@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / d66ac62854
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 15:45 UTC

[Kernel][Bugfix] Delete some more useless code in marlin_moe_ops.cu (#8643)

Tyler Michael Smith committed Sep 21, 2024 at 19:45 UTC d66ac62854e04c8fda83506dc93ef7971ebf593a
1 file changed -3
csrc/moe/marlin_moe_ops.cu
-3
@@ -1704,9 +1704,6 @@ exec_config_t determine_thread_config(int prob_m, int prob_n, int prob_k,
1704 }
1705
1706 #define CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, NUM_THREADS) \
1707 - __CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, true, 0, NUM_THREADS) \
1708 - __CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, true, 0, NUM_THREADS) \
1709 - __CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, true, 0, NUM_THREADS) \
1707 __CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, true, 0, NUM_THREADS) \
1708 \
1709 __CALL_IF_MOE(W_TYPE, N_BLOCKS, K_BLOCKS, false, -1, NUM_THREADS) \