[v3,09/37] target/ppc: Move Vector Compare Equal/Not Equal/Greater Than to decodetree

Message ID	20220210123447.3933301-10-matheus.ferst@eldorado.org.br (mailing list archive)
State	New, archived
Headers	show Return-Path: <qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org> From: matheus.ferst@eldorado.org.br To: qemu-devel@nongnu.org, qemu-ppc@nongnu.org Subject: [PATCH v3 09/37] target/ppc: Move Vector Compare Equal/Not Equal/Greater Than to decodetree Date: Thu, 10 Feb 2022 09:34:19 -0300 Message-Id: <20220210123447.3933301-10-matheus.ferst@eldorado.org.br> In-Reply-To: <20220210123447.3933301-1-matheus.ferst@eldorado.org.br> References: <20220210123447.3933301-1-matheus.ferst@eldorado.org.br> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Received-SPF: pass client-ip=187.72.171.209; envelope-from=matheus.ferst@eldorado.org.br; helo=outlook.eldorado.org.br X-Spam_score_int: -10 X-Spam_score: -1.1 X-Spam_bar: - X-Spam_report: (-1.1 / 5.0 requ) BAYES_00=-1.9, PDS_HP_HELO_NORDNS=0.001, RDNS_NONE=0.793, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=no autolearn_force=no X-Spam_action: no action Precedence: list Cc: danielhb413@gmail.com, richard.henderson@linaro.org, groug@kaod.org, clg@kaod.org, Matheus Ferst <matheus.ferst@eldorado.org.br>, david@gibson.dropbear.id.au Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" <qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org>
Series	target/ppc: PowerISA Vector/VSX instruction batch \| expand [v3,00/37] target/ppc: PowerISA Vector/VSX instruction batch [v3,01/37] target/ppc: Introduce TRANSFLAGS macros [v3,02/37] target/ppc: moved vector even and odd multiplication to decodetree [v3,03/37] target/ppc: Moved vector multiply high and low to decodetree [v3,04/37] target/ppc: vmulh instructions use gvec [v3,05/37] target/ppc: Implement vmsumcud instruction [v3,06/37] target/ppc: Implement vmsumudm instruction [v3,07/37] target/ppc: Move vexts[bhw]2[wd] to decodetree [v3,08/37] target/ppc: Implement vextsd2q [v3,09/37] target/ppc: Move Vector Compare Equal/Not Equal/Greater Than to decodetree [v3,10/37] target/ppc: Move Vector Compare Not Equal or Zero to decodetree [v3,11/37] target/ppc: Implement Vector Compare Equal Quadword [v3,12/37] target/ppc: Implement Vector Compare Greater Than Quadword [v3,13/37] target/ppc: Implement Vector Compare Quadword [v3,14/37] target/ppc: implement vstri[bh][lr] [v3,15/37] target/ppc: implement vclrlb [v3,16/37] target/ppc: implement vclrrb [v3,17/37] target/ppc: implement vcntmb[bhwd] [v3,18/37] target/ppc: implement vgnb [v3,19/37] target/ppc: Move vsel and vperm/vpermr to decodetree [v3,20/37] target/ppc: Move xxsel to decodetree [v3,21/37] target/ppc: move xxperm/xxpermr to decodetree [v3,22/37] target/ppc: Move xxpermdi to decodetree [v3,23/37] target/ppc: Implement xxpermx instruction [v3,24/37] tcg/tcg-op-gvec.c: Introduce tcg_gen_gvec_4i [v3,25/37] target/ppc: Implement xxeval [v3,26/37] target/ppc: Implement xxgenpcv[bhwd]m instruction [v3,27/37] target/ppc: move xs[n]madd[am][ds]p/xs[n]msub[am][ds]p to decodetree [v3,28/37] target/ppc: implement xs[n]maddqp[o]/xs[n]msubqp[o] [v3,29/37] target/ppc: Implement xvtlsbb instruction [v3,30/37] target/ppc: Remove xscmpnedp instruction [v3,31/37] target/ppc: Refactor VSX_SCALAR_CMP_DP [v3,32/37] target/ppc: Implement xscmp{eq,ge,gt}qp [v3,33/37] target/ppc: Move xscmp{eq,ge,gt}dp to decodetree [v3,34/37] target/ppc: Move xs{max, min}[cj]dp to use do_helper_XX3 [v3,35/37] target/ppc: Refactor VSX_MAX_MINC helper [v3,36/37] target/ppc: Implement xs{max,min}cqp [v3,37/37] target/ppc: Implement xvcvbf16spn and xvcvspbf16 instructions

