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) \