Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 updates from Will Deacon:
 "A sizeable pile of arm64 updates for 5.8.

  Summary below, but the big two features are support for Branch Target
  Identification and Clang's Shadow Call stack. The latter is currently
  arm64-only, but the high-level parts are all in core code so it could
  easily be adopted by other architectures pending toolchain support

  Branch Target Identification (BTI):

   - Support for ARMv8.5-BTI in both user- and kernel-space. This allows
     branch targets to limit the types of branch from which they can be
     called and additionally prevents branching to arbitrary code,
     although kernel support requires a very recent toolchain.

   - Function annotation via SYM_FUNC_START() so that assembly functions
     are wrapped with the relevant "landing pad" instructions.

   - BPF and vDSO updates to use the new instructions.

   - Addition of a new HWCAP and exposure of BTI capability to userspace
     via ID register emulation, along with ELF loader support for the
     BTI feature in .note.gnu.property.

   - Non-critical fixes to CFI unwind annotations in the sigreturn
     trampoline.

  Shadow Call Stack (SCS):

   - Support for Clang's Shadow Call Stack feature, which reserves
     platform register x18 to point at a separate stack for each task
     that holds only return addresses. This protects function return
     control flow from buffer overruns on the main stack.

   - Save/restore of x18 across problematic boundaries (user-mode,
     hypervisor, EFI, suspend, etc).

   - Core support for SCS, should other architectures want to use it
     too.

   - SCS overflow checking on context-switch as part of the existing
     stack limit check if CONFIG_SCHED_STACK_END_CHECK=y.

  CPU feature detection:

   - Removed numerous "SANITY CHECK" errors when running on a system
     with mismatched AArch32 support at EL1. This is primarily a concern
     for KVM, which disabled support for 32-bit guests on such a system.

   - Addition of new ID registers and fields as the architecture has
     been extended.

  Perf and PMU drivers:

   - Minor fixes and cleanups to system PMU drivers.

  Hardware errata:

   - Unify KVM workarounds for VHE and nVHE configurations.

   - Sort vendor errata entries in Kconfig.

  Secure Monitor Call Calling Convention (SMCCC):

   - Update to the latest specification from Arm (v1.2).

   - Allow PSCI code to query the SMCCC version.

  Software Delegated Exception Interface (SDEI):

   - Unexport a bunch of unused symbols.

   - Minor fixes to handling of firmware data.

  Pointer authentication:

   - Add support for dumping the kernel PAC mask in vmcoreinfo so that
     the stack can be unwound by tools such as kdump.

   - Simplification of key initialisation during CPU bringup.

  BPF backend:

   - Improve immediate generation for logical and add/sub instructions.

  vDSO:

   - Minor fixes to the linker flags for consistency with other
     architectures and support for LLVM's unwinder.

   - Clean up logic to initialise and map the vDSO into userspace.

  ACPI:

   - Work around for an ambiguity in the IORT specification relating to
     the "num_ids" field.

   - Support _DMA method for all named components rather than only PCIe
     root complexes.

   - Minor other IORT-related fixes.

  Miscellaneous:

   - Initialise debug traps early for KGDB and fix KDB cacheflushing
     deadlock.

   - Minor tweaks to early boot state (documentation update, set
     TEXT_OFFSET to 0x0, increase alignment of PE/COFF sections).

   - Refactoring and cleanup"

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (148 commits)
  KVM: arm64: Move __load_guest_stage2 to kvm_mmu.h
  KVM: arm64: Check advertised Stage-2 page size capability
  arm64/cpufeature: Add get_arm64_ftr_reg_nowarn()
  ACPI/IORT: Remove the unused __get_pci_rid()
  arm64/cpuinfo: Add ID_MMFR4_EL1 into the cpuinfo_arm64 context
  arm64/cpufeature: Add remaining feature bits in ID_AA64PFR1 register
  arm64/cpufeature: Add remaining feature bits in ID_AA64PFR0 register
  arm64/cpufeature: Add remaining feature bits in ID_AA64ISAR0 register
  arm64/cpufeature: Add remaining feature bits in ID_MMFR4 register
  arm64/cpufeature: Add remaining feature bits in ID_PFR0 register
  arm64/cpufeature: Introduce ID_MMFR5 CPU register
  arm64/cpufeature: Introduce ID_DFR1 CPU register
  arm64/cpufeature: Introduce ID_PFR2 CPU register
  arm64/cpufeature: Make doublelock a signed feature in ID_AA64DFR0
  arm64/cpufeature: Drop TraceFilt feature exposure from ID_DFR0 register
  arm64/cpufeature: Add explicit ftr_id_isar0[] for ID_ISAR0 register
  arm64: mm: Add asid_gen_match() helper
  firmware: smccc: Fix missing prototype warning for arm_smccc_version_init
  arm64: vdso: Fix CFI directives in sigreturn trampoline
  arm64: vdso: Don't prefix sigreturn trampoline with a BTI C instruction
  ...

