diff options
Diffstat (limited to 'target/ppc/translate/vmx-impl.c.inc')
| -rw-r--r-- | target/ppc/translate/vmx-impl.c.inc | 1348 |
1 files changed, 1245 insertions, 103 deletions
diff --git a/target/ppc/translate/vmx-impl.c.inc b/target/ppc/translate/vmx-impl.c.inc index d5e02fd..f91bee8 100644 --- a/target/ppc/translate/vmx-impl.c.inc +++ b/target/ppc/translate/vmx-impl.c.inc @@ -798,45 +798,7 @@ static void trans_vclzd(DisasContext *ctx) tcg_temp_free_i64(avr); } -GEN_VXFORM(vmuloub, 4, 0); -GEN_VXFORM(vmulouh, 4, 1); -GEN_VXFORM(vmulouw, 4, 2); GEN_VXFORM_V(vmuluwm, MO_32, tcg_gen_gvec_mul, 4, 2); -GEN_VXFORM_DUAL(vmulouw, PPC_ALTIVEC, PPC_NONE, - vmuluwm, PPC_NONE, PPC2_ALTIVEC_207) -GEN_VXFORM(vmulosb, 4, 4); -GEN_VXFORM(vmulosh, 4, 5); -GEN_VXFORM(vmulosw, 4, 6); -GEN_VXFORM_V(vmulld, MO_64, tcg_gen_gvec_mul, 4, 7); -GEN_VXFORM(vmuleub, 4, 8); -GEN_VXFORM(vmuleuh, 4, 9); -GEN_VXFORM(vmuleuw, 4, 10); -GEN_VXFORM(vmulhuw, 4, 10); -GEN_VXFORM(vmulhud, 4, 11); -GEN_VXFORM_DUAL(vmuleuw, PPC_ALTIVEC, PPC_NONE, - vmulhuw, PPC_NONE, PPC2_ISA310); -GEN_VXFORM(vmulesb, 4, 12); -GEN_VXFORM(vmulesh, 4, 13); -GEN_VXFORM(vmulesw, 4, 14); -GEN_VXFORM(vmulhsw, 4, 14); -GEN_VXFORM_DUAL(vmulesw, PPC_ALTIVEC, PPC_NONE, - vmulhsw, PPC_NONE, PPC2_ISA310); -GEN_VXFORM(vmulhsd, 4, 15); -GEN_VXFORM_V(vslb, MO_8, tcg_gen_gvec_shlv, 2, 4); -GEN_VXFORM_V(vslh, MO_16, tcg_gen_gvec_shlv, 2, 5); -GEN_VXFORM_V(vslw, MO_32, tcg_gen_gvec_shlv, 2, 6); -GEN_VXFORM(vrlwnm, 2, 6); -GEN_VXFORM_DUAL(vslw, PPC_ALTIVEC, PPC_NONE, \ - vrlwnm, PPC_NONE, PPC2_ISA300) -GEN_VXFORM_V(vsld, MO_64, tcg_gen_gvec_shlv, 2, 23); -GEN_VXFORM_V(vsrb, MO_8, tcg_gen_gvec_shrv, 2, 8); -GEN_VXFORM_V(vsrh, MO_16, tcg_gen_gvec_shrv, 2, 9); -GEN_VXFORM_V(vsrw, MO_32, tcg_gen_gvec_shrv, 2, 10); -GEN_VXFORM_V(vsrd, MO_64, tcg_gen_gvec_shrv, 2, 27); -GEN_VXFORM_V(vsrab, MO_8, tcg_gen_gvec_sarv, 2, 12); -GEN_VXFORM_V(vsrah, MO_16, tcg_gen_gvec_sarv, 2, 13); -GEN_VXFORM_V(vsraw, MO_32, tcg_gen_gvec_sarv, 2, 14); -GEN_VXFORM_V(vsrad, MO_64, tcg_gen_gvec_sarv, 2, 15); GEN_VXFORM(vsrv, 2, 28); GEN_VXFORM(vslv, 2, 29); GEN_VXFORM(vslo, 6, 16); @@ -844,6 +806,387 @@ GEN_VXFORM(vsro, 6, 17); GEN_VXFORM(vaddcuw, 0, 6); GEN_VXFORM(vsubcuw, 0, 22); +static bool do_vector_gvec3_VX(DisasContext *ctx, arg_VX *a, int vece, + void (*gen_gvec)(unsigned, uint32_t, uint32_t, + uint32_t, uint32_t, uint32_t)) +{ + REQUIRE_VECTOR(ctx); + + gen_gvec(vece, avr_full_offset(a->vrt), avr_full_offset(a->vra), + avr_full_offset(a->vrb), 16, 16); + + return true; +} + +TRANS_FLAGS(ALTIVEC, VSLB, do_vector_gvec3_VX, MO_8, tcg_gen_gvec_shlv); +TRANS_FLAGS(ALTIVEC, VSLH, do_vector_gvec3_VX, MO_16, tcg_gen_gvec_shlv); +TRANS_FLAGS(ALTIVEC, VSLW, do_vector_gvec3_VX, MO_32, tcg_gen_gvec_shlv); +TRANS_FLAGS2(ALTIVEC_207, VSLD, do_vector_gvec3_VX, MO_64, tcg_gen_gvec_shlv); + +TRANS_FLAGS(ALTIVEC, VSRB, do_vector_gvec3_VX, MO_8, tcg_gen_gvec_shrv); +TRANS_FLAGS(ALTIVEC, VSRH, do_vector_gvec3_VX, MO_16, tcg_gen_gvec_shrv); +TRANS_FLAGS(ALTIVEC, VSRW, do_vector_gvec3_VX, MO_32, tcg_gen_gvec_shrv); +TRANS_FLAGS2(ALTIVEC_207, VSRD, do_vector_gvec3_VX, MO_64, tcg_gen_gvec_shrv); + +TRANS_FLAGS(ALTIVEC, VSRAB, do_vector_gvec3_VX, MO_8, tcg_gen_gvec_sarv); +TRANS_FLAGS(ALTIVEC, VSRAH, do_vector_gvec3_VX, MO_16, tcg_gen_gvec_sarv); +TRANS_FLAGS(ALTIVEC, VSRAW, do_vector_gvec3_VX, MO_32, tcg_gen_gvec_sarv); +TRANS_FLAGS2(ALTIVEC_207, VSRAD, do_vector_gvec3_VX, MO_64, tcg_gen_gvec_sarv); + +TRANS_FLAGS(ALTIVEC, VRLB, do_vector_gvec3_VX, MO_8, tcg_gen_gvec_rotlv) +TRANS_FLAGS(ALTIVEC, VRLH, do_vector_gvec3_VX, MO_16, tcg_gen_gvec_rotlv) +TRANS_FLAGS(ALTIVEC, VRLW, do_vector_gvec3_VX, MO_32, tcg_gen_gvec_rotlv) +TRANS_FLAGS2(ALTIVEC_207, VRLD, do_vector_gvec3_VX, MO_64, tcg_gen_gvec_rotlv) + +static TCGv_vec do_vrl_mask_vec(unsigned vece, TCGv_vec vrb) +{ + TCGv_vec t0 = tcg_temp_new_vec_matching(vrb), + t1 = tcg_temp_new_vec_matching(vrb), + t2 = tcg_temp_new_vec_matching(vrb), + ones = tcg_constant_vec_matching(vrb, vece, -1); + + /* Extract b and e */ + tcg_gen_dupi_vec(vece, t2, (8 << vece) - 1); + + tcg_gen_shri_vec(vece, t0, vrb, 16); + tcg_gen_and_vec(vece, t0, t0, t2); + + tcg_gen_shri_vec(vece, t1, vrb, 8); + tcg_gen_and_vec(vece, t1, t1, t2); + + /* Compare b and e to negate the mask where begin > end */ + tcg_gen_cmp_vec(TCG_COND_GT, vece, t2, t0, t1); + + /* Create the mask with (~0 >> b) ^ ((~0 >> e) >> 1) */ + tcg_gen_shrv_vec(vece, t0, ones, t0); + tcg_gen_shrv_vec(vece, t1, ones, t1); + tcg_gen_shri_vec(vece, t1, t1, 1); + tcg_gen_xor_vec(vece, t0, t0, t1); + + /* negate the mask */ + tcg_gen_xor_vec(vece, t0, t0, t2); + + tcg_temp_free_vec(t1); + tcg_temp_free_vec(t2); + + return t0; +} + +static void gen_vrlnm_vec(unsigned vece, TCGv_vec vrt, TCGv_vec vra, + TCGv_vec vrb) +{ + TCGv_vec mask, n = tcg_temp_new_vec_matching(vrt); + + /* Create