Re: [PATCH v2 2/5] x86/virt/tdx: Configure add-on features on TDX module init

[Edgecombe, Rick P]

On Tue, 2026-09-15 at 18:26 +0800, Xu Yilun wrote:
> @@ -1012,12 +1021,22 @@ struct tdmr_info_pa_array {
>  static __init int tdx_sys_config(struct tdmr_info_pa_array *tdmr_pa_array,
>  				 unsigned int nr_tdmr_pa, u64 global_keyid)
>  {
> +	u64 addon_features0 = get_tdx_addon_features0();
>  	struct tdx_module_args args = {
>  		.rcx = __pa(tdmr_pa_array),
>  		.rdx = nr_tdmr_pa,
>  		.r8 = global_keyid,
>  	};

It seems inconsistent that global_keyid is passed in even though it is basically
global state, while addon_features0 is retrieved from the global state. Passing
the result of get_tdx_addon_features0() in from the caller doesn't make a ton of
sense. And we already reference tdx_global_keyid globally from another seamcall
wrapper. So I'd think dropping the arg when you create the wrapper would be best
in the end.

The smallest change to stay consistent would be go add an addon_features0 arg to
tdx_sys_config(), but it's hard to justify as good code. I'd maybe go with the
global_keyid change and it probably is a separate patch, but we are brushing up
against mixing cleanup and feature enabling... thoughts?


>  
> +	/*
> +	 * Use SEAMCALL version 1 that supports add-on features if any are
> +	 * requested. Otherwise use version 0 for backward compatibility.
> +	 */
> +	if (addon_features0) {
> +		args.r9 = addon_features0;
> +		args.version = 1;
> +	}
> +
>  	return seamcall_prerr(TDH_SYS_CONFIG, &args);
>  }