Merge branch 'bpf-override-return'

Josef Bacik says:

====================
This is the same as v8, just rebased onto the bpf tree.

v8->v9:
- rebased onto the bpf tree.

v7->v8:
- removed the _ASM_KPROBE_ERROR_INJECT since it was not needed.

v6->v7:
- moved the opt-in macro to bpf.h out of kprobes.h.

v5->v6:
- add BPF_ALLOW_ERROR_INJECTION() tagging for functions that will support this
  feature.  This way only functions that opt-in will be allowed to be
  overridden.
- added a btrfs patch to allow error injection for open_ctree() so that the bpf
  sample actually works.

v4->v5:
- disallow kprobe_override programs from being put in the prog map array so we
  don't tail call into something we didn't check.  This allows us to make the
  normal path still fast without a bunch of percpu operations.

v3->v4:
- fix a build error found by kbuild test bot (I didn't wait long enough
  apparently.)
- Added a warning message as per Daniels suggestion.

v2->v3:
- added a ->kprobe_override flag to bpf_prog.
- added some sanity checks to disallow attaching bpf progs that have
  ->kprobe_override set that aren't for ftrace kprobes.
- added the trace_kprobe_ftrace helper to check if the trace_event_call is a
  ftrace kprobe.
- renamed bpf_kprobe_state to bpf_kprobe_override, fixed it so we only read this
  value in the kprobe path, and thus only write to it if we're overriding or
  clearing the override.

v1->v2:
- moved things around to make sure that bpf_override_return could really only be
  used for an ftrace kprobe.
- killed the special return values from trace_call_bpf.
- renamed pc_modified to bpf_kprobe_state so bpf_override_return could tell if
  it was being called from an ftrace kprobe context.
- reworked the logic in kprobe_perf_func to take advantage of bpf_kprobe_state.
- updated the test as per Alexei's review.

- Original message -

A lot of our error paths are not well tested because we have no good way of
injecting errors generically.  Some subystems (block, memory) have ways to
inject errors, but they are random so it's hard to get reproduceable results.

With BPF we can add determinism to our error injection.  We can use kprobes and
other things to verify we are injecting errors at the exact case we are trying
to test.  This patch gives us the tool to actual do the error injection part.
It is very simple, we just set the return value of the pt_regs we're given to
whatever we provide, and then override the PC with a dummy function that simply
returns.

Right now this only works on x86, but it would be simple enough to expand to
other architectures.  Thanks,

Josef
====================

In patch "bpf: add a bpf_override_function helper" Alexei moved
"ifdef CONFIG_BPF_KPROBE_OVERRIDE" few lines to fail program loading
when kprobe_override is not available instead of failing at run-time.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

9 files changed, 286 insertions, 8 deletions

diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index b16c6f8..d32bebf 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1320,6 +1320,9 @@ EVAL4(PROG_NAME_LIST, 416, 448, 480, 512)
 bool bpf_prog_array_compatible(struct bpf_array *array,
 			       const struct bpf_prog *fp)
 {
+	if (fp->kprobe_override)
+		return false;
+
 	if (!array->owner_prog_type) {
 		/* There's no owner yet where we could check for
 		 * compatibility.
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 7afa92e..e807bda 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -4413,6 +4413,8 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env)
 			prog->dst_needed = 1;
 		if (insn->imm == BPF_FUNC_get_prandom_u32)
 			bpf_user_rnd_init_once();
+		if (insn->imm == BPF_FUNC_override_return)
+			prog->kprobe_override = 1;
 		if (insn->imm == BPF_FUNC_tail_call) {
 			/* If we tail call into other programs, we
 			 * cannot make any assumptions since they can
diff --git a/kernel/events/core.c b/kernel/events/core.c
index f10609e..5857c500 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8080,6 +8080,13 @@ static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd)
 		return -EINVAL;
 	}
 
+	/* Kprobe override only works for kprobes, not uprobes. */
+	if (prog->kprobe_override &&
+	    !(event->tp_event->flags & TRACE_EVENT_FL_KPROBE)) {
+		bpf_prog_put(prog);
+		return -EINVAL;
+	}
+
 	if (is_tracepoint || is_syscall_tp) {
 		int off = trace_event_get_offsets(event->tp_event);
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index da2ccf1..b4aab48 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -83,6 +83,16 @@ static raw_spinlock_t *kretprobe_table_lock_ptr(unsigned long hash)
 	return &(kretprobe_table_locks[hash].lock);
 }
 