diff --git a/target/ppc/helper.h b/target/ppc/helper.h index 0084080fad..4f0f3e3a08 100644 --- a/target/ppc/helper.h +++ b/target/ppc/helper.h @@ -141,46 +141,16 @@ DEF_HELPER_3(vabsduw, void, avr, avr, avr) DEF_HELPER_3(vavgsb, void, avr, avr, avr) DEF_HELPER_3(vavgsh, void, avr, avr, avr) DEF_HELPER_3(vavgsw, void, avr, avr, avr) -DEF_HELPER_4(vcmpequb, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequh, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequw, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequd, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpneb, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpneh, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpnew, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezb, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezh, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezw, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtub, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtuh, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtuw, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtud, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsb, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsh, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsw, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsd, void, env, avr, avr, avr) DEF_HELPER_4(vcmpeqfp, void, env, avr, avr, avr) DEF_HELPER_4(vcmpgefp, void, env, avr, avr, avr) DEF_HELPER_4(vcmpgtfp, void, env, avr, avr, avr) DEF_HELPER_4(vcmpbfp, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequb_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequh_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequw_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpequd_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpneb_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpneh_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpnew_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezb_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezh_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpnezw_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtub_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtuh_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtuw_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtud_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsb_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsh_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsw_dot, void, env, avr, avr, avr) -DEF_HELPER_4(vcmpgtsd_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpeqfp_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpgefp_dot, void, env, avr, avr, avr) DEF_HELPER_4(vcmpgtfp_dot, void, env, avr, avr, avr) diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode index 1c935eeac8..bfb36e6969 100644 --- a/target/ppc/insn32.decode +++ b/target/ppc/insn32.decode @@ -51,6 +51,9 @@ &VA vrt vra vrb rc @VA ...... vrt:5 vra:5 vrb:5 rc:5 ...... &VA +&VC vrt vra vrb rc:bool +@VC ...... vrt:5 vra:5 vrb:5 rc:1 .......... &VC + &VN vrt vra vrb sh @VN ...... vrt:5 vra:5 vrb:5 .. sh:3 ...... &VN @@ -373,6 +376,27 @@ DSCLIQ 111111 ..... ..... ...... 001000010 . @Z22_tap_sh_rc DSCRI 111011 ..... ..... ...... 001100010 . @Z22_ta_sh_rc DSCRIQ 111111 ..... ..... ...... 001100010 . @Z22_tap_sh_rc +## Vector Integer Instructions + +VCMPEQUB 000100 ..... ..... ..... . 0000000110 @VC +VCMPEQUH 000100 ..... ..... ..... . 0001000110 @VC +VCMPEQUW 000100 ..... ..... ..... . 0010000110 @VC +VCMPEQUD 000100 ..... ..... ..... . 0011000111 @VC + +VCMPGTSB 000100 ..... ..... ..... . 1100000110 @VC +VCMPGTSH 000100 ..... ..... ..... . 1101000110 @VC +VCMPGTSW 000100 ..... ..... ..... . 1110000110 @VC +VCMPGTSD 000100 ..... ..... ..... . 1111000111 @VC + +VCMPGTUB 000100 ..... ..... ..... . 1000000110 @VC +VCMPGTUH 000100 ..... ..... ..... . 1001000110 @VC +VCMPGTUW 000100 ..... ..... ..... . 1010000110 @VC +VCMPGTUD 000100 ..... ..... ..... . 1011000111 @VC + +VCMPNEB 000100 ..... ..... ..... . 0000000111 @VC +VCMPNEH 000100 ..... ..... ..... . 0001000111 @VC +VCMPNEW 000100 ..... ..... ..... . 0010000111 @VC + ## Vector Bit Manipulation Instruction VCFUGED 000100 ..... ..... ..... 