Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
vllm_cscc
Commits
d36deb1a
Commit
d36deb1a
authored
Jun 04, 2025
by
zhuwenwen
Browse files
[feat]add 120cu nn fused_moe configs of qwen3, mixtral8*7b and mixtral8*22b
parent
2e3bfb1e
Changes
3
Hide whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
388 additions
and
60 deletions
+388
-60
vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=K100_AI_nn.json
.../fused_moe/configs/E=64,N=640,device_name=K100_AI_nn.json
+164
-0
vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=K100_AI_nn.json
.../fused_moe/configs/E=8,N=2048,device_name=K100_AI_nn.json
+60
-60
vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=K100_AI_nn.json
.../fused_moe/configs/E=8,N=3584,device_name=K100_AI_nn.json
+164
-0
No files found.
vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=K100_AI_nn.json
0 → 100644
View file @
d36deb1a
{
"1"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"2"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"4"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_K"
:
128
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"8"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
5
,
"num_ldmatrixes"
:
1
},
"16"
:
{
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
5
,
"num_ldmatrixes"
:
1
},
"24"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
5
,
"num_ldmatrixes"
:
1
},
"32"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"48"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"64"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"96"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"128"
:
{
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"256"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"512"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"1024"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"1536"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"2048"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"3072"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"4096"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
}
}
vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=K100_AI_nn.json
View file @
d36deb1a
...
@@ -3,97 +3,97 @@
...
@@ -3,97 +3,97 @@
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
64
,
"num_warps"
:
4
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"2"
:
{
"2"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
2
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"4"
:
{
"4"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
32
,
"num_warps"
:
2
,
"num_warps"
:
2
,
"num_stages"
:
3
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"8"
:
{
"8"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
32
,
"num_warps"
:
4
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"16"
:
{
"16"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
64
,
"num_warps"
:
4
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"24"
:
{
"24"
:
{
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
4
,
"num_warps"
:
2
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"32"
:
{
"32"
:
{
"BLOCK_SIZE_M"
:
1
28
,
"BLOCK_SIZE_M"
:
1
6
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
32
,
"num_warps"
:
4
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"48"
:
{
"48"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_M"
:
32
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
8
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"64"
:
{
"64"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
32
,
"num_warps"
:
4
,
"num_warps"
:
8
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"96"
:
{
"96"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
4
,
"num_warps"
:
8
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"128"
:
{
"128"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
4
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
...
@@ -102,14 +102,14 @@
...
@@ -102,14 +102,14 @@
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
6
,
"num_warps"
:
4
,
"num_warps"
:
8
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"512"
:
{
"512"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_warps"
:
4
,
...
@@ -117,17 +117,17 @@
...
@@ -117,17 +117,17 @@
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"1024"
:
{
"1024"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"1536"
:
{
"1536"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_warps"
:
8
,
...
@@ -135,8 +135,8 @@
...
@@ -135,8 +135,8 @@
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"2048"
:
{
"2048"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_warps"
:
8
,
...
@@ -144,20 +144,20 @@
...
@@ -144,20 +144,20 @@
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"3072"
:
{
"3072"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
},
},
"4096"
:
{
"4096"
:
{
"BLOCK_SIZE_M"
:
256
,
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
"num_ldmatrixes"
:
1
}
}
...
...
vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=K100_AI_nn.json
0 → 100644
View file @
d36deb1a
{
"1"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
64
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
16
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"2"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
32
,
"num_warps"
:
2
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"4"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
64
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"8"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"16"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"24"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"32"
:
{
"BLOCK_SIZE_M"
:
16
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
2
,
"num_stages"
:
4
,
"num_ldmatrixes"
:
1
},
"48"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"64"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
32
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"96"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"128"
:
{
"BLOCK_SIZE_M"
:
64
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
16
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"256"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"512"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
256
,
"BLOCK_SIZE_K"
:
32
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
3
,
"num_ldmatrixes"
:
1
},
"1024"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"1536"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"2048"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
8
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"3072"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
},
"4096"
:
{
"BLOCK_SIZE_M"
:
128
,
"BLOCK_SIZE_N"
:
128
,
"BLOCK_SIZE_K"
:
64
,
"GROUP_SIZE_M"
:
1
,
"num_warps"
:
4
,
"num_stages"
:
2
,
"num_ldmatrixes"
:
1
}
}
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment