Skip to content

x86: KVM vdso and clock improvements

Andy Lutomirski kvm linux-mm lkml 1 version NAK

Status from this message: NACK.

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest5 patches 16 repliesNAK

Patches in v1

#Subject and filesSizeRepliesApplied
1/5x86/kvm: On KVM re-enable (e.g. after suspend), update clocksarch/x86/kvm/x86.c+3 -728 replies
1/5x86/kvm: On KVM re-enable (e.g. after suspend), update clocksarch/x86/kvm/x86.c+3 -720 replies
2/5x86, vdso, pvclock: Simplify and speed up the vdso pvclock readerarch/x86/entry/vdso/vclock_gettime.c+46 -354 replies
2/5x86, vdso, pvclock: Simplify and speed up the vdso pvclock readerarch/x86/entry/vdso/vclock_gettime.c+46 -350 replies
3/5x86/vdso: Get pvclock data from the vvar VMA instead of the fixmaparch/x86/entry/vdso/vclock_gettime.c arch/x86/entry/vdso/vdso-layout.lds.S arch/x86/entry/vdso/vdso2c.c +4 more+41 -131 reply
3/5x86/vdso: Get pvclock data from the vvar VMA instead of the fixmaparch/x86/entry/vdso/vclock_gettime.c arch/x86/entry/vdso/vdso-layout.lds.S arch/x86/entry/vdso/vdso2c.c +4 more+41 -130 replies
4/5x86/vdso: Remove pvclock fixmap machineryarch/x86/entry/vdso/vclock_gettime.c arch/x86/entry/vdso/vma.c arch/x86/include/asm/fixmap.h +3 more+1 -413 replies
4/5x86/vdso: Remove pvclock fixmap machineryarch/x86/entry/vdso/vclock_gettime.c arch/x86/entry/vdso/vma.c arch/x86/include/asm/fixmap.h +3 more+1 -410 replies
5/5x86/vdso: Enable vdso pvclock access on all vdso variantsarch/x86/entry/vdso/vclock_gettime.c+40 -511 reply
5/5x86/vdso: Enable vdso pvclock access on all vdso variantsarch/x86/entry/vdso/vclock_gettime.c+40 -510 replies

Reviews trailers per patch

8 of 10 patches have a Reviewed-by or Acked-by.

PatchPaolo Bonzini
1/5
1/5
2/5R-b
2/5R-b
3/5A-b
3/5A-b
4/5A-b
4/5A-b
5/5A-b
5/5A-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH 0/5] x86: KVM vdso and clock improvements

NB: patch 1 doesn't really belong here, but it makes this a lot easier for me to test. Patch 1, if it's okay at all, should go though the kvm tree. The rest should probably go through tip:x86/vdso once they're reviewed.

I'll do a followup to enable vdso pvclock on 32-bit guests. I'm not currently set up to test it. (The KVM people could also do it very easily on top of these patches.)

Andy Lutomirski (5):
x86/kvm: On KVM re-enable (e.g. after suspend), update clocks x86, vdso, pvclock: Simplify and speed up the vdso pvclock reader x86/vdso: Get pvclock data from the vvar VMA instead of the fixmap x86/vdso: Remove pvclock fixmap machinery
x86/vdso: Enable vdso pvclock access on all vdso variants

 arch/x86/entry/vdso/vclock_gettime.c  | 151 ++++++++++++++++------------------
 arch/x86/entry/vdso/vdso-layout.lds.S |   3 +-
 arch/x86/entry/vdso/vdso2c.c          |   3 +
 arch/x86/entry/vdso/vma.c             |  14 ++++
 arch/x86/include/asm/fixmap.h         |   5 --
 arch/x86/include/asm/pvclock.h        |  14 ++--
 arch/x86/include/asm/vdso.h           |   1 +
 arch/x86/kernel/kvmclock.c            |  11 ++-
 arch/x86/kernel/pvclock.c             |  24 ------
 arch/x86/kvm/x86.c                    |  75 +----------------
 10 files changed, 110 insertions(+), 191 deletions(-)
Signature · 1 line
2.5.0