Lines Matching full:tile
11 %t2 = call <256 x i32> @llvm.x86.cast.tile.to.vector.v256i32(x86_amx %t1)
26 %t2 = call <256 x i32> @llvm.x86.cast.tile.to.vector.v256i32(x86_amx %t1)
38 %t2 = call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %t1)
52 %t2 = call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %t1)
67 %t2 = call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %t1)
69 %t3 = call <256 x i32> @llvm.x86.cast.tile.to.vector.v256i32(x86_amx %t2)
84 %t2 = call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %t1)
86 %t3 = call <256 x i32> @llvm.x86.cast.tile.to.vector.v256i32(x86_amx %t2)
91 ; the shape is loaded after tile.
120 %b.tile.ptr = getelementptr inbounds i8, ptr %b, i64 64
121 %b.tile = load <256 x i32>, ptr %b.tile.ptr, align 64
125 %a.tile.ptr = getelementptr inbounds i8, ptr %a, i64 64
126 %a.tile = load <256 x i32>, ptr %a.tile.ptr, align 64
127 %c.tile.ptr = getelementptr inbounds %struct.__tile1024i_str, ptr %c, i64 0, i32 3
128 %c.tile = load <256 x i32>, ptr %c.tile.ptr, align 64
129 %c.amx = tail call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %c.tile)
130 %a.amx = tail call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %a.tile)
131 %b.amx = tail call x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32> %b.tile)
139 ; CHECK-NEXT: [[TILE:%.*]] = call x86_amx @llvm.x86.tileloadd64.internal(i16 8, i16 32, ptr [[SR…
140 … void @llvm.x86.tilestored64.internal(i16 8, i16 32, ptr [[DST_PTR:%.*]], i64 32, x86_amx [[TILE]])
144 %tile = call x86_amx @llvm.x86.tileloadd64.internal(i16 8, i16 32, ptr %src_ptr, i64 64)
145 %vec = call <256 x i8> @llvm.x86.cast.tile.to.vector.v256i8(x86_amx %tile)
159 %tile = call x86_amx @llvm.x86.cast.vector.to.tile.v256i8(<256 x i8> %vec)
160 call void @llvm.x86.tilestored64.internal(i16 8, i16 32, ptr %dst_ptr, i64 32, x86_amx %tile)
164 declare x86_amx @llvm.x86.cast.vector.to.tile.v256i32(<256 x i32>)
165 declare <256 x i32> @llvm.x86.cast.tile.to.vector.v256i32(x86_amx)
166 declare x86_amx @llvm.x86.cast.vector.to.tile.v256i8(<256 x i8>)
167 declare <256 x i8> @llvm.x86.cast.tile.to.vector.v256i8(x86_amx)