+/* List of symbols that can be overriden for error injection. */
+static LIST_HEAD(kprobe_error_injection_list);
+static DEFINE_MUTEX(kprobe_ei_mutex);
+struct kprobe_ei_entry {
+	struct list_head list;
+	unsigned long start_addr;
+	unsigned long end_addr;
+	void *priv;
+};
+
 /* Blacklist -- list of struct kprobe_blacklist_entry */
 static LIST_HEAD(kprobe_blacklist);
 
@@ -1394,6 +1404,17 @@ bool within_kprobe_blacklist(unsigned long addr)
 	return false;
 }
 
+bool within_kprobe_error_injection_list(unsigned long addr)
+{
+	struct kprobe_ei_entry *ent;
+
+	list_for_each_entry(ent, &kprobe_error_injection_list, list) {
+		if (addr >= ent->start_addr && addr < ent->end_addr)
+			return true;
+	}
+	return false;
+}
+
 /*
  * If we have a symbol_name argument, look it up and add the offset field
  * to it. This way, we can specify a relative address to a symbol.
@@ -2168,6 +2189,86 @@ static int __init populate_kprobe_blacklist(unsigned long *start,
 	return 0;
 }
 
+#ifdef CONFIG_BPF_KPROBE_OVERRIDE
+/* Markers of the _kprobe_error_inject_list section */
+extern unsigned long __start_kprobe_error_inject_list[];
+extern unsigned long __stop_kprobe_error_inject_list[];
+
+/*
+ * Lookup and populate the kprobe_error_injection_list.
+ *
+ * For safety reasons we only allow certain functions to be overriden with
+ * bpf_error_injection, so we need to populate the list of the symbols that have
+ * been marked as safe for overriding.
+ */
+static void populate_kprobe_error_injection_list(unsigned long *start,
+						 unsigned long *end,
+						 void *priv)
+{
+	unsigned long *iter;
+	struct kprobe_ei_entry *ent;
+	unsigned long entry, offset = 0, size = 0;
+
+	mutex_lock(&kprobe_ei_mutex);
+	for (iter = start; iter < end; iter++) {
+		entry = arch_deref_entry_point((void *)*iter);
+
+		if (!kernel_text_address(entry) ||
+		    !kallsyms_lookup_size_offset(entry, &size, &offset)) {
+			pr_err("Failed to find error inject entry at %p\n",
+				(void *)entry);
+			continue;
+		}
+
+		ent = kmalloc(sizeof(*ent), GFP_KERNEL);
+		if (!ent)
+			break;
+		ent->start_addr = entry;
+		ent->end_addr = entry + size;
+		ent->priv = priv;
+		INIT_LIST_HEAD(&ent->list);
+		list_add_tail(&ent->list, &kprobe_error_injection_list);
+	}
+	mutex_unlock(&kprobe_ei_mutex);
+}
+
+static void __init populate_kernel_kprobe_ei_list(void)
+{
+	populate_kprobe_error_injection_list(__start_kprobe_error_inject_list,
+					     __stop_kprobe_error_inject_list,
+					     NULL);
+}
+
+static void module_load_kprobe_ei_list(struct module *mod)
+{
+	if (!mod->num_kprobe_ei_funcs)
+		return;
+	populate_kprobe_error_injection_list(mod->kprobe_ei_funcs,
+					     mod->kprobe_ei_funcs +
+					     mod->num_kprobe_ei_funcs, mod);
+}
+
+static void module_unload_kprobe_ei_list(struct module *mod)
+{
+	struct kprobe_ei_entry *ent, *n;
+	if (!mod->num_kprobe_ei_funcs)
+		return;
+
+	mutex_lock(&kprobe_ei_mutex);
+	list_for_each_entry_safe(ent, n, &kprobe_error_injection_list, list) {
+		if (ent->priv == mod) {
+			list_del_init(&ent->list);
+			kfree(ent);
+		}
+	}
+	mutex_unlock(&kprobe_ei_mutex);
+}
+#else
+static inline void __init populate_kernel_kprobe_ei_list(void) {}
+static inline void module_load_kprobe_ei_list(struct module *m) {}
+static inline void module_unload_kprobe_ei_list(struct module *m) {}
+#endif
+
 /* Module notifier call back, checking kprobes on the module */
 static int kprobes_module_callback(struct notifier_block *nb,
 				   unsigned long val, void *data)
@@ -2178,6 +2279,11 @@ static int kprobes_module_callback(struct notifier_block *nb,
 	unsigned int i;
 	int checkcore = (val == MODULE_STATE_GOING);
 
