lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Tue, 22 Feb 2022 22:35:28 -0600
From:   Brijesh Singh <brijesh.singh@....com>
To:     <x86@...nel.org>, <linux-kernel@...r.kernel.org>,
        <kvm@...r.kernel.org>, <linux-coco@...ts.linux.dev>
CC:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Joerg Roedel <jroedel@...e.de>,
        Tom Lendacky <thomas.lendacky@....com>,
        "H. Peter Anvin" <hpa@...or.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        "Sean Christopherson" <seanjc@...gle.com>,
        Andy Lutomirski <luto@...nel.org>,
        "Dave Hansen" <dave.hansen@...ux.intel.com>,
        Peter Gonda <pgonda@...gle.com>,
        "Peter Zijlstra" <peterz@...radead.org>,
        David Rientjes <rientjes@...gle.com>,
        Borislav Petkov <bp@...en8.de>,
        Michael Roth <michael.roth@....com>,
        "Kirill A . Shutemov" <kirill.shutemov@...ux.intel.com>,
        Andi Kleen <ak@...ux.intel.com>,
        Brijesh Singh <brijesh.singh@....com>
Subject: [PATCH] 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
callback under x86_platform.cc. TDX will provide own version of the
callbacks.

Signed-off-by: Brijesh Singh <brijesh.singh@....com>
---

Depends on Krill's CC cleanup
https://lore.kernel.org/all/20220222185740.26228-1-kirill.shutemov@linux.intel.com/

 arch/x86/include/asm/set_memory.h |  1 -
 arch/x86/include/asm/x86_init.h   | 21 +++++++++
 arch/x86/mm/mem_encrypt_amd.c     | 75 ++++++++++++++++++++++---------
 arch/x86/mm/pat/set_memory.c      | 20 +++++----
 4 files changed, 85 insertions(+), 32 deletions(-)

diff --git a/arch/x86/include/asm/set_memory.h b/arch/x86/include/asm/set_memory.h
index ff0f2d90338a..ce8dd215f5b3 100644
--- a/arch/x86/include/asm/set_memory.h
+++ b/arch/x86/include/asm/set_memory.h
@@ -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;
 
diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
index 22b7412c08f6..dce92e2cb9e1 100644
--- a/arch/x86/include/asm/x86_init.h
+++ b/arch/x86/include/asm/x86_init.h
@@ -141,6 +141,26 @@ struct x86_init_acpi {
 	void (*reduced_hw_early_init)(void);
 };
 
+/**
+ * struct x86_cc_runtime - 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	Flush the TLB before changing the encryption status.
+ *
+ * @enc_cache_flush_required	Flush the caches before changing the encryption status.
+ */
+struct x86_cc_runtime {
+	void (*enc_status_change_prepare)(unsigned long vaddr, int npages, bool enc);
+	void (*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 +307,7 @@ struct x86_platform_ops {
 	struct x86_legacy_features legacy;
 	void (*set_legacy_features)(void);
 	struct x86_hyper_runtime hyper;
+	const struct x86_cc_runtime *cc;
 };
 
 struct x86_apic_ops {
diff --git a/arch/x86/mm/mem_encrypt_amd.c b/arch/x86/mm/mem_encrypt_amd.c
index 2b2d018ea345..22b86af5edf1 100644
--- a/arch/x86/mm/mem_encrypt_amd.c
+++ b/arch/x86/mm/mem_encrypt_amd.c
@@ -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 !this_cpu_has(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 +278,18 @@ 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)
+{
+}
+
+static void amd_enc_status_change_finish(unsigned long vaddr, int npages, bool enc)
+{
+	if (cc_platform_has(CC_ATTR_HOST_MEM_ENCRYPT))
+		return;
+
+	enc_dec_hypercall(vaddr, npages, enc);
+}
+
 static void __init __set_clr_pte_enc(pte_t *kpte, int level, bool enc)
 {
 	pgprot_t old_prot, new_prot;
@@ -392,7 +395,7 @@ 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 +413,35 @@ 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);
+}
+
+static const struct x86_cc_runtime amd_cc_runtime = {
+	.enc_status_change_prepare = amd_enc_status_change_prepare,
+	.enc_status_change_finish = amd_enc_status_change_finish,
+	.enc_tlb_flush_required = amd_enc_tlb_flush_required,
+	.enc_cache_flush_required = amd_enc_cache_flush_required,
+};
+
+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.cc = &amd_cc_runtime;
 }
 
 void __init mem_encrypt_free_decrypted_mem(void)
diff --git a/arch/x86/mm/pat/set_memory.c b/arch/x86/mm/pat/set_memory.c
index af77dbfd143c..4de2a7509039 100644
--- a/arch/x86/mm/pat/set_memory.c
+++ b/arch/x86/mm/pat/set_memory.c
@@ -1997,6 +1997,8 @@ static int __set_memory_enc_pgtable(unsigned long addr, int numpages, bool enc)
 	if (WARN_ONCE(addr & ~PAGE_MASK, "misaligned address: %#lx\n", addr))
 		addr &= PAGE_MASK;
 
+	BUG_ON(!x86_platform.cc);
+
 	memset(&cpa, 0, sizeof(cpa));
 	cpa.vaddr = &addr;
 	cpa.numpages = numpages;
@@ -2008,10 +2010,12 @@ 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.cc->enc_tlb_flush_required(enc))
+		cpa_flush(&cpa, x86_platform.cc->enc_cache_flush_required());
+
+	/* Notify hypervisor that we are about to set/clr encryption attribute. */
+	x86_platform.cc->enc_status_change_prepare(addr, numpages, enc);
 
 	ret = __change_page_attr_set_clr(&cpa, 1);
 
@@ -2024,11 +2028,9 @@ 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)
+		x86_platform.cc->enc_status_change_finish(addr, numpages, enc);
 
 	return ret;
 }
-- 
2.25.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