the mask */ + mask = do_vrl_mask_vec(vece, vrb); + + /* Extract n */ + tcg_gen_dupi_vec(vece, n, (8 << vece) - 1); + tcg_gen_and_vec(vece, n, vrb, n); + + /* Rotate and mask */ + tcg_gen_rotlv_vec(vece, vrt, vra, n); + tcg_gen_and_vec(vece, vrt, vrt, mask); + + tcg_temp_free_vec(n); + tcg_temp_free_vec(mask); +} + +static bool do_vrlnm(DisasContext *ctx, arg_VX *a, int vece) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_cmp_vec, INDEX_op_rotlv_vec, INDEX_op_sari_vec, + INDEX_op_shli_vec, INDEX_op_shri_vec, INDEX_op_shrv_vec, 0 + }; + static const GVecGen3 ops[2] = { + { + .fniv = gen_vrlnm_vec, + .fno = gen_helper_VRLWNM, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 + }, + { + .fniv = gen_vrlnm_vec, + .fno = gen_helper_VRLDNM, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 + } + }; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VSX(ctx); + + tcg_gen_gvec_3(avr_full_offset(a->vrt), avr_full_offset(a->vra), + avr_full_offset(a->vrb), 16, 16, &ops[vece - 2]); + + return true; +} + +TRANS(VRLWNM, do_vrlnm, MO_32) +TRANS(VRLDNM, do_vrlnm, MO_64) + +static void gen_vrlmi_vec(unsigned vece, TCGv_vec vrt, TCGv_vec vra, + TCGv_vec vrb) +{ + TCGv_vec mask, n = tcg_temp_new_vec_matching(vrt), + tmp = tcg_temp_new_vec_matching(vrt); + + /* Create the mask */ + mask = do_vrl_mask_vec(vece, vrb); + + /* Extract n */ + tcg_gen_dupi_vec(vece, n, (8 << vece) - 1); + tcg_gen_and_vec(vece, n, vrb, n); + + /* Rotate and insert */ + tcg_gen_rotlv_vec(vece, tmp, vra, n); + tcg_gen_bitsel_vec(vece, vrt, mask, tmp, vrt); + + tcg_temp_free_vec(n); + tcg_temp_free_vec(tmp); + tcg_temp_free_vec(mask); +} + +static bool do_vrlmi(DisasContext *ctx, arg_VX *a, int vece) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_cmp_vec, INDEX_op_rotlv_vec, INDEX_op_sari_vec, + INDEX_op_shli_vec, INDEX_op_shri_vec, INDEX_op_shrv_vec, 0 + }; + static const GVecGen3 ops[2] = { + { + .fniv = gen_vrlmi_vec, + .fno = gen_helper_VRLWMI, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_32 + }, + { + .fniv = gen_vrlnm_vec, + .fno = gen_helper_VRLDMI, + .opt_opc = vecop_list, + .load_dest = true, + .vece = MO_64 + } + }; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VSX(ctx); + + tcg_gen_gvec_3(avr_full_offset(a->vrt), avr_full_offset(a->vra), + avr_full_offset(a->vrb), 16, 16, &ops[vece - 2]); + + return true; +} + +TRANS(VRLWMI, do_vrlmi, MO_32) +TRANS(VRLDMI, do_vrlmi, MO_64) + +static bool do_vector_shift_quad(DisasContext *ctx, arg_VX *a, bool right, + bool alg) +{ + TCGv_i64 hi, lo, t0, t1, n, zero = tcg_constant_i64(0); + + REQUIRE_VECTOR(ctx); + + n = tcg_temp_new_i64(); + hi = tcg_temp_new_i64(); + lo = tcg_temp_new_i64(); + t0 = tcg_temp_new_i64(); + t1 = tcg_const_i64(0); + + get_avr64(lo, a->vra, false); + get_avr64(hi, a->vra, true); + + get_avr64(n, a->vrb, true); + + tcg_gen_andi_i64(t0, n, 64); + if (right) { + tcg_gen_movcond_i64(TCG_COND_NE, lo, t0, zero, hi, lo); + if (alg) { + tcg_gen_sari_i64(t1, lo, 63); + } + tcg_gen_movcond_i64(TCG_COND_NE, hi, t0, zero, t1, hi); + } else { + tcg_gen_movcond_i64(TCG_COND_NE, hi, t0, zero, lo, hi); + tcg_gen_movcond_i64(TCG_COND_NE, lo, t0, zero, zero, lo); + } + tcg_gen_andi_i64(n, n, 0x3F); + + if (right) { + if (alg) { + tcg_gen_sar_i64(t0, hi, n); + } else { + tcg_gen_shr_i64(t0, hi, n); + } + } else { + tcg_gen_shl_i64(t0, lo, n); + } + set_avr64(a->vrt, t0, right); + + if (right) { + tcg_gen_shr_i64(lo, lo, n); + } else { + tcg_gen_shl_i64(hi, hi, n); + } + tcg_gen_xori_i64(n, n, 63); + if (right) { + tcg_gen_shl_i64(hi, hi, n); + tcg_gen_shli_i64(hi, hi, 1); + } else { + tcg_gen_shr_i64(lo, lo, n); + tcg_gen_shri_i64(lo, lo, 1); + } + tcg_gen_or_i64(hi, hi, lo); + set_avr64(a->vrt, hi, !right); + + tcg_temp_free_i64(hi); + tcg_temp_free_i64(lo); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(n); + + return true; +} + +TRANS_FLAGS2(ISA310, VSLQ, do_vector_shift_quad, false, false); +TRANS_FLAGS2(ISA310, VSRQ, do_vector_shift_quad, true, false); +TRANS_FLAGS2(ISA310, VSRAQ, do_vector_shift_quad, true, true); + +static void do_vrlq_mask(TCGv_i64 mh, TCGv_i64 ml, TCGv_i64 b, TCGv_i64 e) +{ + TCGv_i64 th, tl, t0, t1, zero = tcg_constant_i64(0), + ones = tcg_constant_i64(-1); + + th = tcg_temp_new_i64(); + tl = tcg_temp_new_i64(); + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + + /* m = ~0 >> b */ + tcg_gen_andi_i64(t0, b, 64); + tcg_gen_movcond_i64(TCG_COND_NE, t1, t0, zero, zero, ones); + tcg_gen_andi_i64(t0, b, 0x3F); + tcg_gen_shr_i64(mh, t1, t0); + tcg_gen_shr_i64(ml, ones, t0); + tcg_gen_xori_i64(t0, t0, 63); + tcg_gen_shl_i64(t1, t1, t0); + tcg_gen_shli_i64(t1, t1, 1); + tcg_gen_or_i64(ml, t1, ml); + + /* t = ~0 >> e */ + tcg_gen_andi_i64(t0, e, 64); + tcg_gen_movcond_i64(TCG_COND_NE, t1, t0, zero, zero, ones); + tcg_gen_andi_i64(t0, e, 