diff options
| author | Avi Kivity <avi@qumranet.com> | 2007-09-15 17:34:36 +0300 |
|---|---|---|
| committer | Avi Kivity <avi@qumranet.com> | 2007-10-13 10:18:29 +0200 |
| commit | 0967b7bf1c22b55777aba46ff616547feed0b141 (patch) | |
| tree | f855d1f25943dfd8cb6e40f0e81744f1e32d98e0 /drivers/kvm/kvm_main.c | |
| parent | 535eabcf0e55804b53d27fe45217d874b19bcfa9 (diff) | |
| download | blackbird-op-linux-0967b7bf1c22b55777aba46ff616547feed0b141.tar.gz blackbird-op-linux-0967b7bf1c22b55777aba46ff616547feed0b141.zip | |
KVM: Skip pio instruction when it is emulated, not executed
If we defer updating rip until pio instructions are executed, we have a
problem with reset: a pio reset updates rip, and when the instruction
completes we skip the emulated instruction, pointing rip somewhere completely
unrelated.
Fix by updating rip when we see decode the instruction, not after emulation.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Diffstat (limited to 'drivers/kvm/kvm_main.c')
| -rw-r--r-- | drivers/kvm/kvm_main.c | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c index 3b046507ebc6..353e58527d15 100644 --- a/drivers/kvm/kvm_main.c +++ b/drivers/kvm/kvm_main.c @@ -1815,8 +1815,6 @@ static int complete_pio(struct kvm_vcpu *vcpu) io->count -= io->cur_count; io->cur_count = 0; - if (!io->count) - kvm_x86_ops->skip_emulated_instruction(vcpu); return 0; } @@ -1876,6 +1874,8 @@ int kvm_emulate_pio (struct kvm_vcpu *vcpu, struct kvm_run *run, int in, memcpy(vcpu->pio_data, &vcpu->regs[VCPU_REGS_RAX], 4); kvm_x86_ops->decache_regs(vcpu); + kvm_x86_ops->skip_emulated_instruction(vcpu); + pio_dev = vcpu_find_pio_dev(vcpu, port); if (pio_dev) { kernel_pio(pio_dev, vcpu, vcpu->pio_data); @@ -1938,6 +1938,9 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, vcpu->run->io.count = now; vcpu->pio.cur_count = now; + if (vcpu->pio.cur_count == vcpu->pio.count) + kvm_x86_ops->skip_emulated_instruction(vcpu); + for (i = 0; i < nr_pages; ++i) { mutex_lock(&vcpu->kvm->lock); page = gva_to_page(vcpu, address + i * PAGE_SIZE); |
