Merge branch 'uprobes/core' of...

Merge branch 'uprobes/core' of git://git.kernel.org/pub/scm/linux/kernel/git/oleg/misc into perf/core

Improve uprobes performance by adding 'pre-filtering' support,
by Oleg Nesterov:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m0.040s
	user    0m0.027s
	sys     0m0.010s

	# perf probe -x /lib/libc.so.6 syscall
	# perf record -e probe_libc:syscall sleep 100 &

Before this series:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m1.714s
	user    0m0.103s
	sys     0m1.607s

After:

	# time perl -e 'syscall -1 for 1..100_000'
	real    0m0.037s
	user    0m0.013s
	sys     0m0.023s
Signed-off-by: Ingo Molnar <mingo@kernel.org>

arch/x86/kernel/uprobes.c

@@ -680,8 +680,10 @@ static bool __skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
 		if (auprobe->insn[i] == 0x66)
 			continue;
 
-		if (auprobe->insn[i] == 0x90)
+		if (auprobe->insn[i] == 0x90) {
+			regs->ip += i + 1;
 			return true;
+		}
 
 		break;
 	}

include/linux/perf_event.h

@@ -135,16 +135,21 @@ struct hw_perf_event {
 		struct { /* software */
 			struct hrtimer	hrtimer;
 		};
+		struct { /* tracepoint */
+			struct task_struct	*tp_target;
+			/* for tp_event->class */
+			struct list_head	tp_list;
+		};
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 		struct { /* breakpoint */
-			struct arch_hw_breakpoint	info;
-			struct list_head		bp_list;
 			/*
 			 * Crufty hack to avoid the chicken and egg
 			 * problem hw_breakpoint has with context
 			 * creation and event initalization.
 			 */
 			struct task_struct		*bp_target;
+			struct arch_hw_breakpoint	info;
+			struct list_head		bp_list;
 		};
 #endif
 	};

include/linux/uprobes.h

@@ -35,13 +35,20 @@ struct inode;
 # include <asm/uprobes.h>
 #endif
 
+#define UPROBE_HANDLER_REMOVE		1
+#define UPROBE_HANDLER_MASK		1
+
+enum uprobe_filter_ctx {
+	UPROBE_FILTER_REGISTER,
+	UPROBE_FILTER_UNREGISTER,
+	UPROBE_FILTER_MMAP,
+};
+
 struct uprobe_consumer {
 	int (*handler)(struct uprobe_consumer *self, struct pt_regs *regs);
-	/*
-	 * filter is optional; If a filter exists, handler is run
-	 * if and only if filter returns true.
-	 */
-	bool (*filter)(struct uprobe_consumer *self, struct task_struct *task);
+	bool (*filter)(struct uprobe_consumer *self,
+				enum uprobe_filter_ctx ctx,
+				struct mm_struct *mm);
 
 	struct uprobe_consumer *next;
 };
@@ -94,6 +101,7 @@ extern int __weak set_swbp(struct arch_uprobe *aup, struct mm_struct *mm, unsign
 extern int __weak set_orig_insn(struct arch_uprobe *aup, struct mm_struct *mm, unsigned long vaddr);
 extern bool __weak is_swbp_insn(uprobe_opcode_t *insn);
 extern int uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *uc);
+extern int uprobe_apply(struct inode *inode, loff_t offset, struct uprobe_consumer *uc, bool);
 extern void uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consumer *uc);
 extern int uprobe_mmap(struct vm_area_struct *vma);
 extern void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end);
@@ -117,6 +125,11 @@ uprobe_register(struct inode *inode, loff_t offset, struct uprobe_consumer *uc)
 {
 	return -ENOSYS;
 }
+static inline int
+uprobe_apply(struct inode *inode, loff_t offset, struct uprobe_consumer *uc, bool add)
+{
+	return -ENOSYS;
+}
 static inline void
 uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consumer *uc)
 {

kernel/events/core.c

@@ -6162,11 +6162,14 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu,
 
 	if (task) {
 		event->attach_state = PERF_ATTACH_TASK;
+
+		if (attr->type == PERF_TYPE_TRACEPOINT)
+			event->hw.tp_target = task;
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 		/*
 		 * hw_breakpoint is a bit difficult here..
 		 */
-		if (attr->type == PERF_TYPE_BREAKPOINT)
+		else if (attr->type == PERF_TYPE_BREAKPOINT)
 			event->hw.bp_target = task;
 #endif
 	}

kernel/ptrace.c

@@ -712,6 +712,12 @@ static int ptrace_regset(struct task_struct *task, int req, unsigned int type,
 					     kiov->iov_len, kiov->iov_base);
 }
 
+/*
+ * This is declared in linux/regset.h and defined in machine-dependent
+ * code.  We put the export here, near the primary machine-neutral use,
+ * to ensure no machine forgets it.
+ */
+EXPORT_SYMBOL_GPL(task_user_regset_view);
 #endif
 
 int ptrace_request(struct task_struct *child, long request,

kernel/trace/trace_probe.h

@@ -66,7 +66,6 @@
 #define TP_FLAG_TRACE		1
 #define TP_FLAG_PROFILE		2
 #define TP_FLAG_REGISTERED	4
-#define TP_FLAG_UPROBE		8
 
 
 /* data_rloc: data relative location, compatible with u32 */