Commit 1e8c5971 authored by Brijesh Singh's avatar Brijesh Singh Committed by Borislav Petkov
Browse files

x86/mm/cpa: Generalize __set_memory_enc_pgtable()



The kernel provides infrastructure to set or clear the encryption mask
from the pages for AMD SEV, but TDX requires few tweaks.

- TDX and SEV have different requirements to the cache and TLB
  flushing.

- TDX has own routine to notify VMM about page encryption status change.

Modify __set_memory_enc_pgtable() and make it flexible enough to cover
both AMD SEV and Intel TDX. The AMD-specific behavior is isolated in the
callbacks under x86_platform.guest. TDX will provide own version of said
callbacks.

  [ bp: Beat into submission. ]

Signed-off-by: default avatarBrijesh Singh <brijesh.singh@amd.com>
Signed-off-by: default avatarBorislav Petkov <bp@suse.de>
Acked-by: default avatarKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Link: https://lore.kernel.org/r/20220223043528.2093214-1-brijesh.singh@amd.com
parent b577f542
Loading
Loading
Loading
Loading
+0 −1
Original line number Diff line number Diff line
@@ -84,7 +84,6 @@ int set_pages_rw(struct page *page, int numpages);
int set_direct_map_invalid_noflush(struct page *page);
int set_direct_map_default_noflush(struct page *page);
bool kernel_page_present(struct page *page);
void notify_range_enc_status_changed(unsigned long vaddr, int npages, bool enc);

extern int kernel_set_to_readonly;

+16 −0
Original line number Diff line number Diff line
@@ -141,6 +141,21 @@ struct x86_init_acpi {
	void (*reduced_hw_early_init)(void);
};

/**
 * struct x86_guest - Functions used by misc guest incarnations like SEV, TDX, etc.
 *
 * @enc_status_change_prepare	Notify HV before the encryption status of a range is changed
 * @enc_status_change_finish	Notify HV after the encryption status of a range is changed
 * @enc_tlb_flush_required	Returns true if a TLB flush is needed before changing page encryption status
 * @enc_cache_flush_required	Returns true if a cache flush is needed before changing page encryption status
 */
struct x86_guest {
	void (*enc_status_change_prepare)(unsigned long vaddr, int npages, bool enc);
	bool (*enc_status_change_finish)(unsigned long vaddr, int npages, bool enc);
	bool (*enc_tlb_flush_required)(bool enc);
	bool (*enc_cache_flush_required)(void);
};

/**
 * struct x86_init_ops - functions for platform specific setup
 *
@@ -287,6 +302,7 @@ struct x86_platform_ops {
	struct x86_legacy_features legacy;
	void (*set_legacy_features)(void);
	struct x86_hyper_runtime hyper;
	struct x86_guest guest;
};

struct x86_apic_ops {
+14 −2
Original line number Diff line number Diff line
@@ -129,6 +129,11 @@ struct x86_cpuinit_ops x86_cpuinit = {

static void default_nmi_init(void) { };

static void enc_status_change_prepare_noop(unsigned long vaddr, int npages, bool enc) { }
static bool enc_status_change_finish_noop(unsigned long vaddr, int npages, bool enc) { return false; }
static bool enc_tlb_flush_required_noop(bool enc) { return false; }
static bool enc_cache_flush_required_noop(void) { return false; }

struct x86_platform_ops x86_platform __ro_after_init = {
	.calibrate_cpu			= native_calibrate_cpu_early,
	.calibrate_tsc			= native_calibrate_tsc,
@@ -141,6 +146,13 @@ struct x86_platform_ops x86_platform __ro_after_init = {
	.save_sched_clock_state		= tsc_save_sched_clock_state,
	.restore_sched_clock_state	= tsc_restore_sched_clock_state,
	.hyper.pin_vcpu			= x86_op_int_noop,

	.guest = {
		.enc_status_change_prepare = enc_status_change_prepare_noop,
		.enc_status_change_finish  = enc_status_change_finish_noop,
		.enc_tlb_flush_required	   = enc_tlb_flush_required_noop,
		.enc_cache_flush_required  = enc_cache_flush_required_noop,
	},
};

EXPORT_SYMBOL_GPL(x86_platform);
+50 −22
Original line number Diff line number Diff line
@@ -177,25 +177,6 @@ void __init sme_map_bootdata(char *real_mode_data)
	__sme_early_map_unmap_mem(__va(cmdline_paddr), COMMAND_LINE_SIZE, true);
}

void __init sme_early_init(void)
{
	unsigned int i;

	if (!sme_me_mask)
		return;

	early_pmd_flags = __sme_set(early_pmd_flags);

	__supported_pte_mask = __sme_set(__supported_pte_mask);

	/* Update the protection map with memory encryption mask */
	for (i = 0; i < ARRAY_SIZE(protection_map); i++)
		protection_map[i] = pgprot_encrypted(protection_map[i]);

