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
  ...

4 files changed, 61 insertions, 11 deletions

diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile
index 3862cad2410c..556d424c6f52 100644
--- a/arch/arm64/kernel/vdso/Makefile
+++ b/arch/arm64/kernel/vdso/Makefile
@@ -17,15 +17,19 @@ obj-vdso := vgettimeofday.o note.o sigreturn.o
 targets := $(obj-vdso) vdso.so vdso.so.dbg
 obj-vdso := $(addprefix $(obj)/, $(obj-vdso))
 
+btildflags-$(CONFIG_ARM64_BTI_KERNEL) += -z force-bti
+
+# -Bsymbolic has been added for consistency with arm, the compat vDSO and
+# potential future proofing if we end up with internal calls to the exported
+# routines, as x86 does (see 6f121e548f83 ("x86, vdso: Reimplement vdso.so
+# preparation in build-time C")).
 ldflags-y := -shared -nostdlib -soname=linux-vdso.so.1 --hash-style=sysv \
-		--build-id -n -T
+		-Bsymbolic --eh-frame-hdr --build-id -n $(btildflags-y) -T
 
 ccflags-y := -fno-common -fno-builtin -fno-stack-protector -ffixed-x18
 ccflags-y += -DDISABLE_BRANCH_PROFILING
 
-VDSO_LDFLAGS := -Bsymbolic
-
-CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) -Os
+CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) -Os $(CC_FLAGS_SCS)
 KBUILD_CFLAGS			+= $(DISABLE_LTO)
 KASAN_SANITIZE			:= n
 UBSAN_SANITIZE			:= n
diff --git a/arch/arm64/kernel/vdso/note.S b/arch/arm64/kernel/vdso/note.S
index 0ce6ec75a525..3d4e82290c80 100644
--- a/arch/arm64/kernel/vdso/note.S
+++ b/arch/arm64/kernel/vdso/note.S
@@ -12,9 +12,12 @@
 #include <linux/version.h>
 #include <linux/elfnote.h>
 #include <linux/build-salt.h>
+#include <asm/assembler.h>
 
 ELFNOTE_START(Linux, 0, "a")
 	.long LINUX_VERSION_CODE
 ELFNOTE_END
 
 BUILD_SALT
+
+emit_aarch64_feature_1_and
diff --git a/arch/arm64/kernel/vdso/sigreturn.S b/arch/arm64/kernel/vdso/sigreturn.S
index 12324863d5c2..620a3ef837b7 100644
--- a/arch/arm64/kernel/vdso/sigreturn.S
+++ b/arch/arm64/kernel/vdso/sigreturn.S
@@ -1,7 +1,11 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Sigreturn trampoline for returning from a signal when the SA_RESTORER
- * flag is not set.
+ * flag is not set. It serves primarily as a hall of shame for crappy
+ * unwinders and features an exciting but mysterious NOP instruction.
+ *
+ * It's also fragile as hell, so please think twice before changing anything
+ * in here.
  *
  * Copyright (C) 2012 ARM Limited
  *
@@ -9,18 +13,54 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/assembler.h>
 #include <asm/unistd.h>
 
 	.text
 
-	nop
-SYM_FUNC_START(__kernel_rt_sigreturn)
+/* Ensure that the mysterious NOP can be associated with a function. */
 	.cfi_startproc
+
+/*
+ * .cfi_signal_frame causes the corresponding Frame Description Entry in the
+ * .eh_frame section to be annotated as a signal frame. This allows DWARF
+ * unwinders (e.g. libstdc++) to implement _Unwind_GetIPInfo(), which permits
+ * unwinding out of the signal trampoline without the need for the mysterious
+ * NOP.
+ */
 	.cfi_signal_frame
-	.cfi_def_cfa	x29, 0
-	.cfi_offset	x29, 0 * 8
-	.cfi_offset	x30, 1 * 8
+
+/*
+ * Tell the unwinder where to locate the frame record linking back to the
+ * interrupted context. We don't provide unwind info for registers other
+ * than the frame pointer and the link register here; in practice, this
+ * is sufficient for unwinding in C/C++ based runtimes and the values in
+ * the sigcontext may have been modified by this point anyway. Debuggers
+ * already have baked-in strategies for attempting to unwind out of signals.
+ */
+	.cfi_def_cfa    x29, 0
+	.cfi_offset     x29, 0 * 8
+	.cfi_offset     x30, 1 * 8
+
+/*
+ * This mysterious NOP is required for some unwinders (e.g. libc++) that
+ * unconditionally subtract one from the result of _Unwind_GetIP() in order to
+ * identify the calling function.
+ * Hack borrowed from arch/powerpc/kernel/vdso64/sigtramp.S.
+ */
+	nop	// Mysterious NOP
+
+/*
+ * GDB relies on being able to identify the sigreturn instruction sequence to
+ * unwind from signal handlers. We cannot, therefore, use SYM_FUNC_START()
+ * here, as it will emit a BTI C instruction and break the unwinder. Thankfully,
+ * this function is only ever called from a RET and so omitting the landing pad
+ * is perfectly fine.
+ */
+SYM_CODE_START(__kernel_rt_sigreturn)
 	mov	x8, #__NR_rt_sigreturn
 	svc	#0
 	.cfi_endproc
-SYM_FUNC_END(__kernel_rt_sigreturn)
+SYM_CODE_END(__kernel_rt_sigreturn)
+
+emit_aarch64_feature_1_and
diff --git a/arch/arm64/kernel/vdso/vdso.S b/arch/arm64/kernel/vdso/vdso.S
index d1414fee5274..c4b1990bf2be 100644
--- a/arch/arm64/kernel/vdso/vdso.S
+++ b/arch/arm64/kernel/vdso/vdso.S
@@ -8,6 +8,7 @@
 #include <linux/init.h>
 #include <linux/linkage.h>
 #include <linux/const.h>
+#include <asm/assembler.h>
 #include <asm/page.h>
 
 	.globl vdso_start, vdso_end
@@ -19,3 +20,5 @@ vdso_start:
 vdso_end:
 
 	.previous
+
+emit_aarch64_feature_1_and