On 02.12.2024 10:49, Teddy Astie wrote:
> In many places of x86 HVM code, constants integer are used to indicate in
> what mode is
> running the CPU (real, vm86, 16-bits, 32-bits, 64-bits). However, these
> constants are
> are written directly as integer which hides the actual meaning of these modes.
>
> This patch introduces X86_MODE_* macros and replace those occurences with it.
>
> Signed-off-by Teddy Astie <teddy.as...@vates.tech>
Reviewed-by: Jan Beulich <jbeul...@suse.com>
with further style adjustments (see below) and ideally with ...
> ---
> v2:
> Formatting changes (alignment, ...)
> Renamed v86 to vm86. (Jan)
... this extended to ...
> @@ -2952,11 +2954,11 @@ static const char *guest_x86_mode_to_str(int mode)
> {
> switch ( mode )
> {
> - case 0: return "Real";
> - case 1: return "v86";
> - case 2: return "16bit";
> - case 4: return "32bit";
> - case 8: return "64bit";
> + case X86_MODE_REAL: return "Real";
> + case X86_MODE_VM86: return "v86";
... the string literal here.
> @@ -112,23 +113,23 @@ int hvm_hypercall(struct cpu_user_regs *regs)
>
> switch ( mode )
> {
> - case 8:
> + case X86_MODE_64BIT:
> eax = regs->rax;
> fallthrough;
> - case 4:
> - case 2:
> + case X86_MODE_32BIT:
> + case X86_MODE_16BIT:
> if ( currd->arch.monitor.guest_request_userspace_enabled &&
> eax == __HYPERVISOR_hvm_op &&
> - (mode == 8 ? regs->rdi : regs->ebx) ==
> HVMOP_guest_request_vm_event )
> + (mode == X86_MODE_64BIT ? regs->rdi : regs->ebx) ==
> HVMOP_guest_request_vm_event )
This line is too long now.
> @@ -2073,7 +2073,8 @@ int nvmx_handle_vmx_insn(struct cpu_user_regs *regs,
> unsigned int exit_reason)
>
> if ( !(curr->arch.hvm.guest_cr[4] & X86_CR4_VMXE) ||
> !nestedhvm_enabled(curr->domain) ||
> - (vmx_guest_x86_mode(curr) < (hvm_long_mode_active(curr) ? 8 : 2)) ||
> + (vmx_guest_x86_mode(curr) < (hvm_long_mode_active(curr) ?
> X86_MODE_64BIT
> + :
> X86_MODE_16BIT)) ||
As are these two.
Likely easy enough to adjust while committing.
Jan
