TensorRT-LLMs/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/kernelUtils.cpp
Pengbo Wang c0e25e5418
[TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264)
Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
2026-01-11 19:26:10 -05:00

227 lines
6.3 KiB
C++

/*
* Copyright (c) 2020-2023, NVIDIA CORPORATION. All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/kernelUtils.h"
#include "tensorrt_llm/common/config.h"
#include "tensorrt_llm/common/utils.h"
#include "tensorrt_llm/kernels/multiHeadAttentionCommon.h"
#include <list>
TRTLLM_NAMESPACE_BEGIN
namespace kernels
{
namespace jit
{
namespace
{
using tensorrt_llm::common::contains;
bool supportConfigCommon(XQAParams const& xqaParams, bool forConfigurePlugin)
{
if (xqaParams.unidirectional != 1)
{
return false;
}
if (xqaParams.mask_type != tensorrt_llm::kernels::AttentionMaskType::CAUSAL)
{
return false;
}
if (xqaParams.cross_attention)
{
return false;
}
if (xqaParams.position_shift_enabled || xqaParams.sink_token_length > 0)
{
return false;
}
if (xqaParams.num_kv_heads != 0 && xqaParams.num_q_heads % xqaParams.num_kv_heads != 0)
{
return false;
}
bool const is_vanilla_mha = !xqaParams.multi_query_tokens
&& (xqaParams.num_kv_heads == 0 || xqaParams.num_q_heads == xqaParams.num_kv_heads);
if (is_vanilla_mha && xqaParams.beam_width == 1)
{
// Do not use XQA kernel for vanilla MHA case for performance reasons.
return false;
}
if (is_vanilla_mha && xqaParams.head_size <= 128)
{
// TODO: remove this when the kernel bug for num_kv_heads <= 128 gets fixed.
return false;
}
if (!contains({PositionEmbeddingType::kROPE_GPTJ, PositionEmbeddingType::kROPE_GPT_NEOX,
PositionEmbeddingType::kROPE_M, PositionEmbeddingType::kLONG_ROPE,
PositionEmbeddingType::kLEARNED_ABSOLUTE, PositionEmbeddingType::kYARN},
xqaParams.position_embedding_type))
{
return false;
}
if (xqaParams.chunked_attention_size != INT_MAX)
{
// TODO: chunked attention is not supported yet.
return false;
}
return true;
}
} // anonymous namespace
bool supportConfigQGMMA(XQAParams const& xqaParams, int SM, bool forConfigurePlugin)
{
if (!supportConfigCommon(xqaParams, forConfigurePlugin))
{
return false;
}
if (SM != kSM_90)
{
return false;
}
if (!contains({DATA_TYPE_FP16, DATA_TYPE_BF16}, xqaParams.data_type))
{
return false;
}
if (!contains({DATA_TYPE_FP16, DATA_TYPE_BF16, DATA_TYPE_E4M3}, xqaParams.kv_cache_data_type))
{
return false;
}
bool const is_skip_softmax = xqaParams.skip_softmax_threshold_scale_factor != 0;
if (!is_skip_softmax && xqaParams.kv_cache_data_type != DATA_TYPE_E4M3)
{
// Only use hopper kernel with fp16/bf16 kv cache data type when skip softmax is enabled
return false;
}
if (xqaParams.beam_width != 1)
{
return false;
}
if (xqaParams.head_size % 16 != 0 || xqaParams.head_size < 16 || xqaParams.head_size > 256)
{
return false;
}
int32_t head_grp_size = xqaParams.num_kv_heads == 0 ? 1 : xqaParams.num_q_heads / xqaParams.num_kv_heads;
if (head_grp_size * xqaParams.beam_width > 32)
{
return false;
}
if (xqaParams.paged_kv_cache && !contains({8, 16, 32, 64, 128}, xqaParams.tokens_per_block))
{
return false;
}
return true;
}
bool supportConfigHMMA(XQAParams const& xqaParams, int SM, bool forConfigurePlugin)
{
if (!supportConfigCommon(xqaParams, forConfigurePlugin))
{
return false;
}
if (SM < kSM_80)
{
return false;
}
if (!contains({DATA_TYPE_FP16, DATA_TYPE_BF16}, xqaParams.data_type))
{
return false;
}
if (!contains({DATA_TYPE_FP16, DATA_TYPE_BF16, DATA_TYPE_INT8, DATA_TYPE_E4M3}, xqaParams.kv_cache_data_type))
{
return false;
}
if (xqaParams.beam_width != 1 && xqaParams.beam_width != 4)
{
return false;
}
if (!forConfigurePlugin)
{
// Inference time checks.
if (xqaParams.host_past_key_value_lengths == nullptr)
{
return false;
}
if (!xqaParams.multi_query_tokens && xqaParams.beam_width != 1
&& xqaParams.max_past_kv_length + 1 > xqaParams.cyclic_attention_window_size)
{
return false;
}
}
if (xqaParams.head_size % 16 != 0 || xqaParams.head_size < 16 || xqaParams.head_size > 256)
{
return false;
}
int32_t head_grp_size = xqaParams.num_kv_heads == 0 ? 1 : xqaParams.num_q_heads / xqaParams.num_kv_heads;
if (head_grp_size * xqaParams.beam_width > 32)
{
return false;
}
if (xqaParams.paged_kv_cache && !contains({16, 32, 64, 128}, xqaParams.tokens_per_block))
{
return false;
}
bool const is_skip_softmax = xqaParams.skip_softmax_threshold_scale_factor != 0;
if (is_skip_softmax)
{
return false;
}
return true;
}
bool supportConfigMLA(XQAParams const& xqaParams, int SM, bool forConfigurePlugin)
{
if (!supportConfigCommon(xqaParams, forConfigurePlugin))
{
return false;
}
if (SM != kSM_120)
{
return false;
}
if (xqaParams.data_type != DATA_TYPE_E4M3)
{
return false;
}
if (xqaParams.kv_cache_data_type != DATA_TYPE_E4M3)
{
return false;
}
if (xqaParams.beam_width != 1)
{
return false;
}
if (!xqaParams.isMLA())
{
return false;
}
if (xqaParams.paged_kv_cache && !contains({8, 16, 32, 64, 128}, xqaParams.tokens_per_block))
{
return false;
}
bool const is_skip_softmax = xqaParams.skip_softmax_threshold_scale_factor != 0;
if (is_skip_softmax)
{
return false;
}
return true;
}
} // namespace jit
} // namespace kernels
TRTLLM_NAMESPACE_END