CUTLASS 2.10 updates (#622)
Co-authored-by: Aniket Shivam <ashivam@nvidia.com>
This commit is contained in:
ANIKET SHIVAM
co-authored by
Aniket Shivam
Aniket Shivam
parent
beae168f90
commit
e773429f7e
+213
-34
@@ -99,9 +99,11 @@ parser.add_argument("-te", "--element_epilogue", default="float32", type=str,
|
||||
parser.add_argument("-ep", "--epilogue_functor", default="LinearCombination",
|
||||
type=str, choices=['LinearCombination', 'FastLinearCombinationClamp', 'LinearCombinationClamp'],
|
||||
help="This option describes the epilogue part of the kernel")
|
||||
parser.add_argument("-epv", "--epilogue_visitor", default=None,
|
||||
type=str, choices=['RowReduction', 'ColumnReduction', 'RowBroadcast', 'ColumnBroadcast'], help="epilogue visitor for more complex epilogues")
|
||||
# swizzling
|
||||
parser.add_argument("-sw", "--swizzling_functor", default="IdentitySwizzle1", type=str, choices=[
|
||||
"IdentitySwizzle1", "IdentitySwizzle2", "IdentitySwizzle4", "IdentitySwizzle8", "HorizontalSwizzle"],
|
||||
"IdentitySwizzle1", "IdentitySwizzle2", "IdentitySwizzle4", "IdentitySwizzle8", "HorizontalSwizzle", "BatchedIdentitySwizzle"],
|
||||
help="This option describes how thread blocks are scheduled on GPU")
|
||||
|
||||
# Argument
|
||||
@@ -113,17 +115,22 @@ parser.add_argument("-alpha", "--alpha", default=1.0, type=float,
|
||||
parser.add_argument("-beta", "--beta", default=0.0, type=float,
|
||||
help="Scaling factor of C")
|
||||
parser.add_argument("-gm", "--gemm_mode", default="Gemm", type=str,
|
||||
choices=["Gemm", "GemmSplitKParallel"],
|
||||
choices=["Gemm", "GemmSplitKParallel", "Batched", "Array"],
|
||||
help="GEMM mode. Gemm is used for non-splitK or serial-splitK. \
|
||||
GemmSplitKParallel is used for parallel splitK")
|
||||
parser.add_argument('-k', '--split_k_slices', default=1,
|
||||
type=int, help="Number of split-k partitions. (default 1)")
|
||||
parser.add_argument('-bias', '--bias', action='store_true', help="C is bias vector")
|
||||
parser.add_argument('-batch', '--batch', default=1, type=int, help="batch size for batched GEMM")
|
||||
|
||||
# Activation function
|
||||
parser.add_argument("-activ", "--activation_function", default="identity",
|
||||
choices=["identity", "relu", "leaky_relu", "tanh", "sigmoid", "silu", "hardswish", "gelu"], help="activation function")
|
||||
parser.add_argument("-activ_arg", "--activation_args", default=[], nargs="+", type=float,
|
||||
help="addition arguments for activation")
|
||||
parser.add_argument('--print_cuda', action="store_true",
|
||||
help="print the underlying CUDA kernel")
|
||||
|
||||
# parser.add_argument('-h', '--help', action="store_true",
|
||||
# help="print help information")
|
||||
|
||||
try:
|
||||
args = parser.parse_args()
|
||||
@@ -131,6 +138,9 @@ except:
|
||||
sys.exit(0)
|
||||
|
||||
pycutlass.get_memory_pool(init_pool_size=2**30, max_pool_size=2**32)
|
||||
pycutlass.compiler.nvcc()
|
||||
|
||||
np.random.seed(0)
|
||||
|
||||
element_a = getattr(cutlass, args.element_a)
|
||||
element_b = getattr(cutlass, args.element_b)
|
||||
@@ -146,7 +156,7 @@ math_inst = MathInstruction(
|
||||
|
||||
tile_description = TileDescription(
|
||||
args.threadblock_shape, args.stages, args.warp_count,
|
||||
math_inst, args.compute_capability, args.compute_capability
|
||||
math_inst
|
||||
)
|
||||
|
||||
layout_a = getattr(cutlass, args.layout_a)
|
||||
@@ -166,13 +176,83 @@ C = TensorDescription(
|
||||
)
|
||||
|
||||
element_epilogue = getattr(cutlass, args.element_epilogue)
|
||||
epilogue_functor = getattr(EpilogueFunctor, args.epilogue_functor)
|
||||
if (args.activation_function == "identity"
|
||||
or (args.gemm_mode == "GemmSplitKParallel" and args.split_k_slices > 1)):
|
||||
#
|
||||
epilogue_functor = getattr(pycutlass, args.epilogue_functor)(
|
||||
C.element, C.alignment, math_inst.element_accumulator, element_epilogue)
|
||||
else:
|
||||
epilogue_functor = getattr(pycutlass, "LinearCombinationGeneric")(
|
||||
getattr(pycutlass, args.activation_function)(element_epilogue),
|
||||
C.element, C.alignment, math_inst.element_accumulator, element_epilogue)
|
||||
|
||||
swizzling_functor = getattr(cutlass, args.swizzling_functor)
|
||||
|
||||
visitor = args.epilogue_visitor is not None
|
||||
|
||||
if args.epilogue_visitor == "ColumnReduction":
|
||||
class ColumnReduction_(EpilogueVisitTree):
|
||||
def __call__(
|
||||
self, accum: 'tensor', c: 'tensor',
|
||||
alpha: 'scalar', beta: 'scalar'):
|
||||
#
|
||||
D = alpha * accum + beta * c
|
||||
reduction = reduction_op(D, "column", "Add", args.threadblock_shape[0])
|
||||
return D, reduction
|
||||
epilogue_functor = ColumnReduction_(
|
||||
epilogue_functor, tile_description, math_inst.element_accumulator,
|
||||
C.alignment, element_epilogue, C.element)
|
||||
epilogue_functor.initialize()
|
||||
elif args.epilogue_visitor == "RowReduction":
|
||||
class RowReduction_(EpilogueVisitTree):
|
||||
def __call__(
|
||||
self, accum: 'tensor', c: 'tensor',
|
||||
alpha: 'scalar', beta: 'scalar'):
|
||||
#
|
||||
D = alpha * accum + tanh.numpy(beta * c)
|
||||
reduction = reduction_op(D, "row", "Add", args.threadblock_shape[1])
|
||||
return D, reduction
|
||||
epilogue_functor = RowReduction_(
|
||||
epilogue_functor, tile_description, math_inst.element_accumulator,
|
||||
C.alignment, element_epilogue, C.element)
|
||||
epilogue_functor.initialize()
|
||||
|
||||
elif args.epilogue_visitor == "RowBroadcast":
|
||||
class RowBroadcast_(EpilogueVisitTree):
|
||||
def __call__(
|
||||
self, accum: 'tensor', c: 'tensor',
|
||||
vector: 'row', alpha: 'scalar', beta: 'scalar'):
|
||||
#
|
||||
T = accum + vector
|
||||
scale_T = alpha * T
|
||||
Z = relu.numpy(scale_T + beta * c)
|
||||
return Z, T
|
||||
epilogue_functor = RowBroadcast_(
|
||||
epilogue_functor, tile_description, math_inst.element_accumulator,
|
||||
C.alignment, element_epilogue, C.element)
|
||||
epilogue_functor.initialize()
|
||||
elif args.epilogue_visitor == "ColumnBroadcast":
|
||||
class ColumnBroadcast_(EpilogueVisitTree):
|
||||
def __call__(
|
||||
self, accum: 'tensor', c: 'tensor',
|
||||
vector: 'column', alpha: 'scalar', beta: 'scalar'):
|
||||
#
|
||||
T = accum + vector
|
||||
scale_T = leaky_relu.numpy(alpha * T, 0.2)
|
||||
Z = scale_T + beta * c
|
||||
return Z, T
|
||||
epilogue_functor = ColumnBroadcast_(
|
||||
epilogue_functor, tile_description, math_inst.element_accumulator,
|
||||
C.alignment, element_epilogue, C.element)
|
||||
epilogue_functor.initialize()
|
||||
else:
|
||||
epilogue_functor = epilogue_functor
|
||||
|
||||
operation = GemmOperationUniversal(
|
||||
arch=args.compute_capability, tile_description=tile_description,
|
||||
A=A, B=B, C=C, element_epilogue=element_epilogue,
|
||||
epilogue_functor=epilogue_functor, swizzling_functor=swizzling_functor
|
||||
A=A, B=B, C=C,
|
||||
epilogue_functor=epilogue_functor, swizzling_functor=swizzling_functor,
|
||||
visitor=visitor
|
||||
)
|
||||
|
||||
if args.print_cuda:
|
||||
@@ -181,10 +261,19 @@ if args.print_cuda:
|
||||
operations = [operation, ]
|
||||
|
||||
if args.gemm_mode == "GemmSplitKParallel":
|
||||
if (args.activation_function == "identity"):
|
||||
epilogue_functor_reduction = getattr(pycutlass, args.epilogue_functor)(
|
||||
C.element, C.alignment, math_inst.element_accumulator, element_epilogue)
|
||||
else:
|
||||
epilogue_functor_reduction = getattr(pycutlass, "LinearCombinationGeneric")(
|
||||
getattr(pycutlass, args.activation_function)(element_epilogue),
|
||||
C.element, C.alignment, math_inst.element_accumulator, element_epilogue)
|
||||
|
||||
reduction_operation = ReductionOperation(
|
||||
shape=cutlass.MatrixCoord(4, 32 * C.alignment),
|
||||
C=C, element_accumulator=element_acc,
|
||||
element_compute=element_epilogue,
|
||||
element_compute=element_epilogue,
|
||||
epilogue_functor=epilogue_functor_reduction,
|
||||
count=C.alignment
|
||||
)
|
||||
operations.append(reduction_operation)
|
||||
@@ -196,47 +285,102 @@ pycutlass.compiler.add_module(operations)
|
||||
problem_size = cutlass.gemm.GemmCoord(
|
||||
args.problem_size[0], args.problem_size[1], args.problem_size[2])
|
||||
|
||||
tensor_a_size = args.batch * problem_size.m() * problem_size.k()
|
||||
if args.element_a != "int8":
|
||||
if args.element_a == "bfloat16":
|
||||
tensor_A = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.m()
|
||||
* problem_size.k(),))).astype(bfloat16)
|
||||
tensor_A = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_a_size,))
|
||||
).astype(bfloat16)
|
||||
else:
|
||||
tensor_A = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.m()
|
||||
* problem_size.k(),))).astype(getattr(np, args.element_a))
|
||||
tensor_A = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_a_size,))
|
||||
).astype(getattr(np, args.element_a))
|
||||
else:
|
||||
tensor_A = np.random.uniform(low=-2, high=2, size=(problem_size.m()
|
||||
* problem_size.k(),)).astype(getattr(np, args.element_a))
|
||||
tensor_A = np.random.uniform(
|
||||
low=-2, high=2,size=(tensor_a_size,)
|
||||
).astype(getattr(np, args.element_a))
|
||||
|
||||
tensor_b_size = args.batch * problem_size.k() * problem_size.n()
|
||||
if args.element_b != "int8":
|
||||
if args.element_b == "bfloat16":
|
||||
tensor_B = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.k()
|
||||
* problem_size.n(),))).astype(bfloat16)
|
||||
tensor_B = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_b_size,))
|
||||
).astype(bfloat16)
|
||||
else:
|
||||
tensor_B = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.k()
|
||||
* problem_size.n(),))).astype(getattr(np, args.element_b))
|
||||
tensor_B = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_b_size,))
|
||||
).astype(getattr(np, args.element_b))
|
||||
else:
|
||||
tensor_B = np.random.uniform(low=-2, high=2, size=(problem_size.k()
|
||||
* problem_size.n(),)).astype(getattr(np, args.element_b))
|
||||
tensor_B = np.random.uniform(