+	if (val == MODULE_STATE_COMING)
+		module_load_kprobe_ei_list(mod);
+	else if (val == MODULE_STATE_GOING)
+		module_unload_kprobe_ei_list(mod);
+
 	if (val != MODULE_STATE_GOING && val != MODULE_STATE_LIVE)
 		return NOTIFY_DONE;
 
@@ -2240,6 +2346,8 @@ static int __init init_kprobes(void)
 		pr_err("Please take care of using kprobes.\n");
 	}
 
+	populate_kernel_kprobe_ei_list();
+
 	if (kretprobe_blacklist_size) {
 		/* lookup the function address from its name */
 		for (i = 0; kretprobe_blacklist[i].name != NULL; i++) {
@@ -2407,6 +2515,56 @@ static const struct file_operations debugfs_kprobe_blacklist_ops = {
 	.release        = seq_release,
 };
 
+/*
+ * kprobes/error_injection_list -- shows which functions can be overriden for
+ * error injection.
+ * */
+static void *kprobe_ei_seq_start(struct seq_file *m, loff_t *pos)
+{
+	mutex_lock(&kprobe_ei_mutex);
+	return seq_list_start(&kprobe_error_injection_list, *pos);
+}
+
+static void kprobe_ei_seq_stop(struct seq_file *m, void *v)
+{
+	mutex_unlock(&kprobe_ei_mutex);
+}
+
+static void *kprobe_ei_seq_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	return seq_list_next(v, &kprobe_error_injection_list, pos);
+}
+
+static int kprobe_ei_seq_show(struct seq_file *m, void *v)
+{
+	char buffer[KSYM_SYMBOL_LEN];
+	struct kprobe_ei_entry *ent =
+		list_entry(v, struct kprobe_ei_entry, list);
+
+	sprint_symbol(buffer, ent->start_addr);
+	seq_printf(m, "%s\n", buffer);
+	return 0;
+}
+
+static const struct seq_operations kprobe_ei_seq_ops = {
+	.start = kprobe_ei_seq_start,
+	.next  = kprobe_ei_seq_next,
+	.stop  = kprobe_ei_seq_stop,
+	.show  = kprobe_ei_seq_show,
+};
+
+static int kprobe_ei_open(struct inode *inode, struct file *filp)
+{
+	return seq_open(filp, &kprobe_ei_seq_ops);
+}
+
+static const struct file_operations debugfs_kprobe_ei_ops = {
+	.open           = kprobe_ei_open,
+	.read           = seq_read,
+	.llseek         = seq_lseek,
+	.release        = seq_release,
+};
+
 static void arm_all_kprobes(void)
 {
 	struct hlist_head *head;
@@ -2548,6 +2706,11 @@ static int __init debugfs_kprobe_init(void)
 	if (!file)
 		goto error;
 
+	file = debugfs_create_file("error_injection_list", 0444, dir, NULL,
+				  &debugfs_kprobe_ei_ops);
+	if (!file)
+		goto error;
+
 	return 0;
 
 error:
diff --git a/kernel/module.c b/kernel/module.c
index dea01ac..bd695bf 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -3118,7 +3118,11 @@ static int find_module_sections(struct module *mod, struct load_info *info)
 					     sizeof(*mod->ftrace_callsites),
 					     &mod->num_ftrace_callsites);
 #endif
-
+#ifdef CONFIG_BPF_KPROBE_OVERRIDE
+	mod->kprobe_ei_funcs = section_objs(info, "_kprobe_error_inject_list",
+					    sizeof(*mod->kprobe_ei_funcs),
+					    &mod->num_kprobe_ei_funcs);
+#endif
 	mod->extable = section_objs(info, "__ex_table",
 				    sizeof(*mod->extable), &mod->num_exentries);
 
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index af7dad1..3e6fd58 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -529,6 +529,17 @@ config FUNCTION_PROFILER
 
 	  If in doubt, say N.
 
