This patch passes the number of units to aarch64_reverse_mask, which avoids a to_constant () once GET_MODE_NUNITS is variable.
2017-10-26 Richard Sandiford <richard.sandif...@linaro.org> Alan Hayward <alan.hayw...@arm.com> David Sherwood <david.sherw...@arm.com> gcc/ * config/aarch64/aarch64-protos.h (aarch64_reverse_mask): Take the number of units too. * config/aarch64/aarch64.c (aarch64_reverse_mask): Likewise. * config/aarch64/aarch64-simd.md (vec_load_lanesoi<mode>) (vec_store_lanesoi<mode>, vec_load_lanesci<mode>) (vec_store_lanesci<mode>, vec_load_lanesxi<mode>) (vec_store_lanesxi<mode>): Update accordingly. Index: gcc/config/aarch64/aarch64-protos.h =================================================================== --- gcc/config/aarch64/aarch64-protos.h 2017-10-27 14:12:00.601693018 +0100 +++ gcc/config/aarch64/aarch64-protos.h 2017-10-27 14:12:04.192082112 +0100 @@ -365,7 +365,7 @@ bool aarch64_mask_and_shift_for_ubfiz_p bool aarch64_zero_extend_const_eq (machine_mode, rtx, machine_mode, rtx); bool aarch64_move_imm (HOST_WIDE_INT, machine_mode); bool aarch64_mov_operand_p (rtx, machine_mode); -rtx aarch64_reverse_mask (machine_mode); +rtx aarch64_reverse_mask (machine_mode, unsigned int); bool aarch64_offset_7bit_signed_scaled_p (machine_mode, HOST_WIDE_INT); char *aarch64_output_scalar_simd_mov_immediate (rtx, scalar_int_mode); char *aarch64_output_simd_mov_immediate (rtx, unsigned, Index: gcc/config/aarch64/aarch64.c =================================================================== --- gcc/config/aarch64/aarch64.c 2017-10-27 14:12:00.603550436 +0100 +++ gcc/config/aarch64/aarch64.c 2017-10-27 14:12:04.193939530 +0100 @@ -13945,16 +13945,18 @@ aarch64_vectorize_vec_perm_const_ok (mac return ret; } +/* Generate a byte permute mask for a register of mode MODE, + which has NUNITS units. */ + rtx -aarch64_reverse_mask (machine_mode mode) +aarch64_reverse_mask (machine_mode mode, unsigned int nunits) { /* We have to reverse each vector because we dont have a permuted load that can reverse-load according to ABI rules. */ rtx mask; rtvec v = rtvec_alloc (16); - int i, j; - int nunits = GET_MODE_NUNITS (mode); - int usize = GET_MODE_UNIT_SIZE (mode); + unsigned int i, j; + unsigned int usize = GET_MODE_UNIT_SIZE (mode); gcc_assert (BYTES_BIG_ENDIAN); gcc_assert (AARCH64_VALID_SIMD_QREG_MODE (mode)); Index: gcc/config/aarch64/aarch64-simd.md =================================================================== --- gcc/config/aarch64/aarch64-simd.md 2017-10-27 14:12:00.602621727 +0100 +++ gcc/config/aarch64/aarch64-simd.md 2017-10-27 14:12:04.193010821 +0100 @@ -4632,7 +4632,7 @@ (define_expand "vec_load_lanesoi<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (OImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_simd_ld2<mode> (tmp, operands[1])); emit_insn (gen_aarch64_rev_reglistoi (operands[0], tmp, mask)); } @@ -4676,7 +4676,7 @@ (define_expand "vec_store_lanesoi<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (OImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_rev_reglistoi (tmp, operands[1], mask)); emit_insn (gen_aarch64_simd_st2<mode> (operands[0], tmp)); } @@ -4730,7 +4730,7 @@ (define_expand "vec_load_lanesci<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (CImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_simd_ld3<mode> (tmp, operands[1])); emit_insn (gen_aarch64_rev_reglistci (operands[0], tmp, mask)); } @@ -4774,7 +4774,7 @@ (define_expand "vec_store_lanesci<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (CImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_rev_reglistci (tmp, operands[1], mask)); emit_insn (gen_aarch64_simd_st3<mode> (operands[0], tmp)); } @@ -4828,7 +4828,7 @@ (define_expand "vec_load_lanesxi<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (XImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_simd_ld4<mode> (tmp, operands[1])); emit_insn (gen_aarch64_rev_reglistxi (operands[0], tmp, mask)); } @@ -4872,7 +4872,7 @@ (define_expand "vec_store_lanesxi<mode>" if (BYTES_BIG_ENDIAN) { rtx tmp = gen_reg_rtx (XImode); - rtx mask = aarch64_reverse_mask (<MODE>mode); + rtx mask = aarch64_reverse_mask (<MODE>mode, <nunits>); emit_insn (gen_aarch64_rev_reglistxi (tmp, operands[1], mask)); emit_insn (gen_aarch64_simd_st4<mode> (operands[0], tmp)); }