Commit eac90a5b authored by Chao Gao's avatar Chao Gao Committed by Sean Christopherson
Browse files

x86/virt/tdx: KVM: Consolidate TDX CPU hotplug handling



The core kernel registers a CPU hotplug callback to do VMX and TDX init
and deinit while KVM registers a separate CPU offline callback to block
offlining the last online CPU in a socket.

Splitting TDX-related CPU hotplug handling across two components is odd
and adds unnecessary complexity.

Consolidate TDX-related CPU hotplug handling by integrating KVM's
tdx_offline_cpu() to the one in the core kernel.

Also move nr_configured_hkid to the core kernel because tdx_offline_cpu()
references it. Since HKID allocation and free are handled in the core
kernel, it's more natural to track used HKIDs there.

Reviewed-by: default avatarDan Williams <dan.j.williams@intel.com>
Signed-off-by: default avatarChao Gao <chao.gao@intel.com>
Tested-by: default avatarChao Gao <chao.gao@intel.com>
Tested-by: default avatarSagi Shahar <sagis@google.com>
Link: https://patch.msgid.link/20260214012702.2368778-14-seanjc@google.com


Signed-off-by: default avatarSean Christopherson <seanjc@google.com>
parent 9900400e
Loading
Loading
Loading
Loading
+1 −66
Original line number Diff line number Diff line
@@ -59,8 +59,6 @@ module_param_named(tdx, enable_tdx, bool, 0444);
#define TDX_SHARED_BIT_PWL_5 gpa_to_gfn(BIT_ULL(51))
#define TDX_SHARED_BIT_PWL_4 gpa_to_gfn(BIT_ULL(47))

static enum cpuhp_state tdx_cpuhp_state __ro_after_init;

static const struct tdx_sys_info *tdx_sysinfo;

void tdh_vp_rd_failed(struct vcpu_tdx *tdx, char *uclass, u32 field, u64 err)
@@ -219,8 +217,6 @@ static int init_kvm_tdx_caps(const struct tdx_sys_info_td_conf *td_conf,
 */
static DEFINE_MUTEX(tdx_lock);

static atomic_t nr_configured_hkid;

static bool tdx_operand_busy(u64 err)
{
	return (err & TDX_SEAMCALL_STATUS_MASK) == TDX_OPERAND_BUSY;
@@ -268,7 +264,6 @@ static inline void tdx_hkid_free(struct kvm_tdx *kvm_tdx)
{
	tdx_guest_keyid_free(kvm_tdx->hkid);
	kvm_tdx->hkid = -1;
	atomic_dec(&nr_configured_hkid);
	misc_cg_uncharge(MISC_CG_RES_TDX, kvm_tdx->misc_cg, 1);
	put_misc_cg(kvm_tdx->misc_cg);
	kvm_tdx->misc_cg = NULL;
@@ -2398,8 +2393,6 @@ static int __tdx_td_init(struct kvm *kvm, struct td_params *td_params,

	ret = -ENOMEM;

	atomic_inc(&nr_configured_hkid);

	tdr_page = alloc_page(GFP_KERNEL);
	if (!tdr_page)
		goto free_hkid;
@@ -3291,51 +3284,10 @@ int tdx_gmem_max_mapping_level(struct kvm *kvm, kvm_pfn_t pfn, bool is_private)
	return PG_LEVEL_4K;
}

static int tdx_online_cpu(unsigned int cpu)
{
	return 0;
}

static int tdx_offline_cpu(unsigned int cpu)
{
	int i;

	/* No TD is running.  Allow any cpu to be offline. */
	if (!atomic_read(&nr_configured_hkid))
		return 0;

	/*
	 * In order to reclaim TDX HKID, (i.e. when deleting guest TD), need to
	 * call TDH.PHYMEM.PAGE.WBINVD on all packages to program all memory
	 * controller with pconfig.  If we have active TDX HKID, refuse to
	 * offline the last online cpu.
	 */
	for_each_online_cpu(i) {
		/*
		 * Found another online cpu on the same package.
		 * Allow to offline.
		 */
		if (i != cpu && topology_physical_package_id(i) ==
				topology_physical_package_id(cpu))
			return 0;
	}

	/*
	 * This is the last cpu of this package.  Don't offline it.
	 *
	 * Because it's hard for human operator to understand the
	 * reason, warn it.
	 */
#define MSG_ALLPKG_ONLINE \
	"TDX requires all packages to have an online CPU. Delete all TDs in order to offline all CPUs of a package.\n"
	pr_warn_ratelimited(MSG_ALLPKG_ONLINE);
	return -EBUSY;
}

static int __init __tdx_bringup(void)
{
	const struct tdx_sys_info_td_conf *td_conf;
	int r, i;
	int i;

	for (i = 0; i < ARRAY_SIZE(tdx_uret_msrs); i++) {
		/*
@@ -3403,23 +3355,7 @@ static int __init __tdx_bringup(void)
	if (misc_cg_set_capacity(MISC_CG_RES_TDX, tdx_get_nr_guest_keyids()))
		return -EINVAL;

	/*
	 * TDX-specific cpuhp callback to disallow offlining the last CPU in a
	 * packing while KVM is running one or more TDs.  Reclaiming HKIDs
	 * requires doing PAGE.WBINVD on every package, i.e. offlining all CPUs
	 * of a package would prevent reclaiming the HKID.
	 */
	r = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "kvm/cpu/tdx:online",
			      tdx_online_cpu, tdx_offline_cpu);
	if (r < 0)
		goto err_cpuhup;

	tdx_cpuhp_state = r;
	return 0;

err_cpuhup:
	misc_cg_set_capacity(MISC_CG_RES_TDX, 0);
	return r;
}

int __init tdx_bringup(void)
@@ -3487,7 +3423,6 @@ void tdx_cleanup(void)
		return;

	misc_cg_set_capacity(MISC_CG_RES_TDX, 0);
	cpuhp_remove_state(tdx_cpuhp_state);
}

void __init tdx_hardware_setup(void)
+46 −3
Original line number Diff line number Diff line
@@ -59,6 +59,8 @@ static LIST_HEAD(tdx_memlist);
static struct tdx_sys_info tdx_sysinfo __ro_after_init;
static bool tdx_module_initialized __ro_after_init;

static atomic_t nr_configured_hkid;

typedef void (*sc_err_func_t)(u64 fn, u64 err, struct tdx_module_args *args);

static inline void seamcall_err(u64 fn, u64 err, struct tdx_module_args *args)
@@ -186,6 +188,40 @@ static int tdx_online_cpu(unsigned int cpu)

static int tdx_offline_cpu(unsigned int cpu)
{
	int i;

	/* No TD is running.  Allow any cpu to be offline. */
	if (!atomic_read(&nr_configured_hkid))
		goto done;

	/*
	 * In order to reclaim TDX HKID, (i.e. when deleting guest TD), need to
	 * call TDH.PHYMEM.PAGE.WBINVD on all packages to program all memory
	 * controller with pconfig.  If we have active TDX HKID, refuse to
	 * offline the last online cpu.
	 */
	for_each_online_cpu(i) {
		/*
		 * Found another online cpu on the same package.
		 * Allow to offline.
		 */
		if (i != cpu && topology_physical_package_id(i) ==
				topology_physical_package_id(cpu))
			goto done;
	}

	/*
	 * This is the last cpu of this package.  Don't offline it.
	 *
	 * Because it's hard for human operator to understand the
	 * reason, warn it.
	 */
#define MSG_ALLPKG_ONLINE \
	"TDX requires all packages to have an online CPU. Delete all TDs in order to offline all CPUs of a package.\n"
	pr_warn_ratelimited(MSG_ALLPKG_ONLINE);
	return -EBUSY;

done:
	x86_virt_put_ref(X86_FEATURE_VMX);
	return 0;
}
@@ -1506,15 +1542,22 @@ EXPORT_SYMBOL_FOR_KVM(tdx_get_nr_guest_keyids);

int tdx_guest_keyid_alloc(void)
{
	return ida_alloc_range(&tdx_guest_keyid_pool, tdx_guest_keyid_start,
	int ret;

	ret = ida_alloc_range(&tdx_guest_keyid_pool, tdx_guest_keyid_start,
			      tdx_guest_keyid_start + tdx_nr_guest_keyids - 1,
			      GFP_KERNEL);
	if (ret >= 0)
		atomic_inc(&nr_configured_hkid);

	return ret;
}
EXPORT_SYMBOL_FOR_KVM(tdx_guest_keyid_alloc);

void tdx_guest_keyid_free(unsigned int keyid)
{
	ida_free(&tdx_guest_keyid_pool, keyid);
	atomic_dec(&nr_configured_hkid);
}
EXPORT_SYMBOL_FOR_KVM(tdx_guest_keyid_free);