Fix various mistakes in ec_GFp_nistp_recode_scalar_bits comment.

Change-Id: I9b94e2da1bdf83a51b3dc219c154c5706e493e85
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/36244
Reviewed-by: Adam Langley <agl@google.com>

crypto/fipsmodule/ec/util.c

diff --git a/crypto/fipsmodule/ec/util.c b/crypto/fipsmodule/ec/util.c
index 7303a15..8951203 100644
--- a/crypto/fipsmodule/ec/util.c
+++ b/crypto/fipsmodule/ec/util.c
@@ -43,13 +43,13 @@
 //     of a nonnegative integer (b_k in {0, 1}), rewrite it in digits 0, 1, -1
 //     by using bit-wise subtraction as follows:
 //
-//        b_k b_(k-1)  ...  b_2  b_1  b_0
-//      -     b_k      ...  b_3  b_2  b_1  b_0
-//       -------------------------------------
-//        s_k b_(k-1)  ...  s_3  s_2  s_1  s_0
+//        b_k     b_(k-1)  ...  b_2  b_1  b_0
+//      -         b_k      ...  b_3  b_2  b_1  b_0
+//       -----------------------------------------
+//        s_(k+1) s_k      ...  s_3  s_2  s_1  s_0
 //
 //     A left-shift followed by subtraction of the original value yields a new
-//     representation of the same value, using signed bits s_i = b_(i+1) - b_i.
+//     representation of the same value, using signed bits s_i = b_(i-1) - b_i.
 //     This representation from Booth's paper has since appeared in the
 //     literature under a variety of different names including "reversed binary
 //     form", "alternating greedy expansion", "mutual opposite form", and
@@ -73,7 +73,7 @@
 // (1961), pp. 67-91), in a radix-2^5 setting.  That is, we always combine five
 // signed bits into a signed digit:
 //
-//       s_(4j + 4) s_(4j + 3) s_(4j + 2) s_(4j + 1) s_(4j)
+//       s_(5j + 4) s_(5j + 3) s_(5j + 2) s_(5j + 1) s_(5j)
 //
 // The sign-alternating property implies that the resulting digit values are
 // integers from -16 to 16.
@@ -81,14 +81,14 @@
 // Of course, we don't actually need to compute the signed digits s_i as an
 // intermediate step (that's just a nice way to see how this scheme relates
 // to the wNAF): a direct computation obtains the recoded digit from the
-// six bits b_(4j + 4) ... b_(4j - 1).
+// six bits b_(5j + 4) ... b_(5j - 1).
 //
-// This function takes those five bits as an integer (0 .. 63), writing the
+// This function takes those six bits as an integer (0 .. 63), writing the
 // recoded digit to *sign (0 for positive, 1 for negative) and *digit (absolute
-// value, in the range 0 .. 8).  Note that this integer essentially provides the
-// input bits "shifted to the left" by one position: for example, the input to
-// compute the least significant recoded digit, given that there's no bit b_-1,
-// has to be b_4 b_3 b_2 b_1 b_0 0.
+// value, in the range 0 .. 16).  Note that this integer essentially provides
+// the input bits "shifted to the left" by one position: for example, the input
+// to compute the least significant recoded digit, given that there's no bit
+// b_-1, has to be b_4 b_3 b_2 b_1 b_0 0.
 void ec_GFp_nistp_recode_scalar_bits(uint8_t *sign, uint8_t *digit,
                                      uint8_t in) {
   uint8_t s, d;