0x3F); + tcg_gen_shr_i64(th, t1, t0); + tcg_gen_shr_i64(tl, ones, t0); + tcg_gen_xori_i64(t0, t0, 63); + tcg_gen_shl_i64(t1, t1, t0); + tcg_gen_shli_i64(t1, t1, 1); + tcg_gen_or_i64(tl, t1, tl); + + /* t = t >> 1 */ + tcg_gen_shli_i64(t0, th, 63); + tcg_gen_shri_i64(tl, tl, 1); + tcg_gen_shri_i64(th, th, 1); + tcg_gen_or_i64(tl, t0, tl); + + /* m = m ^ t */ + tcg_gen_xor_i64(mh, mh, th); + tcg_gen_xor_i64(ml, ml, tl); + + /* Negate the mask if begin > end */ + tcg_gen_movcond_i64(TCG_COND_GT, t0, b, e, ones, zero); + + tcg_gen_xor_i64(mh, mh, t0); + tcg_gen_xor_i64(ml, ml, t0); + + tcg_temp_free_i64(th); + tcg_temp_free_i64(tl); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); +} + +static bool do_vector_rotl_quad(DisasContext *ctx, arg_VX *a, bool mask, + bool insert) +{ + TCGv_i64 ah, al, vrb, n, t0, t1, zero = tcg_constant_i64(0); + + REQUIRE_VECTOR(ctx); + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + + ah = tcg_temp_new_i64(); + al = tcg_temp_new_i64(); + vrb = tcg_temp_new_i64(); + n = tcg_temp_new_i64(); + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + + get_avr64(ah, a->vra, true); + get_avr64(al, a->vra, false); + get_avr64(vrb, a->vrb, true); + + tcg_gen_mov_i64(t0, ah); + tcg_gen_andi_i64(t1, vrb, 64); + tcg_gen_movcond_i64(TCG_COND_NE, ah, t1, zero, al, ah); + tcg_gen_movcond_i64(TCG_COND_NE, al, t1, zero, t0, al); + tcg_gen_andi_i64(n, vrb, 0x3F); + + tcg_gen_shl_i64(t0, ah, n); + tcg_gen_shl_i64(t1, al, n); + + tcg_gen_xori_i64(n, n, 63); + + tcg_gen_shr_i64(al, al, n); + tcg_gen_shri_i64(al, al, 1); + tcg_gen_or_i64(t0, al, t0); + + tcg_gen_shr_i64(ah, ah, n); + tcg_gen_shri_i64(ah, ah, 1); + tcg_gen_or_i64(t1, ah, t1); + + if (mask || insert) { + tcg_gen_shri_i64(n, vrb, 8); + tcg_gen_shri_i64(vrb, vrb, 16); + tcg_gen_andi_i64(n, n, 0x7f); + tcg_gen_andi_i64(vrb, vrb, 0x7f); + + do_vrlq_mask(ah, al, vrb, n); + + tcg_gen_and_i64(t0, t0, ah); + tcg_gen_and_i64(t1, t1, al); + + if (insert) { + get_avr64(n, a->vrt, true); + get_avr64(vrb, a->vrt, false); + tcg_gen_not_i64(ah, ah); + tcg_gen_not_i64(al, al); + tcg_gen_and_i64(n, n, ah); + tcg_gen_and_i64(vrb, vrb, al); + tcg_gen_or_i64(t0, t0, n); + tcg_gen_or_i64(t1, t1, vrb); + } + } + + set_avr64(a->vrt, t0, true); + set_avr64(a->vrt, t1, false); + + tcg_temp_free_i64(ah); + tcg_temp_free_i64(al); + tcg_temp_free_i64(vrb); + tcg_temp_free_i64(n); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + + return true; +} + +TRANS(VRLQ, do_vector_rotl_quad, false, false) +TRANS(VRLQNM, do_vector_rotl_quad, true, false) +TRANS(VRLQMI, do_vector_rotl_quad, false, true) + #define GEN_VXFORM_SAT(NAME, VECE, NORM, SAT, OPC2, OPC3) \ static void glue(glue(gen_, NAME), _vec)(unsigned vece, TCGv_vec t, \ TCGv_vec sat, TCGv_vec a, \ @@ -909,20 +1252,7 @@ GEN_VXFORM3(vsubeuqm, 31, 0); GEN_VXFORM3(vsubecuq, 31, 0); GEN_VXFORM_DUAL(vsubeuqm, PPC_NONE, PPC2_ALTIVEC_207, \ vsubecuq, PPC_NONE, PPC2_ALTIVEC_207) -GEN_VXFORM_V(vrlb, MO_8, tcg_gen_gvec_rotlv, 2, 0); -GEN_VXFORM_V(vrlh, MO_16, tcg_gen_gvec_rotlv, 2, 1); -GEN_VXFORM_V(vrlw, MO_32, tcg_gen_gvec_rotlv, 2, 2); -GEN_VXFORM(vrlwmi, 2, 2); -GEN_VXFORM_DUAL(vrlw, PPC_ALTIVEC, PPC_NONE, \ - vrlwmi, PPC_NONE, PPC2_ISA300) -GEN_VXFORM_V(vrld, MO_64, tcg_gen_gvec_rotlv, 2, 3); -GEN_VXFORM(vrldmi, 2, 3); -GEN_VXFORM_DUAL(vrld, PPC_NONE, PPC2_ALTIVEC_207, \ - vrldmi, PPC_NONE, PPC2_ISA300) GEN_VXFORM_TRANS(vsl, 2, 7); -GEN_VXFORM(vrldnm, 2, 7); -GEN_VXFORM_DUAL(vsl, PPC_ALTIVEC, PPC_NONE, \ - vrldnm, PPC_NONE, PPC2_ISA300) GEN_VXFORM_TRANS(vsr, 2, 11); GEN_VXFORM_ENV(vpkuhum, 7, 0); GEN_VXFORM_ENV(vpkuwum, 7, 1); @@ -1008,41 +1338,252 @@ static void glue(gen_, name0##_##name1)(DisasContext *ctx) \ } \ } -GEN_VXRFORM(vcmpequb, 3, 0) -GEN_VXRFORM(vcmpequh, 3, 1) -GEN_VXRFORM(vcmpequw, 3, 2) -GEN_VXRFORM(vcmpequd, 3, 3) -GEN_VXRFORM(vcmpnezb, 3, 4) -GEN_VXRFORM(vcmpnezh, 3, 5) -GEN_VXRFORM(vcmpnezw, 3, 6) -GEN_VXRFORM(vcmpgtsb, 3, 12) -GEN_VXRFORM(vcmpgtsh, 3, 13) -GEN_VXRFORM(vcmpgtsw, 3, 14) -GEN_VXRFORM(vcmpgtsd, 3, 15) -GEN_VXRFORM(vcmpgtub, 3, 8) -GEN_VXRFORM(vcmpgtuh, 3, 9) -GEN_VXRFORM(vcmpgtuw, 3, 10) -GEN_VXRFORM(vcmpgtud, 3, 11) +static void do_vcmp_rc(int vrt) +{ + TCGv_i64 tmp, set, clr; + + tmp = tcg_temp_new_i64(); + set = tcg_temp_new_i64(); + clr = tcg_temp_new_i64(); + + get_avr64(tmp, vrt, true); + tcg_gen_mov_i64(set, tmp); + get_avr64(tmp, vrt, false); + tcg_gen_or_i64(clr, set, tmp); + tcg_gen_and_i64(set, set, tmp); + + tcg_gen_setcondi_i64(TCG_COND_EQ, clr, clr, 0); + tcg_gen_shli_i64(clr, clr, 1); + + tcg_gen_setcondi_i64(TCG_COND_EQ, set, set, -1); + tcg_gen_shli_i64(set, set, 3); + + tcg_gen_or_i64(tmp, set, clr); + tcg_gen_extrl_i64_i32(cpu_crf[6], tmp); + + tcg_temp_free_i64(tmp); + tcg_temp_free_i64(set); + tcg_temp_free_i64(clr); +} + +static bool do_vcmp(DisasContext *ctx, arg_VC *a, TCGCond