/13 | KVM: MMU: do not write-protect large mappingsarch/ arch/ | +11 | 2 replies | |
| 6/13 | KVM: MMU: global page keepingarch/ arch/ include/ | +28 | 1 reply | |
| 7/13 | KVM: MMU: mode specific sync_pagearch/ arch/ include/ | +132 | 3 replies | |
| 8/13 | KVM: MMU: record guest root level on struct guest_walkerarch/ arch/ | +9 | 0 replies | |
| 9/13 | KVM: MMU: out of sync shadow corearch/ arch/ arch/ +1 more | +263 | 5 replies | |
| 10/13 | KVM: MMU: sync roots on mmu reloadarch/ arch/ include/ | +50 | 0 replies | |
| 11/13 | KVM: MMU: sync global pages on cr0/cr4 writesarch/ arch/ include/ | +17 | 0 replies | |
| 12/13 | KVM: x86: trap invlpgarch/ arch/ arch/ +2 more | +68 | 1 reply | |
| 13/13 | KVM: MMU: ignore multiroot when unsyncing global pagesarch/ | +1 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [patch 00/13] RFC: out of sync shadow
Keep shadow pages temporarily out of sync, allowing more efficient guest PTE updates in comparison to trap-emulate + unprotect heuristics. Stolen from Xen :)
This version only allows leaf pagetables to go out of sync, for simplicity, but can be enhanced.
VMX "bypass_guest_pf" feature on prefetch_page breaks it (since new PTE writes need no TLB flush, I assume). Not sure if its worthwhile to convert notrap_nonpresent -> trap_nonpresent on unshadow or just go for unconditional nonpaging_prefetch_page.
* Kernel builds on 4-way 64-bit guest improve 10% (+ 3.7% for get_user_pages_fast).
* lmbench's "lat_proc fork" microbenchmark latency is 40% lower (a shadow worst scenario test).
* The RHEL3 highpte kscand hangs go from 5+ seconds to < 1 second.
* Windows 2003 Server, 32-bit PAE, DDK build (build -cPzM 3):
Windows 2003 Checked 64 Bit Build Environment, 256M RAM 1-vcpu:
vanilla + gup_fast: oos
0:04:37.375 0:03:28.047 (- 25%)
2-vcpus:
vanilla + gup_fast oos
0:02:32.000 0:01:56.031 (- 23%)
Windows 2003 Checked Build Environment, 1GB RAM
2-vcpus:
vanilla + fast_gup oos
0:02:26.078 0:01:50.110 (- 24%)
4-vcpus:
vanilla + fast_gup oos
0:01:59.266 0:01:29.625 (- 25%)
And I think other optimizations are possible now, for example the guest can be responsible for remote TLB flushing on kvm_mmu_pte_write().
Please review.