69093cf2d6
* Optimize `TryAllocateRegWithtoutSpill` a bit * Add a fast path for when all registers are live. * Do not query `GetOverlapPosition` if the register is already in use (i.e: free position is 0). * Do not allocate child split list if not parent * Turn `LiveRange` into a reference struct `LiveRange` is now a reference wrapping struct like `Operand` and `Operation`. It has also been changed into a singly linked-list. In micro-benchmarks traversing the linked-list was faster than binary search on `List<T>`. Even for quite large input sizes (e.g: 1,000,000), surprisingly. Could be because the code gen for traversing the linked-list is much much cleaner and there is no virtual dispatch happening when checking if intervals overlaps. * Turn `LiveInterval` into an iterator The LSRA allocates in forward order and never inspect previous `LiveInterval` once they are expired. Something similar can be done for the `LiveRange`s within the `LiveInterval`s themselves. The `LiveInterval` is turned into a iterator which expires `LiveRange` within it. The iterator is moved forward along with interval walking code, i.e: AllocateInterval(context, interval, cIndex). * Remove `LinearScanAllocator.Sources` Local methods are less susceptible to do allocations than lambdas. * Optimize `GetOverlapPosition(interval)` a bit Time complexity should be in O(n+m) instead of O(nm) now. * Optimize `NumberLocals` a bit Use the same idea as in `HybridAllocator` to store the visited state in the MSB of the Operand's value instead of using a `HashSet<T>`. * Optimize `InsertSplitCopies` a bit Avoid allocating a redundant `CopyResolver`. * Optimize `InsertSplitCopiesAtEdges` a bit Avoid redundant allocations of `CopyResolver`. * Use stack allocation for `freePositions` Avoid redundant computations. * Add `UseList` Replace `SortedIntegerList` with an even more specialized data structure. It allocates memory on the arena allocators and does not require copying use positions when splitting it. * Turn `LiveInterval` into a reference struct `LiveInterval` is now a reference wrapping struct like `Operand` and `Operation`. The rationale behind turning this in a reference wrapping struct is because a `LiveInterval` is associated with each local variable, and these intervals may themselves be split further. I've seen translations having up to 8000 local variables. To make the `LiveInterval` unmanaged, a new data structure called `LiveIntervalList` was added to store child splits. This differs from `SortedList<,>` because it can contain intervals with the same start position. Really wished we got some more of C++ template in C#. :^( * Optimize `GetChildSplit` a bit No need to inspect the remaining ranges if we've reached a range which starts after position, since the split list is ordered. * Optimize `CopyResolver` a bit Lazily allocate the fill, spill and parallel copy structures since most of the time only one of them is needed. * Optimize `BitMap.Enumerator` a bit Marking `MoveNext` as `AggressiveInlining` allows RyuJIT to promote the `Enumerator` struct into registers completely, reducing load/store code a lot since it does not have to store the struct on the stack for ABI purposes. * Use stack allocation for `use/blockedPositions` * Optimize `AllocateWithSpill` a bit * Address feedback * Make `LiveInterval.AddRange(,)` more conservative Produces no diff against master, but just for good measure.
222 lines
No EOL
5.2 KiB
C#
222 lines
No EOL
5.2 KiB
C#
using System;
|
|
using System.Collections;
|
|
using System.Collections.Generic;
|
|
using System.Numerics;
|
|
using System.Runtime.CompilerServices;
|
|
|
|
namespace ARMeilleure.Common
|
|
{
|
|
unsafe class BitMap : IEnumerable<int>, IDisposable
|
|
{
|
|
private const int IntSize = 64;
|
|
private const int IntMask = IntSize - 1;
|
|
|
|
private int _count;
|
|
