KVM/x86: Use assembly instruction mnemonics instead of .byte streams

Recently the minimum required version of binutils was changed to 2.20,
which supports all VMX instruction mnemonics. The patch removes
all .byte #defines and uses real instruction mnemonics instead.

The compiler is now able to pass memory operand to the instruction,
so there is no need for memory clobber anymore. Also, the compiler
adds CC register clobber automatically to all extended asm clauses,
so the patch also removes explicit CC clobber.

The immediate benefit of the patch is removal of many unnecesary
register moves, resulting in 1434 saved bytes in vmx.o:

   text    data     bss     dec     hex filename
 151257   18246    8500  178003   2b753 vmx.o
 152691   18246    8500  179437   2bced vmx-old.o

Some examples of improvement include removal of unneeded moves
of %rsp to %rax in front of invept and invvpid instructions:

    a57e:	b9 01 00 00 00       	mov    $0x1,%ecx
    a583:	48 89 04 24          	mov    %rax,(%rsp)
    a587:	48 89 e0             	mov    %rsp,%rax
    a58a:	48 c7 44 24 08 00 00 	movq   $0x0,0x8(%rsp)
    a591:	00 00
    a593:	66 0f 38 80 08       	invept (%rax),%rcx

to:

    a45c:	48 89 04 24          	mov    %rax,(%rsp)
    a460:	b8 01 00 00 00       	mov    $0x1,%eax
    a465:	48 c7 44 24 08 00 00 	movq   $0x0,0x8(%rsp)
    a46c:	00 00
    a46e:	66 0f 38 80 04 24    	invept (%rsp),%rax

and the ability to use more optimal registers and memory operands
in the instruction:

    8faa:	48 8b 44 24 28       	mov    0x28(%rsp),%rax
    8faf:	4c 89 c2             	mov    %r8,%rdx
    8fb2:	0f 79 d0             	vmwrite %rax,%rdx

to:

    8e7c:	44 0f 79 44 24 28    	vmwrite 0x28(%rsp),%r8

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/arch/x86/include/asm/virtext.h b/arch/x86/include/asm/virtext.h
index 0116b2ee9e64f34ebf612a0380410af39b4dd2e5..449c92da2c91ef24a5cad9ad187f021fd36261e1 100644 (file)
--- a/arch/x86/include/asm/virtext.h
+++ b/arch/x86/include/asm/virtext.h
@@ -40,7 +40,7 @@ static inline int cpu_has_vmx(void)
  */
 static inline void cpu_vmxoff(void)
 {
-       asm volatile (ASM_VMX_VMXOFF : : : "cc");
+       asm volatile ("vmxoff");
        cr4_clear_bits(X86_CR4_VMXE);
 }
 

diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h
index 9527ba5d62daba4fe5ca96d9c1ba2986f9612ccc..ade0f153947db82042e34ecd7124055c474d1ca0 100644 (file)
--- a/arch/x86/include/asm/vmx.h
+++ b/arch/x86/include/asm/vmx.h
@@ -503,19 +503,6 @@ enum vmcs_field {
 
 #define VMX_EPT_IDENTITY_PAGETABLE_ADDR                0xfffbc000ul
 
-
-#define ASM_VMX_VMCLEAR_RAX       ".byte 0x66, 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMLAUNCH          ".byte 0x0f, 0x01, 0xc2"
-#define ASM_VMX_VMRESUME          ".byte 0x0f, 0x01, 0xc3"
-#define ASM_VMX_VMPTRLD_RAX       ".byte 0x0f, 0xc7, 0x30"
-#define ASM_VMX_VMREAD_RDX_RAX    ".byte 0x0f, 0x78, 0xd0"
-#define ASM_VMX_VMWRITE_RAX_RDX   ".byte 0x0f, 0x79, 0xd0"
-#define ASM_VMX_VMWRITE_RSP_RDX   ".byte 0x0f, 0x79, 0xd4"
-#define ASM_VMX_VMXOFF            ".byte 0x0f, 0x01, 0xc4"
-#define ASM_VMX_VMXON_RAX         ".byte 0xf3, 0x0f, 0xc7, 0x30"
-#define ASM_VMX_INVEPT           ".byte 0x66, 0x0f, 0x38, 0x80, 0x08"
-#define ASM_VMX_INVVPID                  ".byte 0x66, 0x0f, 0x38, 0x81, 0x08"
-
 struct vmx_msr_entry {
        u32 index;
        u32 reserved;

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 05a0e8f80f68239fcfee71b48d2eb07b2ccffaf8..c1b9bd15815c9b27db5a6eb352f3a998bd6f202e 100644 (file)
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2079,9 +2079,8 @@ static inline void __invvpid(unsigned long ext, u16 vpid, gva_t gva)
     } operand = { vpid, 0, gva };
     bool error;
 
-    asm volatile (__ex(ASM_VMX_INVVPID) CC_SET(na)
-                 : CC_OUT(na) (error) : "a"(&operand), "c"(ext)
-                 : "memory");
+    asm volatile (__ex("invvpid %2, %1") CC_SET(na)
+                 : CC_OUT(na) (error) : "r"(ext), "m"(operand));
     BUG_ON(error);
 }
 
@@ -2092,9 +2091,8 @@ static inline void __invept(unsigned long ext, u64 eptp, gpa_t gpa)
        } operand = {eptp, gpa};
        bool error;
 
