Skip to content

[WIP] Experimental kernels #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
michaelfeil wants to merge 3 commits into
base: master
Choose a base branch
from
Open

[WIP] Experimental kernels #1

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5a7c8db
add experimental kernel
michaelfeil Apr 4, 2024
fc3b718
reset to working state
michaelfeil Apr 5, 2024
0bd7337
update kernels
michaelfeil Apr 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 24 additions & 17 deletions kernels/ternary_mm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,10 +2,13 @@
# - DeltaBit https://github.com/FasterDecoding/BitDelta/tree/main
# - BitNet
# - https://triton-lang.org/main/getting-started/tutorials/03-matrix-multiplication.html & IBM FMS
# - https://github.com/ROCm/triton/blob/triton-mlir/python/tutorials/11-grouped-gemm.pyw


# COPYRIGHT 2024, Gradient.ai Inc. All Rights Reserved.
# CC-BY-NC-4.0


import torch
import triton
import triton.language as tl
Expand Down Expand Up @@ -59,11 +62,7 @@ def unpack(x: torch.Tensor, n_bits=4):

@triton.autotune(
configs=[
# triton.Config({'BLOCK_SIZE_M': 128, 'BLOCK_SIZE_N': 128, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 8}, num_stages=4, num_warps=8),
triton.Config({'BLOCK_SIZE_M': 128, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 64, 'GROUP_SIZE_M': 8}, num_stages=3,
num_warps=8),
triton.Config({'BLOCK_SIZE_M': 64, 'BLOCK_SIZE_N': 256, 'BLOCK_SIZE_K': 32, 'GROUP_SIZE_M': 8}, num_stages=4,
num_warps=4),

triton.Config({'BLOCK_SIZE_M': 128, 'BLOCK_SIZE_N': 128, 'BLOCK_SIZE_K': 32, 'GROUP_SIZE_M': 8}, num_stages=4,
num_warps=4),
triton.Config({'BLOCK_SIZE_M': 128, 'BLOCK_SIZE_N': 64, 'BLOCK_SIZE_K': 32, 'GROUP_SIZE_M': 8}, num_stages=4,
Expand Down Expand Up @@ -93,7 +92,7 @@ def _ternary_mm_kernel(
a_ptr, b_ptr, c_ptr,
# Matrix dimensions
M, N, K,
n_bits,
n_bits: tl.constexpr,
# The stride variables represent how much to increase the ptr by when moving by 1
# element in a particular dimension. E.g. `stride_am` is how much to increase `a_ptr`
# by to get the element one row down (A has M rows).
Expand All @@ -110,6 +109,9 @@ def _ternary_mm_kernel(
A has shape (M, K), float16
B has shape (K//n_bits, N), uint8, packed
C has shape (M, N),

NEEDS triton>3.0.0 to work / tl.join()/tl.interleave()
install torch, then add triton from source.
"""
# -----------------------------------------------------------
# Map program ids `pid` to the block of C it should compute.
Expand Down Expand Up @@ -140,7 +142,7 @@ def _ternary_mm_kernel(
offs_n = (pid_n * BLOCK_SIZE_N + tl.arange(0, BLOCK_SIZE_N)) % N
# offs_am = tl.max_contiguous(tl.multiple_of(offs_m, BLOCK_SIZE_M), BLOCK_SIZE_M)
offs_bn = tl.max_contiguous(tl.multiple_of(offs_n, BLOCK_SIZE_N), BLOCK_SIZE_N)
offs_k = tl.arange(0, BLOCK_SIZE_K)
offs_k = tl.arange(0, BLOCK_SIZE_K// n_bits)

# a_ptrs = a_ptr + (offs_am[:, None] * stride_am + offs_k[None, :] * stride_ak)
a_block_ptr = tl.make_block_ptr(base=a_ptr, shape=(M,K), strides=(stride_am, stride_ak),
Expand All @@ -149,9 +151,9 @@ def _ternary_mm_kernel(

