views:

276

answers:

5

I've come up with several manual ways of doing this, but i keep wondering if there is something built-in .NET that does this.

Basically, i want to reverse the bit order in a byte, so that the least significant bit becomes the most significant bit.

For example: 1001 1101 = 9D would become 1011 1001 = B9

On of the ways to do this is to use bitwise operations if following this pseudo code:

for (i = 0;i'8;i++)
{
  Y>>1
  x= byte & 1
  byte >>1
  y = x|y;
}

I wonder if there is a function somewhere that will allow me to do all this in one single line. Also, do you know the term for such an operation, i'm sure there is one, but i can't remember it at the moment.

Thanks

+12  A: 

No, there isn't anything in the BCL for this.

But, assuming you want something fast:

  • Since there are only 8 bits, it pays to unroll the loop (use 4 statements instead of the for-loop).

  • For an even faster solution, create a 256 entry lookup table.

And you can of course wrap both methods in a function so that the usage only takes 1 statement.

I found a page for this problem.

Henk Holterman
A lookup table solution is quite common, for problem like this. They are terrific for operations on small primitive types (like byte or short), but they don't scale up to larger types like int or long.
LBushkin
@LBushkin: Although you can't say they scale in the common case, in this case you can use the look-up table repeatedly to flip the individual bytes of a larger type.
Matti Virkkunen
I added an answer with some performance tests if you're interested ;)
digEmAll
Thanks for the advice, hadn't really considered performance. +1
Lily
+1  A: 

You can find bit twiddling algorithms in the fxtbook. Chapter 1.14 gives these bit swapping algorithms:

    static uint bitSwap1(uint x) {
        uint m = 0x55555555;
        return ((x & m) << 1) | ((x & (~m)) >> 1);
    }
    static uint bitSwap2(uint x) {
        uint m = 0x33333333;
        return ((x & m) << 2) | ((x & (~m)) >> 2);
    }
    static uint bitSwap4(uint x) {
        uint m = 0x0f0f0f0f;
        return ((x & m) << 4) | ((x & (~m)) >> 4);
    }

Which makes your byte value bit reversal:

    public static byte swapBits(byte value) {
        return (byte)(bitSwap4(bitSwap2(bitSwap1(value))));
    }

The x86 JIT compiler doesn't do a great job optimizing this code. If speed matters then you could use it to initialize a byte[] to make it a fast lookup instead.

Hans Passant
+1  A: 

Using @Chads link

byte b; 
b = 0x9D;
b = (byte)((b * 0x0202020202 & 0x010884422010) % 1023); 

Edit: Forgot the cast

Conrad Frix
A: 

Please see this comprehensive bit-twiddling hacks, namely you want 'Reverse the bits in a byte with 3 operations (64-bit multiply and modulus division)'

int lVal = 0x9D;
int lNewVal = (int)((((ulong)lVal * 0x0202020202UL) & 0x010884422010UL) % 1023);
System.Diagnostics.Debug.WriteLine(string.Format("{0:X2}", lNewVal));

When you run this you will find that the value gets reversed to 0xB9.

tommieb75
+10  A: 

I decided to do some performance tests about reversing methods.

Using Chad's link I wrote the following methods:

public static byte[] BitReverseTable =
{
    0x00, 0x80, 0x40, 0xc0, 0x20, 0xa0, 0x60, 0xe0,
    0x10, 0x90, 0x50, 0xd0, 0x30, 0xb0, 0x70, 0xf0,
    0x08, 0x88, 0x48, 0xc8, 0x28, 0xa8, 0x68, 0xe8,
    0x18, 0x98, 0x58, 0xd8, 0x38, 0xb8, 0x78, 0xf8,
    0x04, 0x84, 0x44, 0xc4, 0x24, 0xa4, 0x64, 0xe4,
    0x14, 0x94, 0x54, 0xd4, 0x34, 0xb4, 0x74, 0xf4,
    0x0c, 0x8c, 0x4c, 0xcc, 0x2c, 0xac, 0x6c, 0xec,
    0x1c, 0x9c, 0x5c, 0xdc, 0x3c, 0xbc, 0x7c, 0xfc,
    0x02, 0x82, 0x42, 0xc2, 0x22, 0xa2, 0x62, 0xe2,
    0x12, 0x92, 0x52, 0xd2, 0x32, 0xb2, 0x72, 0xf2,
    0x0a, 0x8a, 0x4a, 0xca, 0x2a, 0xaa, 0x6a, 0xea,
    0x1a, 0x9a, 0x5a, 0xda, 0x3a, 0xba, 0x7a, 0xfa,
    0x06, 0x86, 0x46, 0xc6, 0x26, 0xa6, 0x66, 0xe6,
    0x16, 0x96, 0x56, 0xd6, 0x36, 0xb6, 0x76, 0xf6,
    0x0e, 0x8e, 0x4e, 0xce, 0x2e, 0xae, 0x6e, 0xee,
    0x1e, 0x9e, 0x5e, 0xde, 0x3e, 0xbe, 0x7e, 0xfe,
    0x01, 0x81, 0x41, 0xc1, 0x21, 0xa1, 0x61, 0xe1,
    0x11, 0x91, 0x51, 0xd1, 0x31, 0xb1, 0x71, 0xf1,
    0x09, 0x89, 0x49, 0xc9, 0x29, 0xa9, 0x69, 0xe9,
    0x19, 0x99, 0x59, 0xd9, 0x39, 0xb9, 0x79, 0xf9,
    0x05, 0x85, 0x45, 0xc5, 0x25, 0xa5, 0x65, 0xe5,
    0x15, 0x95, 0x55, 0xd5, 0x35, 0xb5, 0x75, 0xf5,
    0x0d, 0x8d, 0x4d, 0xcd, 0x2d, 0xad, 0x6d, 0xed,
    0x1d, 0x9d, 0x5d, 0xdd, 0x3d, 0xbd, 0x7d, 0xfd,
    0x03, 0x83, 0x43, 0xc3, 0x23, 0xa3, 0x63, 0xe3,
    0x13, 0x93, 0x53, 0xd3, 0x33, 0xb3, 0x73, 0xf3,
    0x0b, 0x8b, 0x4b, 0xcb, 0x2b, 0xab, 0x6b, 0xeb,
    0x1b, 0x9b, 0x5b, 0xdb, 0x3b, 0xbb, 0x7b, 0xfb,
    0x07, 0x87, 0x47, 0xc7, 0x27, 0xa7, 0x67, 0xe7,
    0x17, 0x97, 0x57, 0xd7, 0x37, 0xb7, 0x77, 0xf7,
    0x0f, 0x8f, 0x4f, 0xcf, 0x2f, 0xaf, 0x6f, 0xef,
    0x1f, 0x9f, 0x5f, 0xdf, 0x3f, 0xbf, 0x7f, 0xff
};
public static byte ReverseWithLookupTable(byte toReverse)
{
    return BitReverseTable[toReverse];
}
public static byte ReverseBitsWith4Operations(byte b)
{
    return (byte)(((b * 0x80200802ul) & 0x0884422110ul) * 0x0101010101ul >> 32);
}
public static byte ReverseBitsWith3Operations(byte b)
{
    return (byte)((b * 0x0202020202ul & 0x010884422010ul) % 1023);
}
public static byte ReverseBitsWith7Operations(byte b)
{
    return (byte)(((b * 0x0802u & 0x22110u) | (b * 0x8020u & 0x88440u)) * 0x10101u >> 16);
}
public static byte ReverseBitsWithLoop(byte v)
{
    byte r = v; // r will be reversed bits of v; first get LSB of v
    int s = 7; // extra shift needed at end
    for (v >>= 1; v != 0; v >>= 1)
    {
        r <<= 1;
        r |= (byte)(v & 1);
        s--;
    }
    r <<= s; // shift when v's highest bits are zero
    return r;
}
public static byte ReverseWithUnrolledLoop(byte b)
{
    byte r = b;
    b >>= 1;
    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    r <<= 1;
    r |= (byte)(b & 1);
    b >>= 1;

    return r;
}

Then I tested it, and here's the results:

Test features:

  • 100000000 random bytes to reverse
  • OS: Windows 7 x64
  • CPU: AMD Phenom II 955 (4-core @ 3.2 GHz)
  • RAM: 4GB
  • IDE: Visual Studio 2010

Target framework 3.5

-----------------------------------------------------
|    Method     | Ticks(x64 mode) | Ticks(x86 mode) |
-----------------------------------------------------
| Loop          |   4861859       |   4079554       |
| Unrolled Loop |   3241781       |   2948026       |
| Look-up table |   894809        |   312410        |
| 3-Operations  |   2068072       |   6757008       |
| 4-Operations  |   893924        |   1972576       |
| 7-Operations  |   1219189       |   303499        |
-----------------------------------------------------

Target framework 4

-----------------------------------------------------
|    Method     | Ticks(x64 mode) | Ticks(x86 mode) |
-----------------------------------------------------
| Loop          |   4682654       |   4147036       |
| Unrolled Loop |   3154920       |   2851307       |
| Look-up table |   602686        |   313940        |
| 3-Operations  |   2067509       |   6661542       |
| 4-Operations  |   893406        |   2018334       |
| 7-Operations  |   1193200       |   991792        |
-----------------------------------------------------

So, look-up table method is not always the fastest :)

That can be reasonable, because memory access is slower than CPU registers access, so if some method is compiled and optimized enough to avoid mem access (and to do few operations) it is faster. (Anyway, the gap is extremely reduced by CPU mem caching)

It's also interesting to see the different behaviours in case of x64 or x86 mode, and how 3.5 and 4.0 frameworks performs distinct optimizations.

digEmAll
wow, very interesting. While i was hoping to find a method to do it for me, i think i will use your 7-Operations for the performance gain. I've only been doing this for 2 years and don't tend to think about performance enough. Now's a good time to start maybe :)
Lily
@digEmAll - What kind of CPU and how fast?
dbasnett
@dbasnett: Updated ;)
digEmAll
You must be including something other than the actual reverse bit code. On my 2.5GHZ the unrolled loop 100,000,000 times is < 300,000 ticks W7, .Net 4.0.
dbasnett
@dbasnett: my purpose was not show the ticks/time spent for solving the problem, but rather make a comparison amongst the different algorithms. Anyway, to test it I created a 100K random entries array and then read it in a loop for each reversing method.
digEmAll