RFC rework KVM mmu_shrink() code

Dave Hansen kvm lkml 1 version Jun 15, 2010 In review

Get series with b4

b4 am [email protected]

Versions newest first

v1 RFC Latest Jun 15, 20109 patches 21 repliesIn review

Patches in v1 No applies recorded yet

#	Subject and files	Size	Replies
1/9	abstract kvm x86 mmu->n_free_mmu_pages`arch/x86/kvm/mmu.c` `arch/x86/kvm/mmu.h`	+9 -4	1 reply
2/9	rename x86 kvm->arch.n_alloc_mmu_pages`arch/x86/include/asm/kvm_host.h` `arch/x86/kvm/mmu.c` `arch/x86/kvm/x86.c`	+6 -6	0 replies
3/9	replace x86 kvm n_free_mmu_pages with n_used_mmu_pages`arch/x86/include/asm/kvm_host.h` `arch/x86/kvm/mmu.c` `arch/x86/kvm/mmu.h`	+13 ~~-21~~	2 replies
4/9	create aggregate kvm_total_used_mmu_pages value`arch/x86/kvm/mmu.c`	+23 ~~-10~~	5 replies
5/9	break out some mmu_skrink() code`arch/x86/kvm/mmu.c`	+25 ~~-10~~	0 replies
6/9	remove kvm_freed variable`arch/x86/kvm/mmu.c`	+5 ~~-10~~	0 replies
7/9	make kvm_get_kvm() more robust`include/linux/kvm_host.h` `virt/kvm/kvm_main.c`	+30 -4	0 replies
8/9	reduce kvm_lock hold times in mmu_skrink()`arch/x86/kvm/mmu.c` `kernel/profile.c`	+40 ~~-10~~	1 reply
9/9	make kvm mmu shrinker more aggressive`arch/x86/kvm/mmu.c`	+6 -5	9 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [RFC][PATCH 0/9] rework KVM mmu_shrink() code

This is a big RFC for the moment. These need a bunch more runtime testing.

We've seen contention in the mmu_shrink() function. This patch set reworks it to hopefully be more scalable to large numbers of CPUs, as well as large numbers of running VMs.

The patches are ordered with increasing invasiveness.

These seem to boot and run fine. I'm running about 40 VMs at once, while doing "echo 3 > /proc/sys/vm/drop_caches", and killing/restarting VMs constantly.

Seems to be relatively stable, and seems to keep the numbers of kvm_mmu_page_header objects down.

Read the cover letter in its thread