-       asm volatile (__ex(ASM_VMX_INVEPT) CC_SET(na)
-                     : CC_OUT(na) (error) : "a" (&operand), "c" (ext)
-                     : "memory");
+       asm volatile (__ex("invept %2, %1") CC_SET(na)
+                     : CC_OUT(na) (error) : "r"(ext), "m"(operand));
        BUG_ON(error);
 }
 
@@ -2113,9 +2111,8 @@ static void vmcs_clear(struct vmcs *vmcs)
        u64 phys_addr = __pa(vmcs);
        bool error;
 
-       asm volatile (__ex(ASM_VMX_VMCLEAR_RAX) CC_SET(na)
-                     : CC_OUT(na) (error) : "a"(&phys_addr), "m"(phys_addr)
-                     : "memory");
+       asm volatile (__ex("vmclear %1") CC_SET(na)
+                     : CC_OUT(na) (error) : "m"(phys_addr));
        if (unlikely(error))
                printk(KERN_ERR "kvm: vmclear fail: %p/%llx\n",
                       vmcs, phys_addr);
@@ -2138,9 +2135,8 @@ static void vmcs_load(struct vmcs *vmcs)
        if (static_branch_unlikely(&enable_evmcs))
                return evmcs_load(phys_addr);
 
-       asm volatile (__ex(ASM_VMX_VMPTRLD_RAX) CC_SET(na)
-                     : CC_OUT(na) (error) : "a"(&phys_addr), "m"(phys_addr)
-                     : "memory");
+       asm volatile (__ex("vmptrld %1") CC_SET(na)
+                     : CC_OUT(na) (error) : "m"(phys_addr));
        if (unlikely(error))
                printk(KERN_ERR "kvm: vmptrld %p/%llx failed\n",
                       vmcs, phys_addr);
@@ -2316,8 +2312,8 @@ static __always_inline unsigned long __vmcs_readl(unsigned long field)
 {
        unsigned long value;
 
-       asm volatile (__ex_clear(ASM_VMX_VMREAD_RDX_RAX, "%0")
-                     : "=a"(value) : "d"(field) : "cc");
+       asm volatile (__ex_clear("vmread %1, %0", "%0")
+                     : "=r"(value) : "r"(field));
        return value;
 }
 
@@ -2368,8 +2364,8 @@ static __always_inline void __vmcs_writel(unsigned long field, unsigned long val
 {
        bool error;
 
-       asm volatile (__ex(ASM_VMX_VMWRITE_RAX_RDX) CC_SET(na)
-                     : CC_OUT(na) (error) : "a"(value), "d"(field));
+       asm volatile (__ex("vmwrite %2, %1") CC_SET(na)
+                     : CC_OUT(na) (error) : "r"(field), "rm"(value));
        if (unlikely(error))
                vmwrite_error(field, value);
 }
@@ -4396,9 +4392,7 @@ static void kvm_cpu_vmxon(u64 addr)
        cr4_set_bits(X86_CR4_VMXE);
        intel_pt_handle_vmx(1);
 
-       asm volatile (ASM_VMX_VMXON_RAX
-                       : : "a"(&addr), "m"(addr)
-                       : "memory", "cc");
+       asm volatile ("vmxon %0" : : "m"(addr));
 }
 
 static int hardware_enable(void)
@@ -4467,7 +4461,7 @@ static void vmclear_local_loaded_vmcss(void)
  */
 static void kvm_cpu_vmxoff(void)
 {
-       asm volatile (__ex(ASM_VMX_VMXOFF) : : : "cc");
+       asm volatile (__ex("vmxoff"));
 
        intel_pt_handle_vmx(0);
        cr4_clear_bits(X86_CR4_VMXE);
@@ -10712,7 +10706,7 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
                "mov %%" _ASM_SP ", (%%" _ASM_SI ") \n\t"
                "jmp 1f \n\t"
                "2: \n\t"
-               __ex(ASM_VMX_VMWRITE_RSP_RDX) "\n\t"
+               __ex("vmwrite %%" _ASM_SP ", %%" _ASM_DX) "\n\t"
                "1: \n\t"
                /* Reload cr2 if changed */
                "mov %c[cr2](%0), %%" _ASM_AX " \n\t"
@@ -10744,9 +10738,9 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 
                /* Enter guest mode */
                "jne 1f \n\t"
-               __ex(ASM_VMX_VMLAUNCH) "\n\t"
+               __ex("vmlaunch") "\n\t"
                "jmp 2f \n\t"
-               "1: " __ex(ASM_VMX_VMRESUME) "\n\t"
+               "1: " __ex("vmresume") "\n\t"
                "2: "
                /* Save guest registers, load host registers, keep flags */
                "mov %0, %c[wordsize](%%" _ASM_SP ") \n\t"
@@ -12702,15 +12696,15 @@ static int __noclone nested_vmx_check_vmentry_hw(struct kvm_vcpu *vcpu)
 
        asm(
                /* Set HOST_RSP */
-               __ex(ASM_VMX_VMWRITE_RSP_RDX) "\n\t"
+               __ex("vmwrite %%" _ASM_SP ", %%" _ASM_DX) "\n\t"
                "mov %%" _ASM_SP ", %c[host_rsp](%0)\n\t"
 
                /* Check if vmlaunch of vmresume is needed */
                "cmpl $0, %c[launched](%0)\n\t"
                "je 1f\n\t"
-               __ex(ASM_VMX_VMRESUME) "\n\t"
+               __ex("vmresume") "\n\t"
                "jmp 2f\n\t"
-               "1: " __ex(ASM_VMX_VMLAUNCH) "\n\t"
+               "1: " __ex("vmlaunch") "\n\t"
                "jmp 2f\n\t"
                "2: "