[TEST] Fix lit failure from e21efcb

Signed-off-by: Whitney Tsang <[email protected]>

Author: whitneywhtsang

test/TritonGPU/pipeline-lower-loop.mlir

@@ -435,7 +435,7 @@ tt.func @unused_load(%lb : index, %ub : index, %step : index,
 
 // -----
 
-#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, instrShape = [16, 16, 16]}>
+#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, warpsPerCTA = [4, 1], instrShape = [16, 16, 16]}>
 #blocked1 = #ttg.blocked<{sizePerThread = [1, 128], threadsPerWarp = [32, 1], warpsPerCTA = [4, 1], order = [1, 0]}>
 #shared = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = false, elementBitWidth = 16}>
 #shared1 = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = true, elementBitWidth = 16}>
@@ -493,7 +493,7 @@ module attributes {"ttg.num-ctas" = 1 : i32, "ttg.num-warps" = 4 : i32, ttg.targ
 
 // -----
 
-#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, instrShape = [16, 16, 16]}>
+#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, warpsPerCTA = [4, 1], instrShape = [16, 16, 16]}>
 #blocked1 = #ttg.blocked<{sizePerThread = [1, 128], threadsPerWarp = [32, 1], warpsPerCTA = [4, 1], order = [1, 0]}>
 #shared = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = false, elementBitWidth = 32}>
 #shared1 = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = true, elementBitWidth = 32}>
@@ -547,7 +547,7 @@ module attributes {"ttg.num-ctas" = 1 : i32, "ttg.num-warps" = 4 : i32, ttg.targ
 
 // -----
 
-#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, instrShape = [16, 16, 16]}>
+#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, warpsPerCTA = [4, 1], instrShape = [16, 16, 16]}>
 #blocked1 = #ttg.blocked<{sizePerThread = [1, 128], threadsPerWarp = [32, 1], warpsPerCTA = [4, 1], order = [1, 0]}>
 #shared = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = false, elementBitWidth = 16}>
 
@@ -840,7 +840,7 @@ tt.func @tma_reuse_barrier(%lb : index, %ub : index, %step : index,
 
 // -----
 
-#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, instrShape = [16, 16, 16]}>
+#mma = #ttg.nvidia_mma<{versionMajor = 3, versionMinor = 0, warpsPerCTA = [4, 1], instrShape = [16, 16, 16]}>
 #blocked1 = #ttg.blocked<{sizePerThread = [1, 128], threadsPerWarp = [32, 1], warpsPerCTA = [4, 1], order = [1, 0]}>
 #shared = #ttg.nvmma_shared<{swizzlingByteWidth = 128, transposed = false, elementBitWidth = 16}>