Functionrun_flash_mla(q, block_table, blocked_k, max_seqlen_pad, block_size, b, s_q, cache_seqlens, h_q, h_kv, d, dv, causal, dtype
examples/deepseek_mla/benchmark_mla.py:63
Functionrun_flash_mla_tilelang(q, block_table, blocked_k, max_seqlen_pad, block_size, b, s_q, cache_seqlens, h_q, h_kv, d, dv, causal, dtype
examples/deepseek_mla/benchmark_mla.py:416
Functionrun_flash_mla_triton(q, block_table, blocked_k, max_seqlen_pad, block_size, b, s_q, cache_seqlens, h_q, h_kv, d, dv, causal, dtype
examples/deepseek_mla/benchmark_mla.py:384
Functionrun_flash_mla_triton(q, block_table, blocked_k, max_seqlen_pad, block_size, b, s_q, cache_seqlens, h_q, h_kv, d, dv, causal, dtype
examples/deepseek_mla/amd/benchmark_mla_decode_amd_triton.py:305
Functionrun_flashinfer(q, block_table, blocked_k, max_seqlen_pad, block_size, b, s_q, cache_seqlens, h_q, h_kv, d, dv, causal, dtype
examples/deepseek_mla/benchmark_mla.py:88
Functionrun_gemm_rs(
M,
N,
K,
trans_A,
trans_B,
in_dtype,
out_dtype,
dtypeAccum,
block_M,
testing/python/amd/test_tilelang_test_amd.py:198