diff --git a/py/objstr.c b/py/objstr.c
index 3e275afb9a..e0cbea577c 100644
--- a/py/objstr.c
+++ b/py/objstr.c
@@ -375,33 +375,58 @@ STATIC mp_obj_t str_subscr(mp_obj_t self_in, mp_obj_t index, mp_obj_t value) {
             return mp_obj_new_str_of_type(type, self_data + slice.start, slice.stop - slice.start);
         }
 #endif
-        // TODO: Don't use mp_get_index() here
-        uint index_val = mp_get_index(type, unichar_charlen((const char *)self_data, self_len), index, false);
         if (type == &mp_type_bytes) {
+            uint index_val = mp_get_index(type, self_len, index, false);
             return MP_OBJ_NEW_SMALL_INT((mp_small_int_t)self_data[index_val]);
-        } else {
-            // Count non-continuation bytes to count characters.
-            // Assumes that the string is correctly formed - will run past the
-            // end of the buffer if there aren't that many characters in it
-            const char *s;
-            for (s=(const char *)self_data; index_val; ++s) {
+        }
+        const char *s, *top = (const char *)self_data + self_len;
+        machine_int_t i;
+        // Copied from mp_get_index; I don't want bounds checking, just give me
+        // the integer as-is. (I can't bounds-check without scanning the whole
+        // string; an out-of-bounds index will be caught in the loops below.)
+        if (MP_OBJ_IS_SMALL_INT(index)) {
+            i = MP_OBJ_SMALL_INT_VALUE(index);
+        } else if (!mp_obj_get_int_maybe(index, &i)) {
+            nlr_raise(mp_obj_new_exception_msg_varg(&mp_type_TypeError, "%s indices must be integers, not %s", qstr_str(type->name), mp_obj_get_type_str(index)));
+        }
+        if (i < 0)
+        {
+            // Negative indexing is performed by counting from the end of the string.
+            for (s = top - 1; i; --s) {
+                if (s < (const char *)self_data) {
+                    nlr_raise(mp_obj_new_exception_msg_varg(&mp_type_IndexError, "string index out of range"));
+                }
                 if (!UTF8_IS_CONT(*s)) {
-                    --index_val;
+                    ++i;
+                }
+            }
+            ++s;
+        } else {
+            // Positive indexing, correspondingly, counts from the start of the string.
+            // It's assumed that negative indexing will generally be used with small
+            // absolute values (eg str[-1], not str[-1000000]), which means it'll be
+            // more efficient this way.
+            for (s = (const char *)self_data; i; ++s) {
+                if (s >= top) {
+                    nlr_raise(mp_obj_new_exception_msg_varg(&mp_type_IndexError, "string index out of range"));
+                }
+                if (!UTF8_IS_CONT(*s)) {
+                    --i;
                 }
             }
             // Skip continuation bytes after the last lead byte
             while (UTF8_IS_CONT(*s)) {
                 ++s;
             }
-            int len = 1;
-            if (UTF8_IS_NONASCII(*s)) {
-                // Count the number of 1 bits (after the first)
-                for (char mask = 0x40; *s & mask; mask >>= 1) {
-                    ++len;
-                }
-            }
-            return mp_obj_new_str(s, len, true); // This will create a one-character string
         }
+        int len = 1;
+        if (UTF8_IS_NONASCII(*s)) {
+            // Count the number of 1 bits (after the first)
+            for (char mask = 0x40; *s & mask; mask >>= 1) {
+                ++len;
+            }
+        }
+        return mp_obj_new_str(s, len, true); // This will create a one-character string
     } else {
         return MP_OBJ_NULL; // op not supported
     }
diff --git a/tests/basics/unicode.py b/tests/basics/unicode.py
index ee66679b8d..777525e06f 100644
--- a/tests/basics/unicode.py
+++ b/tests/basics/unicode.py
@@ -5,5 +5,5 @@ for i in range(len(s)):
 # Test all three forms of Unicode escape, and
 # all blocks of UTF-8 byte patterns
 s = "a\xA9\xFF\u0123\u0800\uFFEE\U0001F44C"
-for i in range(len(s)):
+for i in range(-len(s), len(s)):
     print("s[%d]: %s   %X"%(i, s[i], ord(s[i])))