Question

I’m reading KVM source code and confronted with x86_64 inline assembly. In the following code, what’s use of “%c”? It it new feature in x86_64 inline assembly? Any reference for new features in x86_64 inline assembly in gcc?

Many thanks

    /* Check if vmlaunch of vmresume is needed */
    "cmp $0, %1 \n\t"
    /* Load guest registers.  Don't clobber flags. */
#ifdef CONFIG_X86_64
    "mov %c[cr2](%3), %%rax \n\t"
    "mov %%rax, %%cr2 \n\t"
    "mov %c[rax](%3), %%rax \n\t"
    "mov %c[rbx](%3), %%rbx \n\t"
    "mov %c[rdx](%3), %%rdx \n\t"

Answer 1

You can see how this works at the end of the asm statement:

      : : "c"(vmx), "d"((unsigned long)HOST_RSP),
    [launched]"i"(offsetof(struct vcpu_vmx, launched)),
    [fail]"i"(offsetof(struct vcpu_vmx, fail)),
    [rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])),
    [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])),

The %3 (it’s %0 in the my source tree) is a reference to the local variable vmx, and %c[rax], %c[cr2] etc are the integer constant offsets of the corresponding values within the struct vcpu_vmx that vmx points to (%c meaning “constant”).

So the line:

mov %c[cr2](%0), %%rax

is moving the contents of vmx->vcpu.arch.cr2 into %rax.