bpf: Add bpf_probe_write_user BPF helper to be called in tracers

This allows user memory to be written to during the course of a kprobe.
It shouldn't be used to implement any kind of security mechanism
because of TOC-TOU attacks, but rather to debug, divert, and
manipulate execution of semi-cooperative processes.

Although it uses probe_kernel_write, we limit the address space
the probe can write into by checking the space with access_ok.
We do this as opposed to calling copy_to_user directly, in order
to avoid sleeping. In addition we ensure the threads's current fs
/ segment is USER_DS and the thread isn't exiting nor a kernel thread.

Given this feature is meant for experiments, and it has a risk of
crashing the system, and running programs, we print a warning on
when a proglet that attempts to use this helper is installed,
along with the pid and process name.

Signed-off-by: Sargun Dhillon <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Acked-by: Alexei Starovoitov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

include/uapi/linux/bpf.h

@@ -365,6 +365,16 @@ enum bpf_func_id {
 	 */
 	BPF_FUNC_get_current_task,
 
+	/**
+	 * bpf_probe_write_user(void *dst, void *src, int len)
+	 * safely attempt to write to a location
+	 * @dst: destination address in userspace
+	 * @src: source address on stack
+	 * @len: number of bytes to copy
+	 * Return: 0 on success or negative error
+	 */
+	BPF_FUNC_probe_write_user,
+
 	__BPF_FUNC_MAX_ID,
 };
 

kernel/trace/bpf_trace.c

@@ -81,6 +81,49 @@ static const struct bpf_func_proto bpf_probe_read_proto = {
 	.arg3_type	= ARG_ANYTHING,
 };
 
+static u64 bpf_probe_write_user(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
+{
+	void *unsafe_ptr = (void *) (long) r1;
+	void *src = (void *) (long) r2;
+	int size = (int) r3;
+
+	/*
+	 * Ensure we're in user context which is safe for the helper to
+	 * run. This helper has no business in a kthread.
+	 *
+	 * access_ok() should prevent writing to non-user memory, but in
+	 * some situations (nommu, temporary switch, etc) access_ok() does
+	 * not provide enough validation, hence the check on KERNEL_DS.
+	 */
+
+	if (unlikely(in_interrupt() ||
+		     current->flags & (PF_KTHREAD | PF_EXITING)))
+		return -EPERM;
+	if (unlikely(segment_eq(get_fs(), KERNEL_DS)))
+		return -EPERM;
+	if (!access_ok(VERIFY_WRITE, unsafe_ptr, size))
+		return -EPERM;
+
+	return probe_kernel_write(unsafe_ptr, src, size);
+}
+
+static const struct bpf_func_proto bpf_probe_write_user_proto = {
+	.func		= bpf_probe_write_user,
+	.gpl_only	= true,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_ANYTHING,
+	.arg2_type	= ARG_PTR_TO_STACK,
+	.arg3_type	= ARG_CONST_STACK_SIZE,
+};
+
+static const struct bpf_func_proto *bpf_get_probe_write_proto(void)
+{
+	pr_warn_ratelimited("%s[%d] is installing a program with bpf_probe_write_user helper that may corrupt user memory!",
+			    current->comm, task_pid_nr(current));
+
+	return &bpf_probe_write_user_proto;
+}
+
 /*
  * limited trace_printk()
  * only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed
@@ -362,6 +405,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id)
 		return &bpf_get_smp_processor_id_proto;
 	case BPF_FUNC_perf_event_read:
 		return &bpf_perf_event_read_proto;
+	case BPF_FUNC_probe_write_user:
+		return bpf_get_probe_write_proto();
 	default:
 		return NULL;
 	}

samples/bpf/bpf_helpers.h

@@ -41,6 +41,8 @@ static int (*bpf_perf_event_output)(void *ctx, void *map, int index, void *data,
 	(void *) BPF_FUNC_perf_event_output;
 static int (*bpf_get_stackid)(void *ctx, void *map, int flags) =
 	(void *) BPF_FUNC_get_stackid;
+static int (*bpf_probe_write_user)(void *dst, void *src, int size) =
+	(void *) BPF_FUNC_probe_write_user;
 
 /* llvm builtin functions that eBPF C program may use to
  * emit BPF_LD_ABS and BPF_LD_IND instructions