9 files changed, 156 insertions, 13 deletions

diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h
index 59494df0f55b..56d6a5c6e353 100644
--- a/include/linux/arm-smccc.h
+++ b/include/linux/arm-smccc.h
@@ -5,12 +5,15 @@
 #ifndef __LINUX_ARM_SMCCC_H
 #define __LINUX_ARM_SMCCC_H
 
+#include <linux/init.h>
 #include <uapi/linux/const.h>
 
 /*
  * This file provides common defines for ARM SMC Calling Convention as
  * specified in
- * http://infocenter.arm.com/help/topic/com.arm.doc.den0028a/index.html
+ * https://developer.arm.com/docs/den0028/latest
+ *
+ * This code is up-to-date with version DEN 0028 C
  */
 
 #define ARM_SMCCC_STD_CALL	        _AC(0,U)
@@ -56,6 +59,7 @@
 
 #define ARM_SMCCC_VERSION_1_0		0x10000
 #define ARM_SMCCC_VERSION_1_1		0x10001
+#define ARM_SMCCC_VERSION_1_2		0x10002
 
 #define ARM_SMCCC_VERSION_FUNC_ID					\
 	ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL,				\
@@ -98,6 +102,19 @@ enum arm_smccc_conduit {
 enum arm_smccc_conduit arm_smccc_1_1_get_conduit(void);
 
 /**
+ * arm_smccc_get_version()
+ *
+ * Returns the version to be used for SMCCCv1.1 or later.
+ *
+ * When SMCCCv1.1 or above is not present, returns SMCCCv1.0, but this
+ * does not imply the presence of firmware or a valid conduit. Caller
+ * handling SMCCCv1.0 must determine the conduit by other means.
+ */
+u32 arm_smccc_get_version(void);
+
+void __init arm_smccc_version_init(u32 version, enum arm_smccc_conduit conduit);
+
+/**
  * struct arm_smccc_res - Result from SMC/HVC call
  * @a0-a3 result values from registers 0 to 3
  */
@@ -314,10 +331,14 @@ asmlinkage void __arm_smccc_hvc(unsigned long a0, unsigned long a1,
  */
 #define arm_smccc_1_1_hvc(...)	__arm_smccc_1_1(SMCCC_HVC_INST, __VA_ARGS__)
 
-/* Return codes defined in ARM DEN 0070A */
+/*
+ * Return codes defined in ARM DEN 0070A
+ * ARM DEN 0070A is now merged/consolidated into ARM DEN 0028 C
+ */
 #define SMCCC_RET_SUCCESS			0
 #define SMCCC_RET_NOT_SUPPORTED			-1
 #define SMCCC_RET_NOT_REQUIRED			-2
+#define SMCCC_RET_INVALID_PARAMETER		-3
 
 /*
  * Like arm_smccc_1_1* but always returns SMCCC_RET_NOT_SUPPORTED.
diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h
index 333a6695a918..790c0c6b8552 100644
--- a/include/linux/compiler-clang.h
+++ b/include/linux/compiler-clang.h
@@ -42,3 +42,7 @@
  * compilers, like ICC.
  */
 #define barrier() __asm__ __volatile__("" : : : "memory")
+
+#if __has_feature(shadow_call_stack)
+# define __noscs	__attribute__((__no_sanitize__("shadow-call-stack")))
+#endif
diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h
index 5da257cbebf1..6fcf73200b67 100644
--- a/include/linux/compiler_types.h
+++ b/include/linux/compiler_types.h
@@ -197,6 +197,10 @@ struct ftrace_likely_data {
 # define randomized_struct_fields_end
 #endif
 
+#ifndef __noscs
+# define __noscs
+#endif
+
 #ifndef asm_volatile_goto
 #define asm_volatile_goto(x...) asm goto(x)
 #endif
diff --git a/include/linux/elf.h b/include/linux/elf.h
index e3649b3e970e..5d5b0321da0b 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -2,6 +2,7 @@
 #ifndef _LINUX_ELF_H
 #define _LINUX_ELF_H
 