# Adapted from GPTQ-Triton (https://github.com/fpgaminer/GPTQ-triton)
# b_ptrs is set up such that it repeats elements along the K axis n_bits times
b_ptrs = b_ptr + ((offs_k[:, None] // n_bits) * stride_bk + offs_bn[None, :] * stride_bn)
b_ptrs = b_ptr + ((offs_k[:, None] ) * stride_bk + offs_bn[None, :] * stride_bn)
# shifter is used to extract each bit of each element in the int matrix
shifter = (offs_k % n_bits)[:, None] * 2 #
shifter = (tl.arange(0, BLOCK_SIZE_K) % n_bits)[:, None] * 2 #
# shifter = shifter
# -----------------------------------------------------------
# Iterate to compute a block of the C matrix.
Expand All @@ -171,9 +173,13 @@ def _ternary_mm_kernel(
a = tl.load(a_block_ptr, boundary_check=(0,1))
# a = tl.load(a_ptrs, mask=offs_k[None, :] < K - k * BLOCK_SIZE_K, other=0.0)
b = tl.load(b_ptrs, mask=offs_k[:, None] < K - k * BLOCK_SIZE_K, other=0.0)

# Convert B from int to a.dtype, for each bit in B, 0 becomes -1.0, 1 becomes 1.0
# b: (BLOCK_SIZE_K, BLOCK_SIZE_N)
b = b.T
b = tl.interleave(tl.interleave(b,b), tl.interleave(b,b)).T
# b = tl.interleave(b, b) # .reshape(b.shape[-1:] + [2 * b.shape[-1]])
# tl.interleave(b,b)
# b = b.reshape(BLOCK_SIZE_K, BLOCK_SIZE_N, n_bits)
b = (b >> shifter) & 0x3
# shift b to -1, 0, 1
b = b.to(a.dtype) - 1
Expand All @@ -184,7 +190,7 @@ def _ternary_mm_kernel(
# Advance the ptrs to the next K block.
# a_ptrs += BLOCK_SIZE_K * stride_ak
a_block_ptr = tl.advance(a_block_ptr, (0, BLOCK_SIZE_K))
b_ptrs += (BLOCK_SIZE_K // n_bits) * stride_bk
b_ptrs += (BLOCK_SIZE_K) * stride_bk
# You can fuse arbitrary activation functions here
# while the accumulator is still in FP32!
# if ACTIVATION == "leaky_relu":
Expand Down Expand Up @@ -240,9 +246,9 @@ def bitmat(a, b, n_bits=4, activation=""):
)
try:
c[0][0].item()
except RuntimeError:
except RuntimeError as ex:
raise RuntimeError(
"Illegal memory access, it means that the kernel failed most probably to OOM, try to reduce batch size or matrix size.")
"Illegal memory access, it means that the kernel failed most probably to OOM, try to reduce batch size or matrix size.") from ex

return c

Expand Down Expand Up @@ -275,7 +281,7 @@ def matmul_f16(A: torch.Tensor, B: torch.Tensor) -> torch.Tensor:

@torch.inference_mode()
def main():
N_BITS=16
N_BITS=4
# Example usage
M, N, K = 32, 16, 128
A = torch.rand((M,K), device='cuda', dtype=torch.float16) * 10
Expand All @@ -285,15 +291,16 @@ def assert_equal(A, B):
A = A.clone()
B = B.clone()
assert (B - unpack(pack(B, n_bits=N_BITS), n_bits=N_BITS) == 0).all()
assert torch.allclose(matmul_f32(A,B), bitmat(A, pack(B, n_bits=N_BITS), n_bits=N_BITS), atol=1e-3, rtol=1e-3)
bitmat(A, pack(B, n_bits=N_BITS), n_bits=N_BITS)
torch.testing.assert_close(matmul_f32(A,B), bitmat(A, pack(B, n_bits=N_BITS), n_bits=N_BITS), atol=1e-3, rtol=1e-3)

assert_equal(A, B)
print("Success for small tensors.")

@triton.testing.perf_report(
triton.testing.Benchmark(
x_names=['N'], # argument names to use as an x-axis for the plot
x_vals=[2**i for i in range(2, 13)], # different possible values for `x_name`
x_vals=[2**i for i in range(5, 10)], # different possible values for `x_name`
line_arg='provider', # argument name whose value corresponds to a different line in the plot
line_vals=[
'triton',
Expand All @@ -306,7 +313,7 @@ def assert_equal(A, B):
styles=[('blue', '-'), ('green', '-')], # ('green', '--')], # line styles
ylabel="ms", # label name for the y-axis
plot_name="add-performance", # name for the plot. Used also as a file name for saving the plot.
args={'M': 1024, "K":8192}, # 'M': 4096 # values for function arguments not in `x_names` and `y_name`
args={'M': 512, "K":128}, # 'M': 4096 # values for function arguments not in `x_names` and `y_name`
))
def benchmark(M, N, K, provider):
assert N % 4 == 0, "N must be a multiple of 4"
Expand Down