cond, int vece) +{ + REQUIRE_VECTOR(ctx); + + tcg_gen_gvec_cmp(cond, vece, avr_full_offset(a->vrt), + avr_full_offset(a->vra), avr_full_offset(a->vrb), 16, 16); + + if (a->rc) { + do_vcmp_rc(a->vrt); + } + + return true; +} + +TRANS_FLAGS(ALTIVEC, VCMPEQUB, do_vcmp, TCG_COND_EQ, MO_8) +TRANS_FLAGS(ALTIVEC, VCMPEQUH, do_vcmp, TCG_COND_EQ, MO_16) +TRANS_FLAGS(ALTIVEC, VCMPEQUW, do_vcmp, TCG_COND_EQ, MO_32) +TRANS_FLAGS2(ALTIVEC_207, VCMPEQUD, do_vcmp, TCG_COND_EQ, MO_64) + +TRANS_FLAGS(ALTIVEC, VCMPGTSB, do_vcmp, TCG_COND_GT, MO_8) +TRANS_FLAGS(ALTIVEC, VCMPGTSH, do_vcmp, TCG_COND_GT, MO_16) +TRANS_FLAGS(ALTIVEC, VCMPGTSW, do_vcmp, TCG_COND_GT, MO_32) +TRANS_FLAGS2(ALTIVEC_207, VCMPGTSD, do_vcmp, TCG_COND_GT, MO_64) +TRANS_FLAGS(ALTIVEC, VCMPGTUB, do_vcmp, TCG_COND_GTU, MO_8) +TRANS_FLAGS(ALTIVEC, VCMPGTUH, do_vcmp, TCG_COND_GTU, MO_16) +TRANS_FLAGS(ALTIVEC, VCMPGTUW, do_vcmp, TCG_COND_GTU, MO_32) +TRANS_FLAGS2(ALTIVEC_207, VCMPGTUD, do_vcmp, TCG_COND_GTU, MO_64) + +TRANS_FLAGS2(ISA300, VCMPNEB, do_vcmp, TCG_COND_NE, MO_8) +TRANS_FLAGS2(ISA300, VCMPNEH, do_vcmp, TCG_COND_NE, MO_16) +TRANS_FLAGS2(ISA300, VCMPNEW, do_vcmp, TCG_COND_NE, MO_32) + +static void gen_vcmpnez_vec(unsigned vece, TCGv_vec t, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t0, t1, zero; + + t0 = tcg_temp_new_vec_matching(t); + t1 = tcg_temp_new_vec_matching(t); + zero = tcg_constant_vec_matching(t, vece, 0); + + tcg_gen_cmp_vec(TCG_COND_EQ, vece, t0, a, zero); + tcg_gen_cmp_vec(TCG_COND_EQ, vece, t1, b, zero); + tcg_gen_cmp_vec(TCG_COND_NE, vece, t, a, b); + + tcg_gen_or_vec(vece, t, t, t0); + tcg_gen_or_vec(vece, t, t, t1); + + tcg_temp_free_vec(t0); + tcg_temp_free_vec(t1); +} + +static bool do_vcmpnez(DisasContext *ctx, arg_VC *a, int vece) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_cmp_vec, 0 + }; + static const GVecGen3 ops[3] = { + { + .fniv = gen_vcmpnez_vec, + .fno = gen_helper_VCMPNEZB, + .opt_opc = vecop_list, + .vece = MO_8 + }, + { + .fniv = gen_vcmpnez_vec, + .fno = gen_helper_VCMPNEZH, + .opt_opc = vecop_list, + .vece = MO_16 + }, + { + .fniv = gen_vcmpnez_vec, + .fno = gen_helper_VCMPNEZW, + .opt_opc = vecop_list, + .vece = MO_32 + } + }; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VECTOR(ctx); + + tcg_gen_gvec_3(avr_full_offset(a->vrt), avr_full_offset(a->vra), + avr_full_offset(a->vrb), 16, 16, &ops[vece]); + + if (a->rc) { + do_vcmp_rc(a->vrt); + } + + return true; +} + +TRANS(VCMPNEZB, do_vcmpnez, MO_8) +TRANS(VCMPNEZH, do_vcmpnez, MO_16) +TRANS(VCMPNEZW, do_vcmpnez, MO_32) + +static bool trans_VCMPEQUQ(DisasContext *ctx, arg_VC *a) +{ + TCGv_i64 t0, t1, t2; + + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + t2 = tcg_temp_new_i64(); + + get_avr64(t0, a->vra, true); + get_avr64(t1, a->vrb, true); + tcg_gen_xor_i64(t2, t0, t1); + + get_avr64(t0, a->vra, false); + get_avr64(t1, a->vrb, false); + tcg_gen_xor_i64(t1, t0, t1); + + tcg_gen_or_i64(t1, t1, t2); + tcg_gen_setcondi_i64(TCG_COND_EQ, t1, t1, 0); + tcg_gen_neg_i64(t1, t1); + + set_avr64(a->vrt, t1, true); + set_avr64(a->vrt, t1, false); + + if (a->rc) { + tcg_gen_extrl_i64_i32(cpu_crf[6], t1); + tcg_gen_andi_i32(cpu_crf[6], cpu_crf[6], 0xa); + tcg_gen_xori_i32(cpu_crf[6], cpu_crf[6], 0x2); + } + + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); + + return true; +} + +static bool do_vcmpgtq(DisasContext *ctx, arg_VC *a, bool sign) +{ + TCGv_i64 t0, t1, t2; + + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + t2 = tcg_temp_new_i64(); + + get_avr64(t0, a->vra, false); + get_avr64(t1, a->vrb, false); + tcg_gen_setcond_i64(TCG_COND_GTU, t2, t0, t1); + + get_avr64(t0, a->vra, true); + get_avr64(t1, a->vrb, true); + tcg_gen_movcond_i64(TCG_COND_EQ, t2, t0, t1, t2, tcg_constant_i64(0)); + tcg_gen_setcond_i64(sign ? TCG_COND_GT : TCG_COND_GTU, t1, t0, t1); + + tcg_gen_or_i64(t1, t1, t2); + tcg_gen_neg_i64(t1, t1); + + set_avr64(a->vrt, t1, true); + set_avr64(a->vrt, t1, false); + + if (a->rc) { + tcg_gen_extrl_i64_i32(cpu_crf[6], t1); + tcg_gen_andi_i32(cpu_crf[6], cpu_crf[6], 0xa); + tcg_gen_xori_i32(cpu_crf[6], cpu_crf[6], 0x2); + } + + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); + + return true; +} + +TRANS(VCMPGTSQ, do_vcmpgtq, true) +TRANS(VCMPGTUQ, do_vcmpgtq, false) + +static bool do_vcmpq(DisasContext *ctx, arg_VX_bf *a, bool sign) +{ + TCGv_i64 vra, vrb; + TCGLabel *gt, *lt, *done; + + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + vra = tcg_temp_local_new_i64(); + vrb = tcg_temp_local_new_i64(); + gt = gen_new_label(); + lt = gen_new_label(); + done = gen_new_label(); + + get_avr64(vra, a->vra, true); + get_avr64(vrb, a->vrb, true); + tcg_gen_brcond_i64((sign ? TCG_COND_GT : TCG_COND_GTU), vra, vrb, gt); + tcg_gen_brcond_i64((sign ? TCG_COND_LT : TCG_COND_LTU), vra, vrb, lt); + + get_avr64(vra, a->vra, false); + get_avr64(vrb, a->vrb, false); + tcg_gen_brcond_i64(TCG_COND_GTU, vra, vrb, gt); + tcg_gen_brcond_i64(TCG_COND_LTU, vra, vrb, lt); + + tcg_gen_movi_i32(cpu_crf[a->bf], CRF_EQ); + tcg_gen_br(done); + + gen_set_label(gt); + tcg_gen_movi_i32(cpu_crf[a->bf], CRF_GT); + tcg_gen_br(done); + + gen_set_label(lt); + tcg_gen_movi_i32(cpu_crf[a->bf], CRF_LT); + tcg_gen_br(done); + + gen_set_label(done); + tcg_temp_free_i64(vra); + tcg_temp_free_i64(vrb); + + return true; +} + +TRANS(VCMPSQ, do_vcmpq, true) +TRANS(VCMPUQ, do_vcmpq, false) + GEN_VXRFORM(vcmpeqfp, 3, 3) GEN_VXRFORM(vcmpgefp, 3, 7) GEN_VXRFORM(vcmpgtfp, 3, 11) GEN_VXRFORM(vcmpbfp, 3, 15) -GEN_VXRFORM(vcmpneb, 3, 0) -GEN_VXRFORM(vcmpneh, 3, 1) -GEN_VXRFORM(vcmpnew, 3, 2) - -GEN_VXRFORM_DUAL(vcmpequb, PPC_ALTIVEC, PPC_NONE, \ - vcmpneb, PPC_NONE, PPC2_ISA300) -GEN_VXRFORM_DUAL(vcmpequh, PPC_ALTIVEC, PPC_NONE, \ - vcmpneh, PPC_NONE, PPC2_ISA300) -GEN_VXRFORM_DUAL(vcmpequw, PPC_ALTIVEC, PPC_NONE, \ - vcmpnew, PPC_NONE, PPC2_ISA300) -GEN_VXRFORM_DUAL(vcmpeqfp, PPC_ALTIVEC, PPC_NONE, \ - vcmpequd, PPC_NONE, PPC2_ALTIVEC_207) -GEN_VXRFORM_DUAL(vcmpbfp, PPC_ALTIVEC, PPC_NONE, \ - vcmpgtsd, PPC_NONE, PPC2_ALTIVEC_207) -GEN_VXRFORM_DUAL(vcmpgtfp, PPC_ALTIVEC, PPC_NONE, \ - vcmpgtud, PPC_NONE, PPC2_ALTIVEC_207) static void gen_vsplti(DisasContext *ctx, int vece) { @@ -1228,6 +1769,141 @@ GEN_VXFORM_DUAL(vsplth, PPC_ALTIVEC, PPC_NONE, GEN_VXFORM_DUAL(vspltw, PPC_ALTIVEC, PPC_NONE, vextractuw, PPC_NONE, PPC2_ISA300); +static bool trans_VGNB(DisasContext *ctx, arg_VX_n *a) +{ + /* + * Similar to do_vextractm, we'll use a sequence of mask-shift-or operations + * to gather the bits. The masks can be created with + * + * uint64_t mask(uint64_t n, uint64_t step) + * { + * uint64_t p = ((1UL << (1UL << step)) - 1UL) << ((n - 1UL) << step), + * plen = n << step, m = 0; + * for(int i = 0; i < 64/plen; i++) { + * m |= p; + * m = ror64(m, plen); + * } + * p >>= plen * DIV_ROUND_UP(64, plen) - 64; + * return m | p; + * } + * + * But since there are few values of N, we'll use a lookup table to avoid + * these calculations at runtime. + */ + static const uint64_t mask[6][5] = { + { + 0xAAAAAAAAAAAAAAAAULL, 0xccccccccccccccccULL, 0xf0f0f0f0f0f0f0f0ULL, + 0xff00ff00ff00ff00ULL, 0xffff0000ffff0000ULL + }, + { + 0x9249249249249249ULL, 0xC30C30C30C30C30CULL, 0xF00F00F00F00F00FULL, + 0xFF0000FF0000FF00ULL, 0xFFFF00000000FFFFULL + }, + { + /* For N >= 4, some mask operations can be elided */ + 0x8888888888888888ULL, 0, 0xf000f000f000f000ULL, 0, + 0xFFFF000000000000ULL + }, + { + 0x8421084210842108ULL, 0, 0xF0000F0000F0000FULL, 0, 0 + }, + { + 0x8208208208208208ULL, 0, 0xF00000F00000F000ULL, 0, 0 + }, + { + 0x8102040810204081ULL, 0, 0xF000000F000000F0ULL, 0, 0 + } + }; + uint64_t m; + int i, sh, nbits = DIV_ROUND_UP(64, a->n); + TCGv_i64 hi, lo, t0, t1; + + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + if (a->n < 2) { + /* + * "N can be any value between 2 and 7, inclusive." Otherwise, the + * result is undefined, so we don't need to change RT. Also, N > 7 is + * impossible since the immediate field is 3 bits only. + */ + return true; + } + + hi = tcg_temp_new_i64(); + lo = tcg_temp_new_i64(); + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + + get_avr64(hi, a->vrb, true); + get_avr64(lo, a->vrb, false); + + /* Align the lower doubleword so we can use the same mask */ + tcg_gen_shli_i64(lo, lo, a->n * nbits - 64); + + /* + * Starting from the most significant bit, gather every Nth bit with a + * sequence of mask-shift-or operation. E.g.: for N=3 + * AxxBxxCxxDxxExxFxxGxxHxxIxxJxxKxxLxxMxxNxxOxxPxxQxxRxxSxxTxxUxxV + * & rep(0b100) + * A..B..C..D..E..F..G..H..I..J..K..L..M..N..O..P..Q..R..S..T..U..V + * << 2 + * .B..C..D..E..F..G..H..I..J..K..L..M..N..O..P..Q..R..S..T..U..V.. + * | + * AB.BC.CD.DE.EF.FG.GH.HI.IJ.JK.KL.LM.MN.NO.OP.PQ.QR.RS.ST.TU.UV.V + * & rep(0b110000) + * AB....CD....EF....GH....IJ....KL....MN....OP....QR....ST....UV.. + * << 4 + * ..CD....EF....GH....IJ....KL....MN....OP....QR....ST....UV...... + * | + * ABCD..CDEF..EFGH..GHIJ..IJKL..KLMN..MNOP..OPQR..QRST..STUV..UV.. + * & rep(0b111100000000) + * ABCD........EFGH........IJKL........MNOP........QRST........UV.. + * << 8 + * ....EFGH........IJKL........MNOP........QRST........UV.......... + * | + * ABCDEFGH....EFGHIJKL....IJKLMNOP....MNOPQRST....QRSTUV......UV.. + * & rep(0b111111110000000000000000) + * ABCDEFGH................IJKLMNOP................QRSTUV.......... + * << 16 + * ........IJKLMNOP................QRSTUV.......................... + * | + * ABCDEFGHIJKLMNOP........IJKLMNOPQRSTUV..........QRSTUV.......... + * & rep(0b111111111111111100000000000000000000000000000000) + * ABCDEFGHIJKLMNOP................................QRSTUV.......... + * << 32 + * ................QRSTUV.......................................... + * | + * ABCDEFGHIJKLMNOPQRSTUV..........................QRSTUV.......... + */ + for (i = 0, sh = a->n - 1; i < 5; i++, sh <<= 1) { + m = mask[a->n - 2][i]; + if (m) { + tcg_gen_andi_i64(hi, hi, m); + tcg_gen_andi_i64(lo, lo, m); + } + if (sh < 64) { + tcg_gen_shli_i64(t0, hi, sh); + tcg_gen_shli_i64(t1, lo, sh); + tcg_gen_or_i64(hi, t0, hi); + tcg_gen_or_i64(lo, t1, lo); + } + } + + tcg_gen_andi_i64(hi, hi, ~(~0ULL >> nbits)); + tcg_gen_andi_i64(lo, lo, ~(~0ULL >> nbits)); + tcg_gen_shri_i64(lo, lo, nbits); + tcg_gen_or_i64(hi, hi, lo); + tcg_gen_trunc_i64_tl(cpu_gpr[a->rt], hi); + + tcg_temp_free_i64(hi); + tcg_temp_free_i64(lo); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + + return true; +} + static bool do_vextdx(DisasContext *ctx, arg_VA *a, int size, bool right, void (*gen_helper)(TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv)) { @@ -1722,6 +2398,124 @@ static bool trans_MTVSRBMI(DisasContext *ctx, arg_DX_b *a) return true; } +static bool do_vcntmb(DisasContext *ctx, arg_VX_mp *a, int vece) +{ + TCGv_i64 rt, vrb, mask; + rt = tcg_const_i64(0); + vrb = tcg_temp_new_i64(); + mask = tcg_constant_i64(dup_const(vece, 1ULL << ((8 << vece) - 1))); + + for (int i = 0; i < 2; i++) { + get_avr64(vrb, a->vrb, i); + if (a->mp) { + tcg_gen_and_i64(vrb, mask, vrb); + } else { + tcg_gen_andc_i64(vrb, mask, vrb); + } + tcg_gen_ctpop_i64(vrb, vrb); + tcg_gen_add_i64(rt, rt, vrb); + } + + tcg_gen_shli_i64(rt, rt, TARGET_LONG_BITS - 8 + vece); + tcg_gen_trunc_i64_tl(cpu_gpr[a->rt], rt); + + tcg_temp_free_i64(vrb); + tcg_temp_free_i64(rt); + + return true; +} + +TRANS(VCNTMBB, do_vcntmb, MO_8) +TRANS(VCNTMBH, do_vcntmb, MO_16) +TRANS(VCNTMBW, do_vcntmb, MO_32) +TRANS(VCNTMBD, do_vcntmb, MO_64) + +static bool do_vstri(DisasContext *ctx, arg_VX_tb_rc *a, + void (*gen_helper)(TCGv_i32, TCGv_ptr, TCGv_ptr)) +{ + TCGv_ptr vrt, vrb; + + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + vrt = gen_avr_ptr(a->vrt); + vrb = gen_avr_ptr(a->vrb); + + if (a->rc) { + gen_helper(cpu_crf[6], vrt, vrb); + } else { + TCGv_i32 discard = tcg_temp_new_i32(); + gen_helper(discard, vrt, vrb); + tcg_temp_free_i32(discard); + } + + tcg_temp_free_ptr(vrt); + tcg_temp_free_ptr(vrb); + + return true; +} + +TRANS(VSTRIBL, do_vstri, gen_helper_VSTRIBL) +TRANS(VSTRIBR, do_vstri, gen_helper_VSTRIBR) +TRANS(VSTRIHL, do_vstri, gen_helper_VSTRIHL) +TRANS(VSTRIHR, do_vstri, gen_helper_VSTRIHR) + +static bool do_vclrb(DisasContext *ctx, arg_VX *a, bool right) +{ + TCGv_i64 rb, mh, ml, tmp, + ones = tcg_constant_i64(-1), + zero = tcg_constant_i64(0); + + rb = tcg_temp_new_i64(); + mh = tcg_temp_new_i64(); + ml = tcg_temp_new_i64(); + tmp = tcg_temp_new_i64(); + + tcg_gen_extu_tl_i64(rb, cpu_gpr[a->vrb]); + tcg_gen_andi_i64(tmp, rb, 7); + tcg_gen_shli_i64(tmp, tmp, 3); + if (right) { + tcg_gen_shr_i64(tmp, ones, tmp); + } else { + tcg_gen_shl_i64(tmp, ones, tmp); + } + tcg_gen_not_i64(tmp, tmp); + + if (right) { + tcg_gen_movcond_i64(TCG_COND_LTU, mh, rb, tcg_constant_i64(8), + tmp, ones); + tcg_gen_movcond_i64(TCG_COND_LTU, ml, rb, tcg_constant_i64(8), + zero, tmp); + tcg_gen_movcond_i64(TCG_COND_LTU, ml, rb, tcg_constant_i64(16), + ml, ones); + } else { + tcg_gen_movcond_i64(TCG_COND_LTU, ml, rb, tcg_constant_i64(8), + tmp, ones); + tcg_gen_movcond_i64(TCG_COND_LTU, mh, rb, tcg_constant_i64(8), + zero, tmp); + tcg_gen_movcond_i64(TCG_COND_LTU, mh, rb, tcg_constant_i64(16), + mh, ones); + } + + get_avr64(tmp, a->vra, true); + tcg_gen_and_i64(tmp, tmp, mh); + set_avr64(a->vrt, tmp, true); + + get_avr64(tmp, a->vra, false); + tcg_gen_and_i64(tmp, tmp, ml); + set_avr64(a->vrt, tmp, false); + + tcg_temp_free_i64(rb); + tcg_temp_free_i64(mh); + tcg_temp_free_i64(ml); + tcg_temp_free_i64(tmp); + + return true; +} + +TRANS(VCLRLB, do_vclrb, false) +TRANS(VCLRRB, do_vclrb, true) + #define GEN_VAFORM_PAIRED(name0, name1, opc2) \ static void glue(gen_, name0##_##name1)(DisasContext *ctx) \ { \ @@ -1765,28 +2559,65 @@ static void gen_vmladduhm(DisasContext *ctx) tcg_temp_free_ptr(rd); } -static void gen_vpermr(DisasContext *ctx) +static bool trans_VPERM(DisasContext *ctx, arg_VA *a) { - TCGv_ptr ra, rb, rc, rd; - if (unlikely(!ctx->altivec_enabled)) { - gen_exception(ctx, POWERPC_EXCP_VPU); - return; - } - ra = gen_avr_ptr(rA(ctx->opcode)); - rb = gen_avr_ptr(rB(ctx->opcode)); - rc = gen_avr_ptr(rC(ctx->opcode)); - rd = gen_avr_ptr(rD(ctx->opcode)); - gen_helper_vpermr(cpu_env, rd, ra, rb, rc); - tcg_temp_free_ptr(ra); - tcg_temp_free_ptr(rb); - tcg_temp_free_ptr(rc); - tcg_temp_free_ptr(rd); + TCGv_ptr vrt, vra, vrb, vrc; + + REQUIRE_INSNS_FLAGS(ctx, ALTIVEC); + REQUIRE_VECTOR(ctx); + + vrt = gen_avr_ptr(a->vrt); + vra = gen_avr_ptr(a->vra); + vrb = gen_avr_ptr(a->vrb); + vrc = gen_avr_ptr(a->rc); + + gen_helper_VPERM(vrt, vra, vrb, vrc); + + tcg_temp_free_ptr(vrt); + tcg_temp_free_ptr(vra); + tcg_temp_free_ptr(vrb); + tcg_temp_free_ptr(vrc); + + return