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

Pull arm64 updates from Will Deacon:
 "Highlights include a major rework of our kPTI page-table rewriting
  code (which makes it both more maintainable and considerably faster in
  the cases where it is required) as well as significant changes to our
  early boot code to reduce the need for data cache maintenance and
  greatly simplify the KASLR relocation dance.

  Summary:

   - Remove unused generic cpuidle support (replaced by PSCI version)

   - Fix documentation describing the kernel virtual address space

   - Handling of some new CPU errata in Arm implementations

   - Rework of our exception table code in preparation for handling
     machine checks (i.e. RAS errors) more gracefully

   - Switch over to the generic implementation of ioremap()

   - Fix lockdep tracking in NMI context

   - Instrument our memory barrier macros for KCSAN

   - Rework of the kPTI G->nG page-table repainting so that the MMU
     remains enabled and the boot time is no longer slowed to a crawl
     for systems which require the late remapping

   - Enable support for direct swapping of 2MiB transparent huge-pages
     on systems without MTE

   - Fix handling of MTE tags with allocating new pages with HW KASAN

   - Expose the SMIDR register to userspace via sysfs

   - Continued rework of the stack unwinder, particularly improving the
     behaviour under KASAN

   - More repainting of our system register definitions to match the
     architectural terminology

   - Improvements to the layout of the vDSO objects

   - Support for allocating additional bits of HWCAP2 and exposing
     FEAT_EBF16 to userspace on CPUs that support it

   - Considerable rework and optimisation of our early boot code to
     reduce the need for cache maintenance and avoid jumping in and out
     of the kernel when handling relocation under KASLR

   - Support for disabling SVE and SME support on the kernel
     command-line

   - Support for the Hisilicon HNS3 PMU

   - Miscellanous cleanups, trivial updates and minor fixes"

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (136 commits)
  arm64: Delay initialisation of cpuinfo_arm64::reg_{zcr,smcr}
  arm64: fix KASAN_INLINE
  arm64/hwcap: Support FEAT_EBF16
  arm64/cpufeature: Store elf_hwcaps as a bitmap rather than unsigned long
  arm64/hwcap: Document allocation of upper bits of AT_HWCAP
  arm64: enable THP_SWAP for arm64
  arm64/mm: use GENMASK_ULL for TTBR_BADDR_MASK_52
  arm64: errata: Remove AES hwcap for COMPAT tasks
  arm64: numa: Don't check node against MAX_NUMNODES
  drivers/perf: arm_spe: Fix consistency of SYS_PMSCR_EL1.CX
  perf: RISC-V: Add of_node_put() when breaking out of for_each_of_cpu_node()
  docs: perf: Include hns3-pmu.rst in toctree to fix 'htmldocs' WARNING
  arm64: kasan: Revert "arm64: mte: reset the page tag in page->flags"
  mm: kasan: Skip page unpoisoning only if __GFP_SKIP_KASAN_UNPOISON
  mm: kasan: Skip unpoisoning of user pages
  mm: kasan: Ensure the tags are visible before the tag in page->flags
  drivers/perf: hisi: add driver for HNS3 PMU
  drivers/perf: hisi: Add description for HNS3 PMU driver
  drivers/perf: riscv_pmu_sbi: perf format
  perf/arm-cci: Use the bitmap API to allocate bitmaps
  ...

1 files changed, 75 insertions, 24 deletions

diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c
index 0467cb79f080..fcaa151b81f1 100644
--- a/arch/arm64/kernel/stacktrace.c
+++ b/arch/arm64/kernel/stacktrace.c
@@ -38,6 +38,8 @@
  * @kr_cur:      When KRETPROBES is selected, holds the kretprobe instance
  *               associated with the most recently encountered replacement lr
  *               value.
+ *
+ * @task:        The task being unwound.
  */
 struct unwind_state {
 	unsigned long fp;
@@ -48,13 +50,13 @@ struct unwind_state {
 #ifdef CONFIG_KRETPROBES
 	struct llist_node *kr_cur;
 #endif
+	struct task_struct *task;
 };
 
