Merge branch 'develop' into ck_tile/fa_bwd_v3

cd4d4629 · danyao12 · 21d12bb7 · 888317e6 · cd4d4629 · cd4d4629
Commit cd4d4629 authored Jan 07, 2025 by danyao12
20 changed files
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n4096_tp_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n4096_tp_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n512_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n512_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n64_n128_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n64_n128_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n768_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_bf16_n768_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n1024_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n1024_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n1536_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n1536_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n2048_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n2048_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n256_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n256_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n3072_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n3072_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n4096_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n4096_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n4096_tp_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n4096_tp_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n512_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n512_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n64_n128_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n64_n128_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n768_instance.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fp16_n768_instance.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fwd_api.cpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_fwd_api.cpp
--- a/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_instance_common.hpp
+++ b/example/ck_tile/14_moe_smoothquant/instances/moe_smoothquant_instance_common.hpp
--- a/example/ck_tile/14_moe_smoothquant/misc/moe-sm.png
+++ b/example/ck_tile/14_moe_smoothquant/misc/moe-sm.png
--- a/example/ck_tile/14_moe_smoothquant/moe_smoothquant.cpp
+++ b/example/ck_tile/14_moe_smoothquant/moe_smoothquant.cpp
--- a/example/ck_tile/14_moe_smoothquant/moe_smoothquant.hpp
+++ b/example/ck_tile/14_moe_smoothquant/moe_smoothquant.hpp
--- a/example/ck_tile/14_moe_smoothquant/script/perf_test.sh
+++ b/example/ck_tile/14_moe_smoothquant/script/perf_test.sh