On 1/19/19 12:14 AM, Bastian Koppelmann wrote:
> static bool trans_slli(DisasContext *ctx, arg_slli *a)
> {
> - gen_arith_imm(ctx, OPC_RISC_SLLI, a->rd, a->rs1, a->shamt);
> + if (a->rd != 0) {
> + TCGv t = tcg_temp_new();
> + gen_get_gpr(t, a->rs1);
> +
> + if (a->shamt >= TARGET_LONG_BITS) {
> + return false;
> + }
I think the shmat test should be first, so that
slli r0, r1, 99
produces SIGILL instead of translating to a nop.
> static bool trans_sraiw(DisasContext *ctx, arg_sraiw *a)
> {
> - gen_arith_imm(ctx, OPC_RISC_SHIFT_RIGHT_IW , a->rd, a->rs1,
> - a->shamt | 0x400);
> + TCGv t = tcg_temp_new();
> + gen_get_gpr(t, a->rs1);
> + tcg_gen_sextract_tl(t, t, a->shamt, 32 - a->shamt);
> + /* sign-extend for W instructions */
> + tcg_gen_ext32s_tl(t, t);
Sign extension of a sign-extracted value?
r~
