Skip to content

Commit b44a20d

Browse files
author
wangzaijun
committed
update configs.
1 parent d02b1f1 commit b44a20d

12 files changed

+13
-13
lines changed
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 8, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 5}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 3}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 8, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 4, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 4}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 4, "BLOCK_DIM": 128, "num_warps": 4, "NUM_STAGE": 3}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 4, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 16, "BLOCK_DIM": 128, "num_warps": 4, "NUM_STAGE": 5}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 8, "NUM_STAGE": 4}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 16, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 16, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 3}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 2}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 5}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 3}}
1+
{"1": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 5}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 2}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 5}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 16, "BLOCK_DIM": 256, "num_warps": 4, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 1, "BLOCK_DIM": 128, "num_warps": 4, "NUM_STAGE": 3}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 4}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 4}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 128, "num_warps": 2, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 32, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 4}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 4}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 5}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 2}, "4096": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 2}}
1+
{"1": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 3}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 3}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 4}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 5}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 8, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 2, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 2, "BLOCK_DIM": 256, "num_warps": 4, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 128, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 5}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 3}, "2048": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 5}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 2}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 3}}
1+
{"1": {"BLOCK_M": 16, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 2}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 5}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 5}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 5}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 5}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 3}, "2048": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 3}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 4}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 1, "BLOCK_DIM": 128, "num_warps": 4, "NUM_STAGE": 5}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "256": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 2, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 4, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 2, "BLOCK_DIM": 128, "num_warps": 4, "NUM_STAGE": 2}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 16, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 2}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 5}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "2048": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "8192": {"BLOCK_M": 4, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}}
1+
{"1": {"BLOCK_M": 4, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 5}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}, "128": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 3}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 8, "NUM_STAGE": 2}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}, "4096": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 3}, "8192": {"BLOCK_M": 4, "BLOCK_DIM": 1024, "num_warps": 2, "NUM_STAGE": 4}}
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
{"1": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 4, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 2, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 1, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 4, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 1, "NUM_STAGE": 1}}
1+
{"1": {"BLOCK_M": 2, "BLOCK_DIM": 512, "num_warps": 8, "NUM_STAGE": 1}, "8": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 16, "NUM_STAGE": 1}, "64": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "128": {"BLOCK_M": 2, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "256": {"BLOCK_M": 1, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "512": {"BLOCK_M": 2, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "1024": {"BLOCK_M": 2, "BLOCK_DIM": 256, "num_warps": 1, "NUM_STAGE": 1}, "2048": {"BLOCK_M": 1, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "4096": {"BLOCK_M": 2, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 1}, "8192": {"BLOCK_M": 4, "BLOCK_DIM": 1024, "num_warps": 1, "NUM_STAGE": 4}}

0 commit comments

Comments
 (0)