Move NEON version to a separate function and fix indentation

lib/compress/zstd_lazy.c

@@ -954,6 +954,29 @@ void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip) {
     ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 0 /* don't use cache */);
 }
 
+/* Returns the mask width of bits group of which will be set to 1. Given not all
+ * architectures have easy movemask instruction, this helps to iterate over
+ * groups of bits easier and faster.
+ */
+FORCE_INLINE_TEMPLATE U32
+ZSTD_row_matchMaskGroupWidth(const U32 rowEntries)
+{
+    assert((rowEntries == 16) || (rowEntries == 32) || rowEntries == 64);
+    assert(rowEntries <= ZSTD_ROW_HASH_MAX_ENTRIES);
+#if defined(ZSTD_ARCH_ARM_NEON)
+    if (rowEntries == 16) {
+        return 4;
+    }
+    if (rowEntries == 32) {
+        return 2;
+    }
+    if (rowEntries == 64) {
+        return 1;
+    }
+#endif
+    return 1;
+}
+
 #if defined(ZSTD_ARCH_X86_SSE2)
 FORCE_INLINE_TEMPLATE ZSTD_VecMask
 ZSTD_row_getSSEMask(int nbChunks, const BYTE* const src, const BYTE tag, const U32 head)
@@ -974,52 +997,11 @@ ZSTD_row_getSSEMask(int nbChunks, const BYTE* const src, const BYTE tag, const U
 }
 #endif
 
-/* Returns the mask width of bits group of which will be set to 1. Given not all
- * architectures have easy movemask instruction, this helps to iterate over
- * groups of bits easier and faster.
- */
-FORCE_INLINE_TEMPLATE U32
-ZSTD_row_matchMaskGroupWidth(const U32 rowEntries) {
-  assert((rowEntries == 16) || (rowEntries == 32) || rowEntries == 64);
-  assert(rowEntries <= ZSTD_ROW_HASH_MAX_ENTRIES);
-  (void)rowEntries;
 #if defined(ZSTD_ARCH_ARM_NEON)
-  if (rowEntries == 16) {
-    return 4;
-  }
-  if (rowEntries == 32) {
-    return 2;
-  }
-  if (rowEntries == 64) {
-    return 1;
-  }
-#endif
-  return 1;
-}
-
-/* Returns a ZSTD_VecMask (U64) that has the nth group (determined by
- * ZSTD_row_matchMaskGroupWidth) of bits set to 1 if the newly-computed "tag"
- * matches the hash at the nth position in a row of the tagTable.
- * Each row is a circular buffer beginning at the value of "headGrouped". So we
- * must rotate the "matches" bitfield to match up with the actual layout of the
- * entries within the hashTable */
 FORCE_INLINE_TEMPLATE ZSTD_VecMask
-ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 headGrouped, const U32 rowEntries)
+ZSTD_row_getNEONMask(const U32 rowEntries, const BYTE* const src, const BYTE tag, const U32 headGrouped)
 {
-    const BYTE* const src = tagRow + ZSTD_ROW_HASH_TAG_OFFSET;
     assert((rowEntries == 16) || (rowEntries == 32) || rowEntries == 64);
-    assert(rowEntries <= ZSTD_ROW_HASH_MAX_ENTRIES);
-    assert(ZSTD_row_matchMaskGroupWidth(rowEntries) * rowEntries <= sizeof(ZSTD_VecMask) * 8);
-
-#if defined(ZSTD_ARCH_X86_SSE2)
-
-    return ZSTD_row_getSSEMask(rowEntries / 16, src, tag, headGrouped);
-
-#else /* SW or NEON-LE */
-
-# if defined(ZSTD_ARCH_ARM_NEON)
-  /* This NEON path only works for little endian - otherwise use SWAR below */
-    if (MEM_isLittleEndian()) {
     if (rowEntries == 16) {
         /* vshrn_n_u16 shifts by 4 every u16 and narrows to 8 lower bits.
          * After that groups of 4 bits represent the equalMask. We lower
@@ -1061,6 +1043,33 @@ ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 headGr
         return ZSTD_rotateRight_U64(matches, headGrouped);
     }
 }
+#endif
+
+/* Returns a ZSTD_VecMask (U64) that has the nth group (determined by
+ * ZSTD_row_matchMaskGroupWidth) of bits set to 1 if the newly-computed "tag"
+ * matches the hash at the nth position in a row of the tagTable.
+ * Each row is a circular buffer beginning at the value of "headGrouped". So we
+ * must rotate the "matches" bitfield to match up with the actual layout of the
+ * entries within the hashTable */
+FORCE_INLINE_TEMPLATE ZSTD_VecMask
+ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 headGrouped, const U32 rowEntries)
+{
+    const BYTE* const src = tagRow + ZSTD_ROW_HASH_TAG_OFFSET;
+    assert((rowEntries == 16) || (rowEntries == 32) || rowEntries == 64);
+    assert(rowEntries <= ZSTD_ROW_HASH_MAX_ENTRIES);
+    assert(ZSTD_row_matchMaskGroupWidth(rowEntries) * rowEntries <= sizeof(ZSTD_VecMask) * 8);
+
+#if defined(ZSTD_ARCH_X86_SSE2)
+
+    return ZSTD_row_getSSEMask(rowEntries / 16, src, tag, headGrouped);
+
+#else /* SW or NEON-LE */
+
+# if defined(ZSTD_ARCH_ARM_NEON)
+  /* This NEON path only works for little endian - otherwise use SWAR below */
+    if (MEM_isLittleEndian()) {
+        return ZSTD_row_getNEONMask(rowEntries, src, tag, headGrouped);
+    }
 # endif /* ZSTD_ARCH_ARM_NEON */
     /* SWAR */
     {   const size_t chunkSize = sizeof(size_t);