| Message ID | 20190124175845.15926-6-sean.j.christopherson@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: VMX: Move vCPU-run to proper asm sub-routine | expand |
On Thu, Jan 24, 2019 at 9:59 AM Sean Christopherson <sean.j.christopherson@intel.com> wrote: > > Load RDX with the VMCS.HOST_RSP field encoding on-demand instead of > delegating to the compiler via an input constraint. In addition to > saving one whole MOV instruction, this allows RDX to be properly > clobbered (in a future patch) instead of being saved/loaded to/from > the stack. > > Despite nested_vmx_check_vmentry_hw() having similar code, leave it > alone, for now. In that case, RDX is unconditionally used and isn't > clobbered, i.e. sending in HOST_RSP as an input is simpler. > > Note that because HOST_RSP is an enum and not a define, it must be > redefined as an immediate instead of using __stringify(HOST_RSP). The > naming "conflict" between host_rsp and HOST_RSP is slightly confusing, > but the former will be removed in a future patch, at which point > HOST_RSP is absolutely what is desired. > > Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Reviewed-by: Jim Mattson <jmattson@google.com>
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index b7cd24dbd01d..f0084726d0c3 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -6389,6 +6389,7 @@ static void __vmx_vcpu_run(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx) "mov %%" _ASM_SP ", (%%" _ASM_SI ") \n\t" "jmp 1f \n\t" "2: \n\t" + "mov $%c[HOST_RSP], %%" _ASM_DX " \n\t" __ex("vmwrite %%" _ASM_SP ", %%" _ASM_DX) "\n\t" "1: \n\t" "add $%c[wordsize], %%" _ASM_SP "\n\t" /* un-adjust RSP */ @@ -6473,10 +6474,11 @@ static void __vmx_vcpu_run(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx) "xor %%edi, %%edi \n\t" "pop %%" _ASM_BP "; pop %%" _ASM_DX " \n\t" : ASM_CALL_CONSTRAINT, "=S"((int){0}) - : "c"(vmx), "d"((unsigned long)HOST_RSP), "S"(evmcs_rsp), + : "c"(vmx), "S"(evmcs_rsp), [launched]"i"(offsetof(struct vcpu_vmx, __launched)), [fail]"i"(offsetof(struct vcpu_vmx, fail)), [host_rsp]"i"(offsetof(struct vcpu_vmx, host_rsp)), + [HOST_RSP]"i"(HOST_RSP), [rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])), [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])), [rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])),
Load RDX with the VMCS.HOST_RSP field encoding on-demand instead of delegating to the compiler via an input constraint. In addition to saving one whole MOV instruction, this allows RDX to be properly clobbered (in a future patch) instead of being saved/loaded to/from the stack. Despite nested_vmx_check_vmentry_hw() having similar code, leave it alone, for now. In that case, RDX is unconditionally used and isn't clobbered, i.e. sending in HOST_RSP as an input is simpler. Note that because HOST_RSP is an enum and not a define, it must be redefined as an immediate instead of using __stringify(HOST_RSP). The naming "conflict" between host_rsp and HOST_RSP is slightly confusing, but the former will be removed in a future patch, at which point HOST_RSP is absolutely what is desired. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> --- arch/x86/kvm/vmx/vmx.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)