Lines Matching full:sext
932 // fold (aext (sext x)) -> (sext x)
1541 // add (sext i1), X -> sub X, (zext i1)
2371 // fold (OP (sext x), (sext y)) -> (sext (OP x, y))
3795 // If truncate is free for the target sext(shl) is likely to result in better
4396 // fold (sext c1) -> c1
4400 // fold (sext (sext x)) -> (sext x)
4401 // fold (sext (aext x)) -> (sext x)
4407 // fold (sext (truncate (load x))) -> (sext (smaller load x))
4408 // fold (sext (truncate (srl (load x), c))) -> (sext (smaller load (x+c/n)))
4421 // eliminate the trunc/sext pair.
4435 // bits, just sext from i32.
4445 // fold (sext (truncate x)) -> (sextinreg x).
4457 // fold (sext (load x)) -> (sext (truncate (sextload x)))
4486 // fold (sext (sextload x)) -> (sext (truncate (sextload x)))
4487 // fold (sext ( extload x)) -> (sext (truncate (sextload x)))
4509 // fold (sext (and/or/xor (load x), cst)) ->
4510 // (and/or/xor (sextload x), (sext cst))
4533 Mask = Mask.sext(VT.getSizeInBits());
4549 // sext(setcc) -> sext_in_reg(vsetcc) for vectors.
4556 // of the same size as the compared operands. Only optimize sext(setcc())
4563 // we know that the element size of the sext'd result matches the
4582 // sext(setcc x, y, cc) -> (select_cc x, y, -1, 0, cc)
4603 // fold (sext
4854 // we know that the element size of the sext'd result matches the
4930 // fold (aext (sext x)) -> (sext x)
5038 // we know that the element size of the sext'd result matches the
5307 // fold (sext_in_reg (sext x)) -> (sext x)
5308 // fold (sext_in_reg (aext x)) -> (sext x)
5714 // If X has a different width than the result/lhs, sext it or truncate it.
