[CK TILE ENGINE] GEMM Multi D Restructure (#3121)

* Renaming old code

* Adding GEMM code with new Architecture

* Partial Progress : Errors

* Partial Progress : Working code

* Changes to element wise function

* Removing Debugging statements

* Working GEMM Multi D code

* Removing Stale Code

* Address Copilot review comments

* Address Copilot review comments

* Changes to validation file

* Changes to common code snippets

* Creating common folder

* Removing duplicate files

* Pointing to right common file

* Pointing to right common file

* Pointing to right common file

* Changing to VERBOSE

* Changing CMAKE messages to verbose

* Updating Cmake with right layout datatype configs

* Working code for GEMM Multi D
This commit is contained in:
Thrupti Raj Lakshmana Gowda
2025-10-31 14:02:46 -05:00
committed by GitHub
parent 04efd282cf
commit a33d98f8e2
22 changed files with 2443 additions and 2002 deletions

View File

@@ -1,80 +0,0 @@
{
"tile_config": {
"tile_m": {
"values": [
256 ]
},
"tile_n": {
"values": [
128
]
},
"tile_k": {
"values": [
32
]
},
"warp_m": {
"values": [
2
]
},
"warp_n": {
"values": [
2
]
},
"warp_k": {
"values": [
1
]
},
"warp_tile_m": {
"values": [
16
]
},
"warp_tile_n": {
"values": [
16
]
},
"warp_tile_k": {
"values": [
16
]
}
},
"trait_config": {
"pipeline": {
"values": [
"compv3"
]
},
"scheduler": {
"values": [
"intrawave"
]
},
"epilogue": {
"values": [
"cshuffle"
]
},
"pad_m": {
"values": [
false
]
},
"pad_n": {
"values": [
false
]
},
"pad_k": {
"values": [
false
]
}
}
}

View File

@@ -1,84 +1,104 @@
{
"tile_config": {
"tile_m": {
"values": [
256
]
"tile_config": {
"tile_m": {
"max": 256,
"min": 64,
"step": 64
},
"tile_n": {
"max": 256,
"min": 64,
"step": 64
},
"tile_k": {
"max": 256,
"min": 64,
"step": 64
},
"warp_m": {
"values": [
4,
2,
1
]
},
"warp_n": {
"values": [
4,
2,
1
]
},
"warp_k": {
"values": [
1
]
},
"warp_tile_m": {
"values": [
4,
16,
32
]
},
"warp_tile_n": {
"values": [
16,
32,
64
]
},
"warp_tile_k": {
"values": [
8,
16,
32,
64,
128
]
}
},
"tile_n": {
"values": [
128
]
"trait_config": {
"pipeline": {
"values": [
"compv3",
"compv4",
"mem"
]
},
"scheduler": {
"values": [
"intrawave",
"interwave"
]
},
"epilogue": {
"values": [
"cshuffle",
"default"
]
},
"pad_m": {
"values": [
false
]
},
"pad_n": {
"values": [
false
]
},
"pad_k": {
"values": [
false
]
},
"persistent": {
"values": [
false,
true
]
}
},
"tile_k": {
"values": [
32
]
},
"warp_m": {
"values": [
2
]
},
"warp_n": {
"values": [
2
]
},
"warp_k": {
"values": [
1
]
},
"warp_tile_m": {
"values": [
16
]
},
"warp_tile_n": {
"values": [
16
]
},
"warp_tile_k": {
"values": [
16
]
}
},
"trait_config": {
"pipeline": {
"values": [
"compv3",
"compv4",
"mem"
]
},
"scheduler": {
"values": [
"intrawave",
"interwave"
]
},
"epilogue": {
"values": [
"cshuffle"
]
},
"pad_m": {
"values": [
false
]
},
"pad_n": {
"values": [
false
]
},
"pad_k": {
"values": [
false
]
}
}
}
"k_block_per_cu": 1
}

View File

@@ -2,12 +2,12 @@
"tile_config": {
"tile_m": {
"values": [
256
64
]
},
"tile_n": {
"values": [
256
192
]
},
"tile_k": {
@@ -42,24 +42,24 @@
},
"warp_tile_k": {
"values": [
16
8
]
}
},
"trait_config": {
"pipeline": {
"values": [
"compv3"
"compv4"
]
},
"scheduler": {
"values": [
"intrawave"
"intrawave"
]
},
"epilogue": {
"values": [
"cshuffle"
"cshuffle"
]
},
"pad_m": {
@@ -76,6 +76,12 @@
"values": [
false
]
},
"persistent": {
"values": [
true
]
}
}
},
"k_block_per_cu": 1
}