Re: [PATCH 1/2] x86/tdx: Add prctl to allow userlevel TDX hypercalls

[Kirill A . Shutemov]

On Wed, Jul 03, 2024 at 11:36:00PM +0000, Tim Merrifield wrote:
> Add a new prctl option to enable/disable user-level hypercalls when
> running in a confidential VM. Add support for checking this flag on
> VMCALL #VE for TDX and transfer control to a hypervisor
> vendor-specific handler.
> 
> Signed-off-by: Tim Merrifield <tim.merrifield@xxxxxxxxxxxx>
> ---
>  arch/x86/coco/tdx/tdx.c            | 18 ++++++++++++++++++
>  arch/x86/include/asm/thread_info.h |  2 ++
>  arch/x86/include/asm/x86_init.h    |  1 +
>  arch/x86/include/uapi/asm/prctl.h  |  3 +++
>  arch/x86/kernel/process.c          | 20 ++++++++++++++++++++
>  5 files changed, 44 insertions(+)
> 
> diff --git a/arch/x86/coco/tdx/tdx.c b/arch/x86/coco/tdx/tdx.c
> index ef8ec2425998..23111e4c1f91 100644
> --- a/arch/x86/coco/tdx/tdx.c
> +++ b/arch/x86/coco/tdx/tdx.c
> @@ -239,6 +239,7 @@ static int ve_instr_len(struct ve_info *ve)
>  	case EXIT_REASON_MSR_WRITE:
>  	case EXIT_REASON_CPUID:
>  	case EXIT_REASON_IO_INSTRUCTION:
> +	case EXIT_REASON_VMCALL:
>  		/* It is safe to use ve->instr_len for #VE due instructions */
>  		return ve->instr_len;
>  	case EXIT_REASON_EPT_VIOLATION:
> @@ -635,6 +636,21 @@ void tdx_get_ve_info(struct ve_info *ve)
>  	ve->instr_info  = upper_32_bits(args.r10);
>  }
>  
> +/*
> + * Handle user-initiated, hypervisor-specific VMCALLs.
> + */
> +static int handle_user_vmcall(struct pt_regs *regs, struct ve_info *ve)
> +{
> +	if (x86_platform.hyper.tdx_hcall &&
> +	    test_thread_flag(TIF_COCO_USER_HCALL)) {
> +		if (!x86_platform.hyper.tdx_hcall(regs))
> +			return -EIO;
> +		return ve_instr_len(ve);
> +	} else {
> +		return -EOPNOTSUPP;
> +	}

Maybe something like this would be more readable:

	if (!x86_platform.hyper.tdx_hcall)
		return -EOPNOTSUPP;

	if (!test_thread_flag(TIF_COCO_USER_HCALL))
		return -EOPNOTSUPP;

	if (!x86_platform.hyper.tdx_hcall(regs))
		return -EIO;

	return ve_instr_len(ve);

BTW, do we want tdx_hcall() to return errno instead of bool?

> +}
> +
>  /*
>   * Handle the user initiated #VE.
>   *
> @@ -646,6 +662,8 @@ static int virt_exception_user(struct pt_regs *regs, struct ve_info *ve)
>  	switch (ve->exit_reason) {
>  	case EXIT_REASON_CPUID:
>  		return handle_cpuid(regs, ve);
> +	case EXIT_REASON_VMCALL:
> +		return handle_user_vmcall(regs, ve);
>  	default:
>  		pr_warn("Unexpected #VE: %lld\n", ve->exit_reason);
>  		return -EIO;
> diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
> index 12da7dfd5ef1..9f69a26a5e68 100644
> --- a/arch/x86/include/asm/thread_info.h
> +++ b/arch/x86/include/asm/thread_info.h
> @@ -106,6 +106,7 @@ struct thread_info {
>  #define TIF_BLOCKSTEP		25	/* set when we want DEBUGCTLMSR_BTF */
>  #define TIF_LAZY_MMU_UPDATES	27	/* task is updating the mmu lazily */
>  #define TIF_ADDR32		29	/* 32-bit address space on 64 bits */
> +#define TIF_COCO_USER_HCALL     30      /* Userland hypercalls allowed in CoCo */

Tabs instead of spaces for alignment, please.