+#include <linux/types.h>
 #include <asm/elf.h>
 #include <uapi/linux/elf.h>
 
@@ -21,6 +22,9 @@
 	SET_PERSONALITY(ex)
 #endif
 
+#define ELF32_GNU_PROPERTY_ALIGN	4
+#define ELF64_GNU_PROPERTY_ALIGN	8
+
 #if ELF_CLASS == ELFCLASS32
 
 extern Elf32_Dyn _DYNAMIC [];
@@ -31,6 +35,7 @@ extern Elf32_Dyn _DYNAMIC [];
 #define elf_addr_t	Elf32_Off
 #define Elf_Half	Elf32_Half
 #define Elf_Word	Elf32_Word
+#define ELF_GNU_PROPERTY_ALIGN	ELF32_GNU_PROPERTY_ALIGN
 
 #else
 
@@ -42,6 +47,7 @@ extern Elf64_Dyn _DYNAMIC [];
 #define elf_addr_t	Elf64_Off
 #define Elf_Half	Elf64_Half
 #define Elf_Word	Elf64_Word
+#define ELF_GNU_PROPERTY_ALIGN	ELF64_GNU_PROPERTY_ALIGN
 
 #endif
 
@@ -56,4 +62,41 @@ static inline int elf_coredump_extra_notes_write(struct coredump_params *cprm) {
 extern int elf_coredump_extra_notes_size(void);
 extern int elf_coredump_extra_notes_write(struct coredump_params *cprm);
 #endif
+
+/*
+ * NT_GNU_PROPERTY_TYPE_0 header:
+ * Keep this internal until/unless there is an agreed UAPI definition.
+ * pr_type values (GNU_PROPERTY_*) are public and defined in the UAPI header.
+ */
+struct gnu_property {
+	u32 pr_type;
+	u32 pr_datasz;
+};
+
+struct arch_elf_state;
+
+#ifndef CONFIG_ARCH_USE_GNU_PROPERTY
+static inline int arch_parse_elf_property(u32 type, const void *data,
+					  size_t datasz, bool compat,
+					  struct arch_elf_state *arch)
+{
+	return 0;
+}
+#else
+extern int arch_parse_elf_property(u32 type, const void *data, size_t datasz,
+				   bool compat, struct arch_elf_state *arch);
+#endif
+
+#ifdef CONFIG_ARCH_HAVE_ELF_PROT
+int arch_elf_adjust_prot(int prot, const struct arch_elf_state *state,
+			 bool has_interp, bool is_interp);
+#else
+static inline int arch_elf_adjust_prot(int prot,
+				       const struct arch_elf_state *state,
+				       bool has_interp, bool is_interp)
+{
+	return prot;
+}
+#endif
+
 #endif /* _LINUX_ELF_H */
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 9280209d1f62..d796ec20d114 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -105,7 +105,7 @@
 
 /* === DEPRECATED annotations === */
 
-#ifndef CONFIG_X86
+#ifndef CONFIG_ARCH_USE_SYM_ANNOTATIONS
 #ifndef GLOBAL
 /* deprecated, use SYM_DATA*, SYM_ENTRY, or similar */
 #define GLOBAL(name) \
@@ -118,10 +118,10 @@
 #define ENTRY(name) \
 	SYM_FUNC_START(name)
 #endif
-#endif /* CONFIG_X86 */
+#endif /* CONFIG_ARCH_USE_SYM_ANNOTATIONS */
 #endif /* LINKER_SCRIPT */
 
-#ifndef CONFIG_X86
+#ifndef CONFIG_ARCH_USE_SYM_ANNOTATIONS
 #ifndef WEAK
 /* deprecated, use SYM_FUNC_START_WEAK* */
 #define WEAK(name)	   \
@@ -143,7 +143,7 @@
 #define ENDPROC(name) \
 	SYM_FUNC_END(name)
 #endif
-#endif /* CONFIG_X86 */
+#endif /* CONFIG_ARCH_USE_SYM_ANNOTATIONS */
 