|
||||
low=-2, high=2, size=(tensor_b_size,)
|
||||
).astype(getattr(np, args.element_b))
|
||||
|
||||
if args.element_c != "int8":
|
||||
if args.element_c == "bfloat16":
|
||||
tensor_C = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.m()
|
||||
* problem_size.n(),))).astype(bfloat16)
|
||||
if args.bias:
|
||||
if args.layout_c == "RowMajor":
|
||||
tensor_c_size = args.batch * problem_size.n()
|
||||
elif args.layout_c == "ColumnMajor":
|
||||
tensor_c_size = args.batch * problem_size.m()
|
||||
else:
|
||||
raise ValueError(args.layout_c)
|
||||
else:
|
||||
tensor_C = np.ceil(np.random.uniform(low=-8.5, high=7.5, size=(problem_size.m()
|
||||
* problem_size.n(),))).astype(getattr(np, args.element_c))
|
||||
tensor_c_size = args.batch * problem_size.m() * problem_size.n()
|
||||
if args.element_c == "bfloat16":
|
||||
tensor_C = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_c_size,))
|
||||
).astype(bfloat16)
|
||||
else:
|
||||
tensor_C = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(tensor_c_size,))
|
||||
).astype(getattr(np, args.element_c))
|
||||
else:
|
||||
tensor_C = np.random.uniform(low=-2, high=2, size=(problem_size.m()
|
||||
* problem_size.n(),)).astype(getattr(np, args.element_c))
|
||||
tensor_C = np.random.uniform(
|
||||
low=-2, high=2, size=(args.batch * problem_size.m() * problem_size.n(),)
|
||||
).astype(getattr(np, args.element_c))
|
||||
|
||||
tensor_D = np.ones_like(tensor_C)
|
||||
tensor_D = np.zeros(
|
||||
shape=(args.batch * problem_size.m() * problem_size.n(),)
|
||||
).astype(getattr(np, args.element_c))
|
||||
|
||||
if args.epilogue_visitor == "RowReduction":
|
||||
cta_n = args.threadblock_shape[1]
|
||||
num_cta_n = (problem_size.n() + cta_n - 1) // cta_n
|
||||
reduction = np.zeros(shape=(args.batch * problem_size.m() * num_cta_n,), dtype=getattr(np, args.element_c))
|
||||
output_op = operation.epilogue_type(
|
||||
D=tensor_D, alpha=args.alpha, beta=args.beta, c=tensor_C, reduction=reduction, problem_size=[problem_size.m(), problem_size.n()]
|
||||
)
|
||||
elif args.epilogue_visitor == "ColumnReduction":
|
||||
cta_m = args.threadblock_shape[0]
|
||||
num_cta_m = (problem_size.m() + cta_m - 1) // cta_m
|
||||
reduction = np.zeros(shape=(args.batch * problem_size.n() * num_cta_m,), dtype=getattr(np, args.element_c))
|
||||
output_op = operation.epilogue_type(
|
||||
D=tensor_D, alpha=args.alpha, beta=args.beta, c=tensor_C, reduction=reduction, problem_size=[problem_size.m(), problem_size.n()]
|
||||
)
|
||||
elif args.epilogue_visitor == "RowBroadcast":
|
||||
vector = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(args.batch, 1, problem_size.n()))
|
||||
).astype(getattr(np, args.element_c))
|
||||
tensor_t = np.empty_like(tensor_D)
|
||||
output_op = operation.epilogue_type(
|
||||
c=tensor_C, vector=vector, alpha=args.alpha, beta=args.beta, Z=tensor_D, T=tensor_t, problem_size=[problem_size.m(), problem_size.n()]
|
||||
)
|
||||
elif args.epilogue_visitor == "ColumnBroadcast":
|
||||
vector = np.ceil(
|
||||
np.random.uniform(low=-8.5, high=7.5, size=(args.batch, problem_size.m(), 1))
|
||||
).astype(getattr(np, args.element_c))
|
||||
tensor_t = np.empty_like(tensor_D)
|
||||
output_op = operation.epilogue_type(
|
||||