private long* _masks;
|
|
private readonly Allocator _allocator;
|
|
|
|
public BitMap(Allocator allocator)
|
|
{
|
|
_allocator = allocator;
|
|
}
|
|
|
|
public BitMap(Allocator allocator, int capacity) : this(allocator)
|
|
{
|
|
EnsureCapacity(capacity);
|
|
}
|
|
|
|
public bool Set(int bit)
|
|
{
|
|
EnsureCapacity(bit + 1);
|
|
|
|
int wordIndex = bit / IntSize;
|
|
int wordBit = bit & IntMask;
|
|
|
|
long wordMask = 1L << wordBit;
|
|
|
|
if ((_masks[wordIndex] & wordMask) != 0)
|
|
{
|
|
return false;
|
|
}
|
|
|
|
_masks[wordIndex] |= wordMask;
|
|
|
|
return true;
|
|
}
|
|
|
|
public void Clear(int bit)
|
|
{
|
|
EnsureCapacity(bit + 1);
|
|
|
|
int wordIndex = bit / IntSize;
|
|
int wordBit = bit & IntMask;
|
|
|
|
long wordMask = 1L << wordBit;
|
|
|
|
_masks[wordIndex] &= ~wordMask;
|
|
}
|
|
|
|
public bool IsSet(int bit)
|
|
{
|
|
EnsureCapacity(bit + 1);
|
|
|
|
int wordIndex = bit / IntSize;
|
|
int wordBit = bit & IntMask;
|
|
|
|
return (_masks[wordIndex] & (1L << wordBit)) != 0;
|
|
}
|
|
|
|
public int FindFirstUnset()
|
|
{
|
|
for (int index = 0; index < _count; index++)
|
|
{
|
|
long mask = _masks[index];
|
|
|
|
if (mask != -1L)
|
|
{
|
|
return BitOperations.TrailingZeroCount(~mask) + index * IntSize;
|
|
}
|
|
}
|
|
|
|
return _count * IntSize;
|
|
}
|
|
|
|
public bool Set(BitMap map)
|
|
{
|
|
EnsureCapacity(map._count * IntSize);
|
|
|
|
bool modified = false;
|
|
|
|
for (int index = 0; index < _count; index++)
|
|
{
|
|
long newValue = _masks[index] | map._masks[index];
|
|
|
|
if (_masks[index] != newValue)
|
|
{
|
|
_masks[index] = newValue;
|
|
|
|
modified = true;
|
|
}
|
|
}
|
|
|
|
return modified;
|
|
}
|
|
|
|
public bool Clear(BitMap map)
|
|
{
|
|
EnsureCapacity(map._count * IntSize);
|
|
|
|
bool modified = false;
|
|
|
|
for (int index = 0; index < _count; index++)
|
|
{
|
|
long newValue = _masks[index] & ~map._masks[index];
|
|
|
|
if (_masks[index] != newValue)
|
|
{
|
|
_masks[index] = newValue;
|
|
|
|
modified = true;
|
|
}
|
|
}
|
|
|
|
return modified;
|
|
}
|
|
|
|
private void EnsureCapacity(int size)
|
|
{
|
|
int count = (size + IntMask) / IntSize;
|
|
|
|
if (count > _count)
|
|
{
|
|
var oldMask = _masks;
|
|
var oldSpan = new Span<long>(_masks, _count);
|
|
|
|
_masks = _allocator.Allocate<long>((uint)count);
|
|
_count = count;
|
|
|
|
var newSpan = new Span<long>(_masks, _count);
|
|
|
|
oldSpan.CopyTo(newSpan);
|
|
newSpan.Slice(oldSpan.Length).Clear();
|
|
|
|
_allocator.Free(oldMask);
|
|
}
|
|
}
|
|
|
|
public void Dispose()
|
|
{
|
|
if (_masks != null)
|
|
{
|
|
_allocator.Free(_masks);
|
|
|
|
_masks = null;
|
|
}
|
|
}
|
|
|
|
IEnumerator IEnumerable.GetEnumerator()
|
|
{
|
|
return GetEnumerator();
|
|
}
|
|
|
|
IEnumerator<int> IEnumerable<int>.GetEnumerator()
|
|
{
|
|
return GetEnumerator();
|
|
}
|
|
|
|
public Enumerator GetEnumerator()
|
|
{
|
|
return new Enumerator(this);
|
|
}
|
|
|
|
public struct Enumerator : IEnumerator<int>
|
|
{
|
|
private long _index;
|
|
private long _mask;
|
|
private int _bit;
|
|
private readonly BitMap _map;
|
|
|
|
public int Current => (int)_index * IntSize + _bit;
|
|
object IEnumerator.Current => Current;
|
|
|
|
public Enumerator(BitMap map)
|
|
{
|
|
_index = -1;
|
|
_mask = 0;
|
|
_bit = 0;
|
|
_map = map;
|
|
}
|
|
|
|
[MethodImpl(MethodImplOptions.AggressiveInlining)]
|
|
public bool MoveNext()
|
|
{
|
|
if (_mask != 0)
|
|
{
|
|
_mask &= ~(1L << _bit);
|
|
}
|
|
|
|
// Manually hoist these loads, because RyuJIT does not.
|
|
long count = (uint)_map._count;
|
|
long* masks = _map._masks;
|
|
|
|
while (_mask == 0)
|
|
{
|
|
if (++_index >= count)
|
|
{
|
|
return false;
|
|
}
|
|
|
|
_mask = masks[_index];
|
|
}
|
|
|
|
_bit = BitOperations.TrailingZeroCount(_mask);
|
|
|
|
return true;
|
|
}
|
|
|
|
public void Reset() { }
|
|
|
|
public void Dispose() { }
|
|
}
|
|
}
|
|
} |