-static notrace void unwind_init(struct unwind_state *state, unsigned long fp,
-				unsigned long pc)
+static void unwind_init_common(struct unwind_state *state,
+			       struct task_struct *task)
 {
-	state->fp = fp;
-	state->pc = pc;
+	state->task = task;
 #ifdef CONFIG_KRETPROBES
 	state->kr_cur = NULL;
 #endif
@@ -72,7 +74,57 @@ static notrace void unwind_init(struct unwind_state *state, unsigned long fp,
 	state->prev_fp = 0;
 	state->prev_type = STACK_TYPE_UNKNOWN;
 }
-NOKPROBE_SYMBOL(unwind_init);
+
+/*
+ * Start an unwind from a pt_regs.
+ *
+ * The unwind will begin at the PC within the regs.
+ *
+ * The regs must be on a stack currently owned by the calling task.
+ */
+static inline void unwind_init_from_regs(struct unwind_state *state,
+					 struct pt_regs *regs)
+{
+	unwind_init_common(state, current);
+
+	state->fp = regs->regs[29];
+	state->pc = regs->pc;
+}
+
+/*
+ * Start an unwind from a caller.
+ *
+ * The unwind will begin at the caller of whichever function this is inlined
+ * into.
+ *
+ * The function which invokes this must be noinline.
+ */
+static __always_inline void unwind_init_from_caller(struct unwind_state *state)
+{
+	unwind_init_common(state, current);
+
+	state->fp = (unsigned long)__builtin_frame_address(1);
+	state->pc = (unsigned long)__builtin_return_address(0);
+}
+
+/*
+ * Start an unwind from a blocked task.
+ *
+ * The unwind will begin at the blocked tasks saved PC (i.e. the caller of
+ * cpu_switch_to()).
+ *
+ * The caller should ensure the task is blocked in cpu_switch_to() for the
+ * duration of the unwind, or the unwind will be bogus. It is never valid to
+ * call this for the current task.
+ */
+static inline void unwind_init_from_task(struct unwind_state *state,
+					 struct task_struct *task)
+{
+	unwind_init_common(state, task);
+
+	state->fp = thread_saved_fp(task);
+	state->pc = thread_saved_pc(task);
+}
 
 /*
  * Unwind from one frame record (A) to the next frame record (B).
@@ -81,9 +133,9 @@ NOKPROBE_SYMBOL(unwind_init);
  * records (e.g. a cycle), determined based on the location and fp value of A
  * and the location (but not the fp value) of B.
  */
-static int notrace unwind_next(struct task_struct *tsk,
-			       struct unwind_state *state)
+static int notrace unwind_next(struct unwind_state *state)
 {
+	struct task_struct *tsk = state->task;
 	unsigned long fp = state->fp;
 	struct stack_info info;
 
@@ -117,15 +169,15 @@ static int notrace unwind_next(struct task_struct *tsk,
 		if (fp <= state->prev_fp)
 			return -EINVAL;
 	} else {
-		set_bit(state->prev_type, state->stacks_done);
+		__set_bit(state->prev_type, state->stacks_done);
 	}
 
 	/*
 	 * Record this frame record's values and location. The prev_fp and
 	 * prev_type are only meaningful to the next unwind_next() invocation.
 	 */
-	state->fp = READ_ONCE_NOCHECK(*(unsigned long *)(fp));
-	state->pc = READ_ONCE_NOCHECK(*(unsigned long *)(fp + 8));
+	state->fp = READ_ONCE(*(unsigned long *)(fp));
+	state->pc = READ_ONCE(*(unsigned long *)(fp + 8));
 	state->prev_fp = fp;
 	state->prev_type = info.type;
 
@@ -157,8 +209,7 @@ static int notrace unwind_next(struct task_struct *tsk,
 }
 NOKPROBE_SYMBOL(unwind_next);
 
-static void notrace unwind(struct task_struct *tsk,
-			   struct unwind_state *state,
+static void notrace unwind(struct unwind_state *state,
 			   stack_trace_consume_fn consume_entry, void *cookie)
 {
 	while (1) {
@@ -166,7 +217,7 @@ static void notrace unwind(struct task_struct *tsk,
 
 		if (!consume_entry(cookie, state->pc))
 			break;
-		ret = unwind_next(tsk, state);
+		ret = unwind_next(state);
 		if (ret < 0)
 			break;
 	}
@@ -212,15 +263,15 @@ noinline notrace void arch_stack_walk(stack_trace_consume_fn consume_entry,
 {
 	struct unwind_state state;
 
-	if (regs)
-		unwind_init(&state, regs->regs[29], regs->pc);
-	else if (task == current)
-		unwind_init(&state,
-				(unsigned long)__builtin_frame_address(1),
-				(unsigned long)__builtin_return_address(0));
-	else
-		unwind_init(&state, thread_saved_fp(task),
-				thread_saved_pc(task));
-
-	unwind(task, &state, consume_entry, cookie);
+	if (regs) {
+		if (task != current)
+			return;
+		unwind_init_from_regs(&state, regs);
+	} else if (task == current) {
+		unwind_init_from_caller(&state);
+	} else {
+		unwind_init_from_task(&state, task);
+	}
+
+	unwind(&state, consume_entry, cookie);
 }