Add some fixes

Author: zanmato1984

cpp/src/arrow/compute/util.cc

@@ -36,7 +36,11 @@ inline uint64_t SafeLoadUpTo8Bytes(const uint8_t* bytes, int num_bytes) {
   } else {
     uint64_t word = 0;
     for (int i = 0; i < num_bytes; ++i) {
+#if ARROW_LITTLE_ENDIAN
       word |= static_cast<uint64_t>(bytes[i]) << (8 * i);
+#else
+      word |= static_cast<uint64_t>(bytes[i]) << (8 * (num_bytes - 1 - i));
+#endif
     }
     return word;
   }
@@ -47,16 +51,13 @@ inline void SafeStoreUpTo8Bytes(uint8_t* bytes, int num_bytes, uint64_t value) {
   if (num_bytes == 8) {
     util::SafeStore(reinterpret_cast<uint64_t*>(bytes), value);
   } else {
-#if ARROW_LITTLE_ENDIAN
     for (int i = 0; i < num_bytes; ++i) {
+#if ARROW_LITTLE_ENDIAN
       bytes[i] = static_cast<uint8_t>(value >> (8 * i));
-    }
 #else
-    // Big-endian: most significant byte first
-    for (int i = 0; i < num_bytes; ++i) {
       bytes[i] = static_cast<uint8_t>(value >> (8 * (num_bytes - 1 - i)));
-    }
 #endif
+    }
   }
 }
 
@@ -102,6 +103,9 @@ void bits_to_indexes_internal(int64_t hardware_flags, const int num_bits,
     *num_indexes = 0;
     for (int i = 0; i < num_bits / unroll; ++i) {
       uint64_t word = util::SafeLoad(&reinterpret_cast<const uint64_t*>(bits)[i]);
+#if !ARROW_LITTLE_ENDIAN
+      word = ::arrow::bit_util::ByteSwap(word);
+#endif
       if (bit_to_search == 0) {
         word = ~word;
       }
@@ -117,21 +121,9 @@ void bits_to_indexes_internal(int64_t hardware_flags, const int num_bits,
   // Optionally process the last partial word with masking out bits outside range
   if (tail) {
     const uint8_t* bits_tail = bits + (num_bits - tail) / 8;
-#if ARROW_LITTLE_ENDIAN
     uint64_t word = SafeLoadUpTo8Bytes(bits_tail, (tail + 7) / 8);
-#else
-    int tail_bytes = (tail + 7) / 8;
-    uint64_t word;
-    if (tail_bytes == 8) {
-      word = util::SafeLoad(reinterpret_cast<const uint64_t*>(bits_tail));
-    } else {
-      // For bit manipulation, always load into least significant bits
-      // to ensure compatibility with CountTrailingZeros on Big-endian systems
-      word = 0;
-      for (int i = 0; i < tail_bytes; ++i) {
-        word |= static_cast<uint64_t>(bits_tail[i]) << (8 * i);
-      }
-    }
+#if !ARROW_LITTLE_ENDIAN
+    word = ::arrow::bit_util::ByteSwap(word);
 #endif
     if (bit_to_search == 0) {
       word = ~word;
@@ -305,6 +297,9 @@ void bytes_to_bits(int64_t hardware_flags, const int num_bits, const uint8_t* by
   constexpr int unroll = 8;
   for (int i = num_processed / unroll; i < num_bits / unroll; ++i) {
     uint64_t bytes_next = util::SafeLoad(&reinterpret_cast<const uint64_t*>(bytes)[i]);
+#if !ARROW_LITTLE_ENDIAN
+    bytes_next = ::arrow::bit_util::ByteSwap(bytes_next);
+#endif
     bytes_next &= 0x0101010101010101ULL;
     bytes_next |= (bytes_next >> 7);  // Pairs of adjacent output bits in individual bytes
     bytes_next |= (bytes_next >> 14);  // 4 adjacent output bits in individual bytes
@@ -313,16 +308,9 @@ void bytes_to_bits(int64_t hardware_flags, const int num_bits, const uint8_t* by
   }
   int tail = num_bits % unroll;
   if (tail) {
-    uint64_t bytes_next;
-#if ARROW_LITTLE_ENDIAN
-    bytes_next = SafeLoadUpTo8Bytes(bytes + num_bits - tail, tail);
-#else
-    // On Big-endian systems, for bytes_to_bits, load all tail bytes in little-endian
-    // order to ensure compatibility with subsequent bit operations
-    bytes_next = 0;
-    for (int i = 0; i < tail; ++i) {
-      bytes_next |= static_cast<uint64_t>((bytes + num_bits - tail)[i]) << (8 * i);
-    }
+    uint64_t bytes_next = SafeLoadUpTo8Bytes(bytes + num_bits - tail, tail);
+#if !ARROW_LITTLE_ENDIAN
+    bytes_next = ::arrow::bit_util::ByteSwap(bytes_next);
 #endif
     bytes_next &= 0x0101010101010101ULL;
     bytes_next |= (bytes_next >> 7);  // Pairs of adjacent output bits in individual bytes