diff --git a/kaitai/kaitaistream.cpp b/kaitai/kaitaistream.cpp
index 81d0d48..7cb1743 100644
--- a/kaitai/kaitaistream.cpp
+++ b/kaitai/kaitaistream.cpp
@@ -432,41 +432,108 @@ std::string kaitai::kstream::bytes_terminate(std::string src, char term, bool in
 // ========================================================================
 
 std::string kaitai::kstream::process_xor_one(std::string data, uint8_t key) {
+    if (key == 0)
+        return data;
+
     size_t len = data.length();
     std::string result(len, ' ');
 
-    for (size_t i = 0; i < len; i++)
+    for (size_t i = 0; i < len; i++) {
         result[i] = data[i] ^ key;
+    }
 
     return result;
 }
 
 std::string kaitai::kstream::process_xor_many(std::string data, std::string key) {
     size_t len = data.length();
-    size_t kl = key.length();
+    if (len == 1)
+        return process_xor_one(data, key[0]);
+
     std::string result(len, ' ');
 
-    size_t ki = 0;
+    size_t k = 0;
+    size_t keylen = key.length();
     for (size_t i = 0; i < len; i++) {
-        result[i] = data[i] ^ key[ki];
-        ki++;
-        if (ki >= kl)
-            ki = 0;
+        result[i] = data[i] ^ key[k];
+        k++;
+        if (k == keylen)
+            k = 0;
     }
 
     return result;
 }
 
-std::string kaitai::kstream::process_rotate_left(std::string data, int amount) {
+uint8_t precomputedSingleRotations[8][256];
+
+// NOTE: code based on StackOverflow answer at https://stackoverflow.com/a/34321324/2375119
+computeSingleRotations {
+    for (int amount = 1; amount < 8; amount++) {
+        int anti_amount = 8 - amount;
+        for (uint8_t i = 0; i < 256; i++) {
+            precomputedSingleRotations[amount][i] = (uint8_t)((i << amount) | (i >> anti_amount));
+        }
+    }
+}
+
+std::string kaitai::kstream::process_rotate_left(std::string data, int amount, int groupSize = 1) {
+    if (groupSize < 1)
+        throw std::runtime_error("process_rotate_left: groupSize must be at least 1");
+
+    amount = mod(amount, groupSize * 8);
+    if (amount == 0)
+        return data;
+
+    int amount_bytes = amount / 8;
     size_t len = data.length();
     std::string result(len, ' ');
 
-    for (size_t i = 0; i < len; i++) {
-        uint8_t bits = data[i];
-        result[i] = (bits << amount) | (bits >> (8 - amount));
+    if (groupSize == 1) {
+        uint8_t *translate = &precomputedSingleRotations[amount][0];
+
+        for (size_t i = 0; i < len; i++) {
+            result[i] = translate[data[i]];
+        }
+
+        return result;
     }
 
-    return result;
+    if (len % groupSize != 0)
+        throw std::runtime_error("process_rotate_left: data length must be a multiple of group size");
+
+    if (amount % 8 == 0) {
+        size_t indices[groupSize];
+        for (size_t i = 0; i < groupSize; i++) {
+            indices[i] = (size_t)((i + amount_bytes) % groupSize);
+        }
+
+        for (size_t i = 0; i < len; i += groupSize) {
+            for (size_t k = 0; k < groupSize; k++) {
+                result[i+k] = data[i + indices[k]];
+            }
+        }
+
+        return result;
+    }
+
+    {
+        int amount1 = amount % 8;
+        int amount2 = 8 - amount1;
+        size_t indices1[groupSize];
+        size_t indices2[groupSize];
+        for (size_t i = 0; i < groupSize; i++) {
+            indices1[i] = (size_t)((i +     amount_bytes) % groupSize);
+            indices2[i] = (size_t)((i + 1 + amount_bytes) % groupSize);
+        }
+
+        for (size_t i = 0; i < len; i += groupSize) {
+            for (size_t k = 0; k < groupSize; k++) {
+                result[i+k] = (uint8_t)((data[i + indices1[k]] << amount1) | (data[i + indices2[k]] >> amount2));
+            }
+        }
+
+        return result;
+    }
 }
 
 #ifdef KS_ZLIB
diff --git a/kaitai/kaitaistream.h b/kaitai/kaitaistream.h
index 9592771..9cceaf7 100644
--- a/kaitai/kaitaistream.h
+++ b/kaitai/kaitaistream.h
@@ -172,6 +172,8 @@ class kstream {
     /**
      * Performs a XOR processing with given data, XORing every byte of input with a single
      * given value.
+     * WARNING: May return same byte array if key is zero.
+     * 
      * @param data data to process
      * @param key value to XOR with
      * @return processed data
@@ -182,6 +184,8 @@ class kstream {
      * Performs a XOR processing with given data, XORing every byte of input with a key
      * array, repeating key array many times, if necessary (i.e. if data array is longer
      * than key array).
+     * WARNING: May return same byte array if key is zero.
+     * 
      * @param data data to process
      * @param key array of bytes to XOR with
      * @return processed data
@@ -192,11 +196,14 @@ class kstream {
      * Performs a circular left rotation shift for a given buffer by a given amount of bits,
      * using groups of 1 bytes each time. Right circular rotation should be performed
      * using this procedure with corrected amount.
+     * WARNING: May return same byte array if amount is zero (modulo-wise).
+     * 
      * @param data source data to process
      * @param amount number of bits to shift by
+     * @param groupSize number of bytes that make a group
      * @return copy of source array with requested shift applied
      */
-    static std::string process_rotate_left(std::string data, int amount);
+    static std::string process_rotate_left(std::string data, int amount, int groupSize = 1);
 
 #ifdef KS_ZLIB
     /**