With this patches loading 100M initrd takes ~10s instead of ~40s without.
Changelog:
v3->v4:
- check rcx for zero after applying address mask
- check for page boundary after linearizing address
v2->v3
- Fix incorrect size parameter for linearize() notices by Marcelo.
- Get rid of linearize() callback in emulation ops.
v1->v2
- add segment check and mask rcx/rdi correctly drying increment.
Gleb Natapov (5):
Provide userspace IO exit completion callback.
KVM: emulator: make x86 emulation modes enum instead of defines
KVM: emulator: move some address manipulation function out of
emulator code.
KVM: emulator: make linearize() callable from outside of emulator
KVM: Provide fast path for "rep ins" emulation if possible.
arch/x86/include/asm/kvm_emulate.h | 38 +++--
arch/x86/include/asm/kvm_host.h | 32 ++++
arch/x86/kvm/emulate.c | 133 ++++++++--------
arch/x86/kvm/svm.c | 20 ++-
arch/x86/kvm/vmx.c | 25 ++-
arch/x86/kvm/x86.c | 309 ++++++++++++++++++++++++++++--------
6 files changed, 399 insertions(+), 158 deletions(-)
--
1.7.10
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