true; +} + +static bool trans_VPERMR(DisasContext *ctx, arg_VA *a) +{ + TCGv_ptr vrt, vra, vrb, vrc; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VECTOR(ctx); + + vrt = gen_avr_ptr(a->vrt); + vra = gen_avr_ptr(a->vra); + vrb = gen_avr_ptr(a->vrb); + vrc = gen_avr_ptr(a->rc); + + gen_helper_VPERMR(vrt, vra, vrb, vrc); + + tcg_temp_free_ptr(vrt); + tcg_temp_free_ptr(vra); + tcg_temp_free_ptr(vrb); + tcg_temp_free_ptr(vrc); + + return true; +} + +static bool trans_VSEL(DisasContext *ctx, arg_VA *a) +{ + REQUIRE_INSNS_FLAGS(ctx, ALTIVEC); + REQUIRE_VECTOR(ctx); + + tcg_gen_gvec_bitsel(MO_64, avr_full_offset(a->vrt), avr_full_offset(a->rc), + avr_full_offset(a->vrb), avr_full_offset(a->vra), + 16, 16); + + return true; } GEN_VAFORM_PAIRED(vmsumubm, vmsummbm, 18) GEN_VAFORM_PAIRED(vmsumuhm, vmsumuhs, 19) GEN_VAFORM_PAIRED(vmsumshm, vmsumshs, 20) -GEN_VAFORM_PAIRED(vsel, vperm, 21) GEN_VAFORM_PAIRED(vmaddfp, vnmsubfp, 23) GEN_VXFORM_NOA(vclzb, 1, 28) @@ -1795,11 +2626,77 @@ GEN_VXFORM_TRANS(vclzw, 1, 30) GEN_VXFORM_TRANS(vclzd, 1, 31) GEN_VXFORM_NOA_2(vnegw, 1, 24, 6) GEN_VXFORM_NOA_2(vnegd, 1, 24, 7) -GEN_VXFORM_NOA_2(vextsb2w, 1, 24, 16) -GEN_VXFORM_NOA_2(vextsh2w, 1, 24, 17) -GEN_VXFORM_NOA_2(vextsb2d, 1, 24, 24) -GEN_VXFORM_NOA_2(vextsh2d, 1, 24, 25) -GEN_VXFORM_NOA_2(vextsw2d, 1, 24, 26) + +static void gen_vexts_i64(TCGv_i64 t, TCGv_i64 b, int64_t s) +{ + tcg_gen_sextract_i64(t, b, 0, 64 - s); +} + +static void gen_vexts_i32(TCGv_i32 t, TCGv_i32 b, int32_t s) +{ + tcg_gen_sextract_i32(t, b, 0, 32 - s); +} + +static void gen_vexts_vec(unsigned vece, TCGv_vec t, TCGv_vec b, int64_t s) +{ + tcg_gen_shli_vec(vece, t, b, s); + tcg_gen_sari_vec(vece, t, t, s); +} + +static bool do_vexts(DisasContext *ctx, arg_VX_tb *a, unsigned vece, int64_t s) +{ + static const TCGOpcode vecop_list[] = { + INDEX_op_shli_vec, INDEX_op_sari_vec, 0 + }; + + static const GVecGen2i op[2] = { + { + .fni4 = gen_vexts_i32, + .fniv = gen_vexts_vec, + .opt_opc = vecop_list, + .vece = MO_32 + }, + { + .fni8 = gen_vexts_i64, + .fniv = gen_vexts_vec, + .opt_opc = vecop_list, + .vece = MO_64 + }, + }; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VECTOR(ctx); + + tcg_gen_gvec_2i(avr_full_offset(a->vrt), avr_full_offset(a->vrb), + 16, 16, s, &op[vece - MO_32]); + + return true; +} + +TRANS(VEXTSB2W, do_vexts, MO_32, 24); +TRANS(VEXTSH2W, do_vexts, MO_32, 16); +TRANS(VEXTSB2D, do_vexts, MO_64, 56); +TRANS(VEXTSH2D, do_vexts, MO_64, 48); +TRANS(VEXTSW2D, do_vexts, MO_64, 32); + +static bool trans_VEXTSD2Q(DisasContext *ctx, arg_VX_tb *a) +{ + TCGv_i64 tmp; + + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + tmp = tcg_temp_new_i64(); + + get_avr64(tmp, a->vrb, false); + set_avr64(a->vrt, tmp, false); + tcg_gen_sari_i64(tmp, tmp, 63); + set_avr64(a->vrt, tmp, true); + + tcg_temp_free_i64(tmp); + return true; +} + GEN_VXFORM_NOA_2(vctzb, 1, 24, 28) GEN_VXFORM_NOA_2(vctzh, 1, 24, 29) GEN_VXFORM_NOA_2(vctzw, 1, 24, 30) @@ -2104,6 +3001,251 @@ static bool trans_VPEXTD(DisasContext *ctx, arg_VX *a) return true; } +static bool trans_VMSUMUDM(DisasContext *ctx, arg_VA *a) +{ + TCGv_i64 rl, rh, src1, src2; + int dw; + + REQUIRE_INSNS_FLAGS2(ctx, ISA300); + REQUIRE_VECTOR(ctx); + + rh = tcg_temp_new_i64(); + rl = tcg_temp_new_i64(); + src1 = tcg_temp_new_i64(); + src2 = tcg_temp_new_i64(); + + get_avr64(rl, a->rc, false); + get_avr64(rh, a->rc, true); + + for (dw = 0; dw < 2; dw++) { + get_avr64(src1, a->vra, dw); + get_avr64(src2, a->vrb, dw); + tcg_gen_mulu2_i64(src1, src2, src1, src2); + tcg_gen_add2_i64(rl, rh, rl, rh, src1, src2); + } + + set_avr64(a->vrt, rl, false); + set_avr64(a->vrt, rh, true); + + tcg_temp_free_i64(rl); + tcg_temp_free_i64(rh); + tcg_temp_free_i64(src1); + tcg_temp_free_i64(src2); + + return true; +} + +static bool trans_VMSUMCUD(DisasContext *ctx, arg_VA *a) +{ + TCGv_i64 tmp0, tmp1, prod1h, prod1l, prod0h, prod0l, zero; + + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + tmp0 = tcg_temp_new_i64(); + tmp1 = tcg_temp_new_i64(); + prod1h = tcg_temp_new_i64(); + prod1l = tcg_temp_new_i64(); + prod0h = tcg_temp_new_i64(); + prod0l = tcg_temp_new_i64(); + zero = tcg_constant_i64(0); + + /* prod1 = vsr[vra+32].dw[1] * vsr[vrb+32].dw[1] */ + get_avr64(tmp0, a->vra, false); + get_avr64(tmp1, a->vrb, false); + tcg_gen_mulu2_i64(prod1l, prod1h, tmp0, tmp1); + + /* prod0 = vsr[vra+32].dw[0] * vsr[vrb+32].dw[0] */ + get_avr64(tmp0, a->vra, true); + get_avr64(tmp1, a->vrb, true); + tcg_gen_mulu2_i64(prod0l, prod0h, tmp0, tmp1); + + /* Sum lower 64-bits elements */ + get_avr64(tmp1, a->rc, false); + tcg_gen_add2_i64(tmp1, tmp0, tmp1, zero, prod1l, zero); + tcg_gen_add2_i64(tmp1, tmp0, tmp1, tmp0, prod0l, zero); + + /* + * Discard lower 64-bits, leaving the carry into bit 64. + * Then sum the