+config BPF_KPROBE_OVERRIDE
+	bool "Enable BPF programs to override a kprobed function"
+	depends on BPF_EVENTS
+	depends on KPROBES_ON_FTRACE
+	depends on HAVE_KPROBE_OVERRIDE
+	depends on DYNAMIC_FTRACE_WITH_REGS
+	default n
+	help
+	 Allows BPF to override the execution of a probed function and
+	 set a different return value.  This is used for error injection.
+
 config FTRACE_MCOUNT_RECORD
 	def_bool y
 	depends on DYNAMIC_FTRACE
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index b143f2a..e009b7e 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -13,6 +13,10 @@
 #include <linux/filter.h>
 #include <linux/uaccess.h>
 #include <linux/ctype.h>
+#include <linux/kprobes.h>
+#include <asm/kprobes.h>
+
+#include "trace_probe.h"
 #include "trace.h"
 
 u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
@@ -76,6 +80,24 @@ unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx)
 }
 EXPORT_SYMBOL_GPL(trace_call_bpf);
 
+#ifdef CONFIG_BPF_KPROBE_OVERRIDE
+BPF_CALL_2(bpf_override_return, struct pt_regs *, regs, unsigned long, rc)
+{
+	__this_cpu_write(bpf_kprobe_override, 1);
+	regs_set_return_value(regs, rc);
+	arch_ftrace_kprobe_override_function(regs);
+	return 0;
+}
+
+static const struct bpf_func_proto bpf_override_return_proto = {
+	.func		= bpf_override_return,
+	.gpl_only	= true,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_ANYTHING,
+};
+#endif
+
 BPF_CALL_3(bpf_probe_read, void *, dst, u32, size, const void *, unsafe_ptr)
 {
 	int ret;
@@ -551,6 +573,10 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func
 		return &bpf_get_stackid_proto;
 	case BPF_FUNC_perf_event_read_value:
 		return &bpf_perf_event_read_value_proto;
+#ifdef CONFIG_BPF_KPROBE_OVERRIDE
+	case BPF_FUNC_override_return:
+		return &bpf_override_return_proto;
+#endif
 	default:
 		return tracing_func_proto(func_id);
 	}
@@ -768,6 +794,15 @@ int perf_event_attach_bpf_prog(struct perf_event *event,
 	struct bpf_prog_array *new_array;
 	int ret = -EEXIST;
 
+	/*
+	 * Kprobe override only works for ftrace based kprobes, and only if they
+	 * are on the opt-in list.
+	 */
+	if (prog->kprobe_override &&
+	    (!trace_kprobe_ftrace(event->tp_event) ||
+	     !trace_kprobe_error_injectable(event->tp_event)))
+		return -EINVAL;
+
 	mutex_lock(&bpf_event_mutex);
 
 	if (event->prog)
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index 492700c..5db8498 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -42,6 +42,7 @@ struct trace_kprobe {
 	(offsetof(struct trace_kprobe, tp.args) +	\
 	(sizeof(struct probe_arg) * (n)))
 
+DEFINE_PER_CPU(int, bpf_kprobe_override);
 
 static nokprobe_inline bool trace_kprobe_is_return(struct trace_kprobe *tk)
 {
@@ -87,6 +88,27 @@ static nokprobe_inline unsigned long trace_kprobe_nhit(struct trace_kprobe *tk)
 	return nhit;
 }
 
+int trace_kprobe_ftrace(struct trace_event_call *call)
+{
+	struct trace_kprobe *tk = (struct trace_kprobe *)call->data;
+	return kprobe_ftrace(&tk->rp.kp);
+}
+
+int trace_kprobe_error_injectable(struct trace_event_call *call)
+{
+	struct trace_kprobe *tk = (struct trace_kprobe *)call->data;
+	unsigned long addr;
+
+	if (tk->symbol) {
+		addr = (unsigned long)
+			kallsyms_lookup_name(trace_kprobe_symbol(tk));
+		addr += tk->rp.kp.offset;
+	} else {
+		addr = (unsigned long)tk->rp.kp.addr;
+	}
+	return within_kprobe_error_injection_list(addr);
+}
+
 static int register_kprobe_event(struct trace_kprobe *tk);
 static int unregister_kprobe_event(struct trace_kprobe *tk);
 