c=tensor_C, vector=vector, alpha=args.alpha, beta=args.beta, Z=tensor_D, T=tensor_t, problem_size=[problem_size.m(), problem_size.n()]
|
||||
)
|
||||
else:
|
||||
output_op = operation.epilogue_type(*([args.alpha, args.beta] + args.activation_args))
|
||||
|
||||
arguments = GemmArguments(
|
||||
operation=operation, problem_size=problem_size,
|
||||
A=tensor_A, B=tensor_B, C=tensor_C, D=tensor_D,
|
||||
output_op=LinearCombinationFunctorArguments(args.alpha, args.beta),
|
||||
output_op=output_op,
|
||||
gemm_mode=getattr(cutlass.gemm.Mode, args.gemm_mode),
|
||||
split_k_slices=args.split_k_slices
|
||||
split_k_slices=args.split_k_slices, batch=args.batch
|
||||
)
|
||||
|
||||
if args.gemm_mode == "GemmSplitKParallel":
|
||||
@@ -245,7 +389,8 @@ if args.gemm_mode == "GemmSplitKParallel":
|
||||
problem_size=[problem_size.m(), problem_size.n()],
|
||||
partitions=args.split_k_slices, workspace=arguments.ptr_D,
|
||||
destination=tensor_D, source=tensor_C,
|
||||
output_op=LinearCombinationFunctorArguments(args.alpha, args.beta)
|
||||
output_op=reduction_operation.epilogue_type(*([args.alpha, args.beta] + args.activation_args)),
|
||||
bias = arguments.bias
|
||||
)
|
||||
|
||||
operation.run(arguments)
|
||||
@@ -259,8 +404,42 @@ else:
|
||||
# run the host reference module
|
||||
reference = ReferenceModule(A, B, C)
|
||||
tensor_D_ref = reference.run(
|
||||
tensor_A, tensor_B, tensor_C, problem_size, args.alpha, args.beta)
|
||||
tensor_A, tensor_B, tensor_C, problem_size, args.alpha, args.beta, args.bias, args.batch)
|
||||
|
||||
assert np.array_equal(tensor_D, tensor_D_ref)
|
||||
if args.epilogue_visitor in ["RowBroadcast", "ColumnBroadcast"]:
|
||||
tensor_D_ref = (tensor_D_ref.reshape((args.batch, problem_size.m(), problem_size.n())) + vector).flatten()
|
||||
tensor_D_ref = getattr(pycutlass, args.activation_function).numpy(*([tensor_D_ref,] + args.activation_args))
|
||||
|
||||
if args.epilogue_visitor in ["RowReduction", "ColumnReduction"]:
|
||||
output_op.sync()
|
||||
accum_ref = reference.run(
|
||||
tensor_A, tensor_B, tensor_C, problem_size, 1.0, 0.0, args.bias, args.batch)
|
||||
tensor_D_ref, reduction_ref = epilogue_functor(
|
||||
accum_ref.reshape((args.batch, problem_size.m(), problem_size.n())),
|
||||
tensor_C.reshape((args.batch, problem_size.m(), problem_size.n())),
|
||||
args.alpha, args.beta
|
||||
)
|
||||
tensor_D_ref = tensor_D_ref.flatten()
|
||||
reduction_ref = reduction_ref.flatten()
|
||||
assert np.allclose(reduction_ref, reduction, atol=1e-2)
|
||||
|
||||
elif args.epilogue_visitor in ["RowBroadcast", "ColumnBroadcast"]:
|
||||
output_op.sync()
|
||||
accum_ref = reference.run(
|
||||
tensor_A, tensor_B, tensor_C, problem_size, 1.0, 0.0, args.bias, args.batch)
|
||||
|
||||
tensor_D_ref, tensor_T_ref = epilogue_functor(
|
||||
accum_ref.reshape((args.batch, problem_size.m(), problem_size.n())),
|
||||
tensor_C.reshape((args.batch, problem_size.m(), problem_size.n())),
|
||||
vector, args.alpha, args.beta)
|
||||
|
||||
tensor_D_ref = tensor_D_ref.flatten()
|
||||
tensor_T_ref = tensor_T_ref.flatten()
|
||||
|
||||
assert np.array_equal(tensor_t, tensor_T_ref)
|
||||
|
||||
try:
|
||||
assert np.array_equal(tensor_D, tensor_D_ref)
|
||||
except:
|
||||
assert np.allclose(tensor_D, tensor_D_ref, atol=1e-5)
|
||||
print("Passed.")
|
||||
|
||||
Reference in New Issue
Block a user