 /* === generic annotations === */
 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index d5b9b7b0bb9c..3c71d1cbabc3 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -325,6 +325,9 @@ extern unsigned int kobjsize(const void *objp);
 #elif defined(CONFIG_SPARC64)
 # define VM_SPARC_ADI	VM_ARCH_1	/* Uses ADI tag for access control */
 # define VM_ARCH_CLEAR	VM_SPARC_ADI
+#elif defined(CONFIG_ARM64)
+# define VM_ARM64_BTI	VM_ARCH_1	/* BTI guarded page, a.k.a. GP bit */
+# define VM_ARCH_CLEAR	VM_ARM64_BTI
 #elif !defined(CONFIG_MMU)
 # define VM_MAPPED_COPY	VM_ARCH_1	/* T if mapped copy of data (nommu mmap) */
 #endif
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 1b9de7d220fb..acffc3bc6178 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -156,6 +156,9 @@ enum zone_stat_item {
 	NR_MLOCK,		/* mlock()ed pages found and moved off LRU */
 	NR_PAGETABLE,		/* used for pagetables */
 	NR_KERNEL_STACK_KB,	/* measured in KiB */
+#if IS_ENABLED(CONFIG_SHADOW_CALL_STACK)
+	NR_KERNEL_SCS_KB,	/* measured in KiB */
+#endif
 	/* Second 128 byte cacheline */
 	NR_BOUNCE,
 #if IS_ENABLED(CONFIG_ZSMALLOC)
diff --git a/include/linux/psci.h b/include/linux/psci.h
index a67712b73b6c..14ad9b9ebcd6 100644
--- a/include/linux/psci.h
+++ b/include/linux/psci.h
@@ -21,11 +21,6 @@ bool psci_power_state_is_valid(u32 state);
 int psci_set_osi_mode(void);
 bool psci_has_osi_support(void);
 
-enum smccc_version {
-	SMCCC_VERSION_1_0,
-	SMCCC_VERSION_1_1,
-};
-
 struct psci_operations {
 	u32 (*get_version)(void);
 	int (*cpu_suspend)(u32 state, unsigned long entry_point);
@@ -35,8 +30,6 @@ struct psci_operations {
 	int (*affinity_info)(unsigned long target_affinity,
 			unsigned long lowest_affinity_level);
 	int (*migrate_info_type)(void);
-	enum arm_smccc_conduit conduit;
-	enum smccc_version smccc_version;
 };
 
 extern struct psci_operations psci_ops;
diff --git a/include/linux/scs.h b/include/linux/scs.h
new file mode 100644
index 000000000000..6dec390cf154
--- /dev/null
+++ b/include/linux/scs.h
@@ -0,0 +1,72 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Shadow Call Stack support.
+ *
+ * Copyright (C) 2019 Google LLC
+ */
+
+#ifndef _LINUX_SCS_H
+#define _LINUX_SCS_H
+
+#include <linux/gfp.h>
+#include <linux/poison.h>
+#include <linux/sched.h>
+#include <linux/sizes.h>
+
+#ifdef CONFIG_SHADOW_CALL_STACK
+
+/*
+ * In testing, 1 KiB shadow stack size (i.e. 128 stack frames on a 64-bit
+ * architecture) provided ~40% safety margin on stack usage while keeping
+ * memory allocation overhead reasonable.
+ */
+#define SCS_SIZE		SZ_1K
+#define GFP_SCS			(GFP_KERNEL | __GFP_ZERO)
+
+/* An illegal pointer value to mark the end of the shadow stack. */
+#define SCS_END_MAGIC		(0x5f6UL + POISON_POINTER_DELTA)
+
+/* Allocate a static per-CPU shadow stack */
+#define DEFINE_SCS(name)						\
+	DEFINE_PER_CPU(unsigned long [SCS_SIZE/sizeof(long)], name)	\
+
+#define task_scs(tsk)		(task_thread_info(tsk)->scs_base)
+#define task_scs_sp(tsk)	(task_thread_info(tsk)->scs_sp)
+
+void scs_init(void);
+int scs_prepare(struct task_struct *tsk, int node);
+void scs_release(struct task_struct *tsk);
+
+static inline void scs_task_reset(struct task_struct *tsk)
+{
+	/*
+	 * Reset the shadow stack to the base address in case the task
+	 * is reused.
+	 */
+	task_scs_sp(tsk) = task_scs(tsk);
+}
+
+static inline unsigned long *__scs_magic(void *s)
+{
+	return (unsigned long *)(s + SCS_SIZE) - 1;
+}
+
+static inline bool task_scs_end_corrupted(struct task_struct *tsk)
+{
+	unsigned long *magic = __scs_magic(task_scs(tsk));
+	unsigned long sz = task_scs_sp(tsk) - task_scs(tsk);
+
+	return sz >= SCS_SIZE - 1 || READ_ONCE_NOCHECK(*magic) != SCS_END_MAGIC;
+}
+
+#else /* CONFIG_SHADOW_CALL_STACK */
+
+static inline void scs_init(void) {}
+static inline void scs_task_reset(struct task_struct *tsk) {}
+static inline int scs_prepare(struct task_struct *tsk, int node) { return 0; }
+static inline void scs_release(struct task_struct *tsk) {}
+static inline bool task_scs_end_corrupted(struct task_struct *tsk) { return false; }
+
+#endif /* CONFIG_SHADOW_CALL_STACK */
+
+#endif /* _LINUX_SCS_H */