@@ -1170,7 +1192,7 @@ static int kretprobe_event_define_fields(struct trace_event_call *event_call)
 #ifdef CONFIG_PERF_EVENTS
 
 /* Kprobe profile handler */
-static void
+static int
 kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs)
 {
 	struct trace_event_call *call = &tk->tp.call;
@@ -1179,12 +1201,29 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs)
 	int size, __size, dsize;
 	int rctx;
 
-	if (bpf_prog_array_valid(call) && !trace_call_bpf(call, regs))
-		return;
+	if (bpf_prog_array_valid(call)) {
+		int ret;
+
+		ret = trace_call_bpf(call, regs);
+
+		/*
+		 * We need to check and see if we modified the pc of the
+		 * pt_regs, and if so clear the kprobe and return 1 so that we
+		 * don't do the instruction skipping.  Also reset our state so
+		 * we are clean the next pass through.
+		 */
+		if (__this_cpu_read(bpf_kprobe_override)) {
+			__this_cpu_write(bpf_kprobe_override, 0);
+			reset_current_kprobe();
+			return 1;
+		}
+		if (!ret)
+			return 0;
+	}
 
 	head = this_cpu_ptr(call->perf_events);
 	if (hlist_empty(head))
-		return;
+		return 0;
 
 	dsize = __get_data_size(&tk->tp, regs);
 	__size = sizeof(*entry) + tk->tp.size + dsize;
@@ -1193,13 +1232,14 @@ kprobe_perf_func(struct trace_kprobe *tk, struct pt_regs *regs)
 
 	entry = perf_trace_buf_alloc(size, NULL, &rctx);
 	if (!entry)
-		return;
+		return 0;
 
 	entry->ip = (unsigned long)tk->rp.kp.addr;
 	memset(&entry[1], 0, dsize);
 	store_trace_args(sizeof(*entry), &tk->tp, regs, (u8 *)&entry[1], dsize);
 	perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs,
 			      head, NULL);
+	return 0;
 }
 NOKPROBE_SYMBOL(kprobe_perf_func);
 
@@ -1275,6 +1315,7 @@ static int kprobe_register(struct trace_event_call *event,
 static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs)
 {
 	struct trace_kprobe *tk = container_of(kp, struct trace_kprobe, rp.kp);
+	int ret = 0;
 
 	raw_cpu_inc(*tk->nhit);
 
@@ -1282,9 +1323,9 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs)
 		kprobe_trace_func(tk, regs);
 #ifdef CONFIG_PERF_EVENTS
 	if (tk->tp.flags & TP_FLAG_PROFILE)
-		kprobe_perf_func(tk, regs);
+		ret = kprobe_perf_func(tk, regs);
 #endif
-	return 0;	/* We don't tweek kernel, so just return 0 */
+	return ret;
 }
 NOKPROBE_SYMBOL(kprobe_dispatcher);
 
diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h
index fb66e3e..5e54d74 100644
--- a/kernel/trace/trace_probe.h
+++ b/kernel/trace/trace_probe.h
@@ -252,6 +252,8 @@ struct symbol_cache;
 unsigned long update_symbol_cache(struct symbol_cache *sc);
 void free_symbol_cache(struct symbol_cache *sc);
 struct symbol_cache *alloc_symbol_cache(const char *sym, long offset);
+int trace_kprobe_ftrace(struct trace_event_call *call);
+int trace_kprobe_error_injectable(struct trace_event_call *call);
 #else
 /* uprobes do not support symbol fetch methods */
 #define fetch_symbol_u8			NULL
@@ -277,6 +279,16 @@ alloc_symbol_cache(const char *sym, long offset)
 {
 	return NULL;
 }
+
+static inline int trace_kprobe_ftrace(struct trace_event_call *call)
+{
+	return 0;
+}
+
+static inline int trace_kprobe_error_injectable(struct trace_event_call *call)
+{
+	return 0;
+}
 #endif /* CONFIG_KPROBE_EVENTS */
 
 struct probe_arg {