>  #define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
>  #define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
> @@ -128,6 +129,7 @@ struct thread_info {
>  #define _TIF_BLOCKSTEP		(1 << TIF_BLOCKSTEP)
>  #define _TIF_LAZY_MMU_UPDATES	(1 << TIF_LAZY_MMU_UPDATES)
>  #define _TIF_ADDR32		(1 << TIF_ADDR32)
> +#define _TIF_COCO_USER_HCALL    (1 << TIF_COCO_USER_HCALL)

Ditto.

>  
>  /* flags to check in __switch_to() */
>  #define _TIF_WORK_CTXSW_BASE					\
> diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
> index 213cf5379a5a..52975bedd33e 100644
> --- a/arch/x86/include/asm/x86_init.h
> +++ b/arch/x86/include/asm/x86_init.h
> @@ -282,6 +282,7 @@ struct x86_hyper_runtime {
>  	void (*sev_es_hcall_prepare)(struct ghcb *ghcb, struct pt_regs *regs);
>  	bool (*sev_es_hcall_finish)(struct ghcb *ghcb, struct pt_regs *regs);
>  	bool (*is_private_mmio)(u64 addr);
> +	bool (*tdx_hcall)(struct pt_regs *regs);
>  };
>  
>  /**
> diff --git a/arch/x86/include/uapi/asm/prctl.h b/arch/x86/include/uapi/asm/prctl.h
> index 384e2cc6ac19..7fa289a1815b 100644
> --- a/arch/x86/include/uapi/asm/prctl.h
> +++ b/arch/x86/include/uapi/asm/prctl.h
> @@ -16,6 +16,9 @@
>  #define ARCH_GET_XCOMP_GUEST_PERM	0x1024
>  #define ARCH_REQ_XCOMP_GUEST_PERM	0x1025
>  
> +#define ARCH_GET_COCO_USER_HCALL        0x1030
> +#define ARCH_SET_COCO_USER_HCALL        0x1031
> +

Ditto.

>  #define ARCH_XCOMP_TILECFG		17
>  #define ARCH_XCOMP_TILEDATA		18
>  
> diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
> index 1b3d417cd6c4..16f8ab6cde2e 100644
> --- a/arch/x86/kernel/process.c
> +++ b/arch/x86/kernel/process.c
> @@ -1039,6 +1039,21 @@ unsigned long __get_wchan(struct task_struct *p)
>  	return addr;
>  }
>  
> +static int get_coco_user_hcall_mode(void)
> +{
> +	return !test_thread_flag(TIF_COCO_USER_HCALL);
> +}
> +
> +static int set_coco_user_hcall_mode(unsigned long enabled)
> +{
> +	if (enabled)
> +		set_thread_flag(TIF_COCO_USER_HCALL);
> +	else
> +		clear_thread_flag(TIF_COCO_USER_HCALL);
> +
> +	return 0;
> +}
> +

Hm. Per-thread flag is odd. I think it should be per-process.

>  long do_arch_prctl_common(int option, unsigned long arg2)
>  {
>  	switch (option) {
> @@ -1052,6 +1067,11 @@ long do_arch_prctl_common(int option, unsigned long arg2)
>  	case ARCH_GET_XCOMP_GUEST_PERM:
>  	case ARCH_REQ_XCOMP_GUEST_PERM:
>  		return fpu_xstate_prctl(option, arg2);
> +	case ARCH_GET_COCO_USER_HCALL:
> +		return get_coco_user_hcall_mode();
> +	case ARCH_SET_COCO_USER_HCALL:
> +		return set_coco_user_hcall_mode(arg2);
> +
>  	}
>  
>  	return -EINVAL;
> -- 
> 2.40.1
> 

-- 
  Kiryl Shutsemau / Kirill A. Shutemov