higher 64-bit elements. + */ + get_avr64(tmp1, a->rc, true); + tcg_gen_add2_i64(tmp1, tmp0, tmp0, zero, tmp1, zero); + tcg_gen_add2_i64(tmp1, tmp0, tmp1, tmp0, prod1h, zero); + tcg_gen_add2_i64(tmp1, tmp0, tmp1, tmp0, prod0h, zero); + + /* Discard 64 more bits to complete the CHOP128(temp >> 128) */ + set_avr64(a->vrt, tmp0, false); + set_avr64(a->vrt, zero, true); + + tcg_temp_free_i64(tmp0); + tcg_temp_free_i64(tmp1); + tcg_temp_free_i64(prod1h); + tcg_temp_free_i64(prod1l); + tcg_temp_free_i64(prod0h); + tcg_temp_free_i64(prod0l); + + return true; +} + +static bool do_vx_helper(DisasContext *ctx, arg_VX *a, + void (*gen_helper)(TCGv_ptr, TCGv_ptr, TCGv_ptr)) +{ + TCGv_ptr ra, rb, rd; + REQUIRE_VECTOR(ctx); + + ra = gen_avr_ptr(a->vra); + rb = gen_avr_ptr(a->vrb); + rd = gen_avr_ptr(a->vrt); + gen_helper(rd, ra, rb); + tcg_temp_free_ptr(ra); + tcg_temp_free_ptr(rb); + tcg_temp_free_ptr(rd); + + return true; +} + +static bool do_vx_vmuleo(DisasContext *ctx, arg_VX *a, bool even, + void (*gen_mul)(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_i64)) +{ + TCGv_i64 vra, vrb, vrt0, vrt1; + REQUIRE_VECTOR(ctx); + + vra = tcg_temp_new_i64(); + vrb = tcg_temp_new_i64(); + vrt0 = tcg_temp_new_i64(); + vrt1 = tcg_temp_new_i64(); + + get_avr64(vra, a->vra, even); + get_avr64(vrb, a->vrb, even); + gen_mul(vrt0, vrt1, vra, vrb); + set_avr64(a->vrt, vrt0, false); + set_avr64(a->vrt, vrt1, true); + + tcg_temp_free_i64(vra); + tcg_temp_free_i64(vrb); + tcg_temp_free_i64(vrt0); + tcg_temp_free_i64(vrt1); + + return true; +} + +static bool trans_VMULLD(DisasContext *ctx, arg_VX *a) +{ + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + tcg_gen_gvec_mul(MO_64, avr_full_offset(a->vrt), avr_full_offset(a->vra), + avr_full_offset(a->vrb), 16, 16); + + return true; +} + +TRANS_FLAGS2(ALTIVEC_207, VMULESB, do_vx_helper, gen_helper_VMULESB) +TRANS_FLAGS2(ALTIVEC_207, VMULOSB, do_vx_helper, gen_helper_VMULOSB) +TRANS_FLAGS2(ALTIVEC_207, VMULEUB, do_vx_helper, gen_helper_VMULEUB) +TRANS_FLAGS2(ALTIVEC_207, VMULOUB, do_vx_helper, gen_helper_VMULOUB) +TRANS_FLAGS2(ALTIVEC_207, VMULESH, do_vx_helper, gen_helper_VMULESH) +TRANS_FLAGS2(ALTIVEC_207, VMULOSH, do_vx_helper, gen_helper_VMULOSH) +TRANS_FLAGS2(ALTIVEC_207, VMULEUH, do_vx_helper, gen_helper_VMULEUH) +TRANS_FLAGS2(ALTIVEC_207, VMULOUH, do_vx_helper, gen_helper_VMULOUH) +TRANS_FLAGS2(ALTIVEC_207, VMULESW, do_vx_helper, gen_helper_VMULESW) +TRANS_FLAGS2(ALTIVEC_207, VMULOSW, do_vx_helper, gen_helper_VMULOSW) +TRANS_FLAGS2(ALTIVEC_207, VMULEUW, do_vx_helper, gen_helper_VMULEUW) +TRANS_FLAGS2(ALTIVEC_207, VMULOUW, do_vx_helper, gen_helper_VMULOUW) +TRANS_FLAGS2(ISA310, VMULESD, do_vx_vmuleo, true , tcg_gen_muls2_i64) +TRANS_FLAGS2(ISA310, VMULOSD, do_vx_vmuleo, false, tcg_gen_muls2_i64) +TRANS_FLAGS2(ISA310, VMULEUD, do_vx_vmuleo, true , tcg_gen_mulu2_i64) +TRANS_FLAGS2(ISA310, VMULOUD, do_vx_vmuleo, false, tcg_gen_mulu2_i64) + +static void do_vx_vmulhw_i64(TCGv_i64 t, TCGv_i64 a, TCGv_i64 b, bool sign) +{ + TCGv_i64 hh, lh, temp; + + uint64_t c; + hh = tcg_temp_new_i64(); + lh = tcg_temp_new_i64(); + temp = tcg_temp_new_i64(); + + c = 0xFFFFFFFF; + + if (sign) { + tcg_gen_ext32s_i64(lh, a); + tcg_gen_ext32s_i64(temp, b); + } else { + tcg_gen_andi_i64(lh, a, c); + tcg_gen_andi_i64(temp, b, c); + } + tcg_gen_mul_i64(lh, lh, temp); + + if (sign) { + tcg_gen_sari_i64(hh, a, 32); + tcg_gen_sari_i64(temp, b, 32); + } else { + tcg_gen_shri_i64(hh, a, 32); + tcg_gen_shri_i64(temp, b, 32); + } + tcg_gen_mul_i64(hh, hh, temp); + + tcg_gen_shri_i64(lh, lh, 32); + tcg_gen_andi_i64(hh, hh, c << 32); + tcg_gen_or_i64(t, hh, lh); + + tcg_temp_free_i64(hh); + tcg_temp_free_i64(lh); + tcg_temp_free_i64(temp); +} + +static void do_vx_vmulhd_i64(TCGv_i64 t, TCGv_i64 a, TCGv_i64 b, bool sign) +{ + TCGv_i64 tlow; + + tlow = tcg_temp_new_i64(); + if (sign) { + tcg_gen_muls2_i64(tlow, t, a, b); + } else { + tcg_gen_mulu2_i64(tlow, t, a, b); + } + + tcg_temp_free_i64(tlow); +} + +static bool do_vx_mulh(DisasContext *ctx, arg_VX *a, bool sign, + void (*func)(TCGv_i64, TCGv_i64, TCGv_i64, bool)) +{ + REQUIRE_INSNS_FLAGS2(ctx, ISA310); + REQUIRE_VECTOR(ctx); + + TCGv_i64 vra, vrb, vrt; + int i; + + vra = tcg_temp_new_i64(); + vrb = tcg_temp_new_i64(); + vrt = tcg_temp_new_i64(); + + for (i = 0; i < 2; i++) { + get_avr64(vra, a->vra, i); + get_avr64(vrb, a->vrb, i); + get_avr64(vrt, a->vrt, i); + + func(vrt, vra, vrb, sign); + + set_avr64(a->vrt, vrt, i); + } + + tcg_temp_free_i64(vra); + tcg_temp_free_i64(vrb); + tcg_temp_free_i64(vrt); + + return true; + +} + +TRANS(VMULHSW, do_vx_mulh, true , do_vx_vmulhw_i64) +TRANS(VMULHSD, do_vx_mulh, true , do_vx_vmulhd_i64) +TRANS(VMULHUW, do_vx_mulh, false, do_vx_vmulhw_i64) +TRANS(VMULHUD, do_vx_mulh, false, do_vx_vmulhd_i64) + #undef GEN_VR_LDX #undef GEN_VR_STX #undef GEN_VR_LVE |