10101001101 @VX diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c index 630fbc579a..83e718ab0e 100644 --- a/target/ppc/int_helper.c +++ b/target/ppc/int_helper.c @@ -749,57 +749,6 @@ VCF(ux, uint32_to_float32, u32) VCF(sx, int32_to_float32, s32) #undef VCF -#define VCMP_DO(suffix, compare, element, record) \ - void helper_vcmp##suffix(CPUPPCState *env, ppc_avr_t *r, \ - ppc_avr_t *a, ppc_avr_t *b) \ - { \ - uint64_t ones = (uint64_t)-1; \ - uint64_t all = ones; \ - uint64_t none = 0; \ - int i; \ - \ - for (i = 0; i < ARRAY_SIZE(r->element); i++) { \ - uint64_t result = (a->element[i] compare b->element[i] ? \ - ones : 0x0); \ - switch (sizeof(a->element[0])) { \ - case 8: \ - r->u64[i] = result; \ - break; \ - case 4: \ - r->u32[i] = result; \ - break; \ - case 2: \ - r->u16[i] = result; \ - break; \ - case 1: \ - r->u8[i] = result; \ - break; \ - } \ - all &= result; \ - none |= result; \ - } \ - if (record) { \ - env->crf[6] = ((all != 0) << 3) | ((none == 0) << 1); \ - } \ - } -#define VCMP(suffix, compare, element) \ - VCMP_DO(suffix, compare, element, 0) \ - VCMP_DO(suffix##_dot, compare, element, 1) -VCMP(equb, ==, u8) -VCMP(equh, ==, u16) -VCMP(equw, ==, u32) -VCMP(equd, ==, u64) -VCMP(gtub, >, u8) -VCMP(gtuh, >, u16) -VCMP(gtuw, >, u32) -VCMP(gtud, >, u64) -VCMP(gtsb, >, s8) -VCMP(gtsh, >, s16) -VCMP(gtsw, >, s32) -VCMP(gtsd, >, s64) -#undef VCMP_DO -#undef VCMP - #define VCMPNE_DO(suffix, element, etype, cmpzero, record) \ void helper_vcmpne##suffix(CPUPPCState *env, ppc_avr_t *r, \ ppc_avr_t *a, ppc_avr_t *b) \ @@ -838,9 +787,6 @@ void helper_vcmpne##suffix(CPUPPCState *env, ppc_avr_t *r, \ VCMPNE(zb, u8, uint8_t, 1) VCMPNE(zh, u16, uint16_t, 1) VCMPNE(zw, u32, uint32_t, 1) -VCMPNE(b, u8, uint8_t, 0) -VCMPNE(h, u16, uint16_t, 0) -VCMPNE(w, u32, uint32_t, 0) #undef VCMPNE_DO #undef VCMPNE diff --git a/target/ppc/translate/vmx-impl.c.inc b/target/ppc/translate/vmx-impl.c.inc index 99606a8718..a32ad92195 100644 --- a/target/ppc/translate/vmx-impl.c.inc +++ b/target/ppc/translate/vmx-impl.c.inc @@ -985,41 +985,76 @@ 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 t0, t1; + + t0 = tcg_temp_new_i64(); + t1 = tcg_temp_new_i64(); + + get_avr64(t0, vrt, true); + tcg_gen_ctpop_i64(t1, t0); + get_avr64(t0, vrt, false); + tcg_gen_ctpop_i64(t0, t0); + tcg_gen_add_i64(t1, t0, t1); + + tcg_gen_setcondi_i64(TCG_COND_EQ, t0, t1, 0); + tcg_gen_shli_i64(t0, t0, 1); + + tcg_gen_setcondi_i64(TCG_COND_EQ, t1, t1, 128); + tcg_gen_shli_i64(t1, t1, 3); + + tcg_gen_or_i64(t0, t0, t1); + tcg_gen_extrl_i64_i32(cpu_crf[6], t0); + + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); +} + +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); + tcg_gen_gvec_shli(vece, avr_full_offset(a->vrt), avr_full_offset(a->vrt), + (8 << vece) - 1, 16, 16); + tcg_gen_gvec_sari(vece, avr_full_offset(a->vrt), avr_full_offset(a->vrt), + (8 << vece) - 1, 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) + 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) { diff --git a/target/ppc/translate/vmx-ops.c.inc b/target/ppc/translate/vmx-ops.c.inc index 6787327f56..80d460c34e 100644 --- a/target/ppc/translate/vmx-ops.c.inc +++ b/target/ppc/translate/vmx-ops.c.inc @@ -187,19 +187,10 @@ GEN_HANDLER2_E(name, str, 0x4, opc2, opc3, 0x00000000, PPC_NONE, PPC2_ISA300), GEN_VXRFORM_300(vcmpnezb, 3, 4) GEN_VXRFORM_300(vcmpnezh, 3, 5) GEN_VXRFORM_300(vcmpnezw, 3, 6) -GEN_VXRFORM(vcmpgtsb, 3, 12) -GEN_VXRFORM(vcmpgtsh, 3, 13) -GEN_VXRFORM(vcmpgtsw, 3, 14) -GEN_VXRFORM(vcmpgtub, 3, 8) -GEN_VXRFORM(vcmpgtuh, 3, 9) -GEN_VXRFORM(vcmpgtuw, 3, 10) -GEN_VXRFORM_DUAL(vcmpeqfp, vcmpequd, 3, 3, PPC_ALTIVEC, PPC_NONE) +GEN_VXRFORM(vcmpeqfp, 3, 3) GEN_VXRFORM(vcmpgefp, 3, 7) -GEN_VXRFORM_DUAL(vcmpgtfp, vcmpgtud, 3, 11, PPC_ALTIVEC, PPC_NONE) -GEN_VXRFORM_DUAL(vcmpbfp, vcmpgtsd, 3, 15, PPC_ALTIVEC, PPC_NONE) -GEN_VXRFORM_DUAL(vcmpequb, vcmpneb, 3, 0, PPC_ALTIVEC, PPC_NONE) -GEN_VXRFORM_DUAL(vcmpequh, vcmpneh, 3, 1, PPC_ALTIVEC, PPC_NONE) -GEN_VXRFORM_DUAL(vcmpequw, vcmpnew, 3, 2, PPC_ALTIVEC, PPC_NONE) +GEN_VXRFORM(vcmpgtfp, 3, 11) +GEN_VXRFORM(vcmpbfp, 3, 15) #define GEN_VXFORM_DUAL_INV(name0, name1, opc2, opc3, inval0, inval1, type) \ GEN_OPCODE_DUAL(name0##_##name1, 0x04, opc2, opc3, inval0, inval1, type, \

[v3,09/37] target/ppc: Move Vector Compare Equal/Not Equal/Greater Than to decodetree

Commit Message

Comments

Patch