	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT))
		swiotlb_force = SWIOTLB_FORCE;
}

void __init sev_setup_arch(void)
{
	phys_addr_t total_mem = memblock_phys_mem_size();
@@ -256,7 +237,17 @@ static unsigned long pg_level_to_pfn(int level, pte_t *kpte, pgprot_t *ret_prot)
	return pfn;
}

void notify_range_enc_status_changed(unsigned long vaddr, int npages, bool enc)
static bool amd_enc_tlb_flush_required(bool enc)
{
	return true;
}

static bool amd_enc_cache_flush_required(void)
{
	return !cpu_feature_enabled(X86_FEATURE_SME_COHERENT);
}

static void enc_dec_hypercall(unsigned long vaddr, int npages, bool enc)
{
#ifdef CONFIG_PARAVIRT
	unsigned long sz = npages << PAGE_SHIFT;
@@ -287,6 +278,19 @@ void notify_range_enc_status_changed(unsigned long vaddr, int npages, bool enc)
#endif
}

static void amd_enc_status_change_prepare(unsigned long vaddr, int npages, bool enc)
{
}

/* Return true unconditionally: return value doesn't matter for the SEV side */
static bool amd_enc_status_change_finish(unsigned long vaddr, int npages, bool enc)
{
	if (!cc_platform_has(CC_ATTR_HOST_MEM_ENCRYPT))
		enc_dec_hypercall(vaddr, npages, enc);

	return true;
}

static void __init __set_clr_pte_enc(pte_t *kpte, int level, bool enc)
{
	pgprot_t old_prot, new_prot;
@@ -392,7 +396,7 @@ static int __init early_set_memory_enc_dec(unsigned long vaddr,

	ret = 0;

	notify_range_enc_status_changed(start, PAGE_ALIGN(size) >> PAGE_SHIFT, enc);
	early_set_mem_enc_dec_hypercall(start, PAGE_ALIGN(size) >> PAGE_SHIFT, enc);
out:
	__flush_tlb_all();
	return ret;
@@ -410,7 +414,31 @@ int __init early_set_memory_encrypted(unsigned long vaddr, unsigned long size)

void __init early_set_mem_enc_dec_hypercall(unsigned long vaddr, int npages, bool enc)
{
	notify_range_enc_status_changed(vaddr, npages, enc);
	enc_dec_hypercall(vaddr, npages, enc);
}

void __init sme_early_init(void)
{
	unsigned int i;

	if (!sme_me_mask)
		return;

	early_pmd_flags = __sme_set(early_pmd_flags);

	__supported_pte_mask = __sme_set(__supported_pte_mask);

	/* Update the protection map with memory encryption mask */
	for (i = 0; i < ARRAY_SIZE(protection_map); i++)
		protection_map[i] = pgprot_encrypted(protection_map[i]);

	if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT))
		swiotlb_force = SWIOTLB_FORCE;

	x86_platform.guest.enc_status_change_prepare = amd_enc_status_change_prepare;
	x86_platform.guest.enc_status_change_finish  = amd_enc_status_change_finish;
	x86_platform.guest.enc_tlb_flush_required    = amd_enc_tlb_flush_required;
	x86_platform.guest.enc_cache_flush_required  = amd_enc_cache_flush_required;
}

void __init mem_encrypt_free_decrypted_mem(void)
+11 −9
Original line number Diff line number Diff line
@@ -2008,10 +2008,12 @@ static int __set_memory_enc_pgtable(unsigned long addr, int numpages, bool enc)
	kmap_flush_unused();
	vm_unmap_aliases();

	/*
	 * Before changing the encryption attribute, we need to flush caches.
	 */
	cpa_flush(&cpa, !this_cpu_has(X86_FEATURE_SME_COHERENT));
	/* Flush the caches as needed before changing the encryption attribute. */
	if (x86_platform.guest.enc_tlb_flush_required(enc))
		cpa_flush(&cpa, x86_platform.guest.enc_cache_flush_required());

	/* Notify hypervisor that we are about to set/clr encryption attribute. */
	x86_platform.guest.enc_status_change_prepare(addr, numpages, enc);

	ret = __change_page_attr_set_clr(&cpa, 1);

@@ -2024,11 +2026,11 @@ static int __set_memory_enc_pgtable(unsigned long addr, int numpages, bool enc)
	 */
	cpa_flush(&cpa, 0);

	/*
	 * Notify hypervisor that a given memory range is mapped encrypted
	 * or decrypted.
	 */
	notify_range_enc_status_changed(addr, numpages, enc);
	/* Notify hypervisor that we have successfully set/clr encryption attribute. */
	if (!ret) {
		if (!x86_platform.guest.enc_status_change_finish(addr, numpages, enc))
			ret = -EIO;
	}

	return ret;
}