[v4,02/47] target/ppc: moved vector even and odd multiplication to decodetree

Message ID	20220222143646.1268606-3-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 v4 02/47] target/ppc: moved vector even and odd multiplication to decodetree Date: Tue, 22 Feb 2022 11:36:00 -0300 Message-Id: <20220222143646.1268606-3-matheus.ferst@eldorado.org.br> In-Reply-To: <20220222143646.1268606-1-matheus.ferst@eldorado.org.br> References: <20220222143646.1268606-1-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: -4 X-Spam_score: -0.5 X-Spam_bar: / X-Spam_report: (-0.5 / 5.0 requ) BAYES_00=-1.9, PDS_HP_HELO_NORDNS=0.659, 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: "Lucas Mateus Castro \(alqotel\)" <lucas.castro@eldorado.org.br>, danielhb413@gmail.com, richard.henderson@linaro.org, groug@kaod.org, Lucas Mateus Castro <lucas.araujo@eldorado.org.br>, 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 [v4,00/47] target/ppc: PowerISA Vector/VSX instruction batch [v4,01/47] target/ppc: Introduce TRANSFLAGS macros [v4,02/47] target/ppc: moved vector even and odd multiplication to decodetree [v4,03/47] target/ppc: Moved vector multiply high and low to decodetree [v4,04/47] target/ppc: vmulh instructions without helpers [v4,05/47] target/ppc: Implement vmsumcud instruction [v4,06/47] target/ppc: Implement vmsumudm instruction [v4,07/47] target/ppc: Move vexts[bhw]2[wd] to decodetree [v4,08/47] target/ppc: Implement vextsd2q [v4,09/47] target/ppc: Move Vector Compare Equal/Not Equal/Greater Than to decodetree [v4,10/47] target/ppc: Move Vector Compare Not Equal or Zero to decodetree [v4,11/47] target/ppc: Implement Vector Compare Equal Quadword [v4,12/47] target/ppc: Implement Vector Compare Greater Than Quadword [v4,13/47] target/ppc: Implement Vector Compare Quadword [v4,14/47] target/ppc: implement vstri[bh][lr] [v4,15/47] target/ppc: implement vclrlb [v4,16/47] target/ppc: implement vclrrb [v4,17/47] target/ppc: implement vcntmb[bhwd] [v4,18/47] target/ppc: implement vgnb [v4,19/47] target/ppc: move vs[lr][a][bhwd] to decodetree [v4,20/47] target/ppc: implement vslq [v4,21/47] target/ppc: implement vsrq [v4,22/47] target/ppc: implement vsraq [v4,23/47] target/ppc: move vrl[bhwd] to decodetree [v4,24/47] target/ppc: move vrl[bhwd]nm/vrl[bhwd]mi to decodetree [v4,25/47] target/ppc: implement vrlq [v4,26/47] target/ppc: Move vsel and vperm/vpermr to decodetree [v4,27/47] target/ppc: Move xxsel to decodetree [v4,28/47] target/ppc: move xxperm/xxpermr to decodetree [v4,29/47] target/ppc: Move xxpermdi to decodetree [v4,30/47] target/ppc: Implement xxpermx instruction [v4,31/47] tcg/tcg-op-gvec.c: Introduce tcg_gen_gvec_4i [v4,32/47] target/ppc: Implement xxeval [v4,33/47] target/ppc: Implement xxgenpcv[bhwd]m instruction [v4,34/47] target/ppc: move xs[n]madd[am][ds]p/xs[n]msub[am][ds]p to decodetree [v4,35/47] target/ppc: implement xs[n]maddqp[o]/xs[n]msubqp[o] [v4,36/47] target/ppc: Implement xvtlsbb instruction [v4,37/47] target/ppc: Remove xscmpnedp instruction [v4,38/47] target/ppc: Refactor VSX_SCALAR_CMP_DP [v4,39/47] target/ppc: Implement xscmp{eq,ge,gt}qp [v4,40/47] target/ppc: Move xscmp{eq,ge,gt}dp to decodetree [v4,41/47] target/ppc: Move xs{max, min}[cj]dp to use do_helper_XX3 [v4,42/47] target/ppc: Refactor VSX_MAX_MINC helper [v4,43/47] target/ppc: Implement xs{max,min}cqp [v4,44/47] target/ppc: Implement xvcvbf16spn and xvcvspbf16 instructions [v4,45/47] target/ppc: implement plxsd/pstxsd [v4,46/47] target/ppc: implement plxssp/pstxssp [v4,47/47] target/ppc: implement lxvr[bhwd]/stxvr[bhwd]x

diff --git a/target/ppc/helper.h b/target/ppc/helper.h index ab008c9d4e..04689522f8 100644 --- a/target/ppc/helper.h +++ b/target/ppc/helper.h @@ -190,18 +190,22 @@ DEF_HELPER_3(vmrglw, void, avr, avr, avr) DEF_HELPER_3(vmrghb, void, avr, avr, avr) DEF_HELPER_3(vmrghh, void, avr, avr, avr) DEF_HELPER_3(vmrghw, void, avr, avr, avr) -DEF_HELPER_3(vmulesb, void, avr, avr, avr) -DEF_HELPER_3(vmulesh, void, avr, avr, avr) -DEF_HELPER_3(vmulesw, void, avr, avr, avr) -DEF_HELPER_3(vmuleub, void, avr, avr, avr) -DEF_HELPER_3(vmuleuh, void, avr, avr, avr) -DEF_HELPER_3(vmuleuw, void, avr, avr, avr) -DEF_HELPER_3(vmulosb, void, avr, avr, avr) -DEF_HELPER_3(vmulosh, void, avr, avr, avr) -DEF_HELPER_3(vmulosw, void, avr, avr, avr) -DEF_HELPER_3(vmuloub, void, avr, avr, avr) -DEF_HELPER_3(vmulouh, void, avr, avr, avr) -DEF_HELPER_3(vmulouw, void, avr, avr, avr) +DEF_HELPER_3(VMULESB, void, avr, avr, avr) +DEF_HELPER_3(VMULESH, void, avr, avr, avr) +DEF_HELPER_3(VMULESW, void, avr, avr, avr) +DEF_HELPER_3(VMULESD, void, avr, avr, avr) +DEF_HELPER_3(VMULEUB, void, avr, avr, avr) +DEF_HELPER_3(VMULEUH, void, avr, avr, avr) +DEF_HELPER_3(VMULEUW, void, avr, avr, avr) +DEF_HELPER_3(VMULEUD, void, avr, avr, avr) +DEF_HELPER_3(VMULOSB, void, avr, avr, avr) +DEF_HELPER_3(VMULOSH, void, avr, avr, avr) +DEF_HELPER_3(VMULOSW, void, avr, avr, avr) +DEF_HELPER_3(VMULOSD, void, avr, avr, avr) +DEF_HELPER_3(VMULOUB, void, avr, avr, avr) +DEF_HELPER_3(VMULOUH, void, avr, avr, avr) +DEF_HELPER_3(VMULOUW, void, avr, avr, avr) +DEF_HELPER_3(VMULOUD, void, avr, avr, avr) DEF_HELPER_3(vmulhsw, void, avr, avr, avr) DEF_HELPER_3(vmulhuw, void, avr, avr, avr) DEF_HELPER_3(vmulhsd, void, avr, avr, avr) diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode index 2a9c91a423..092ea79618 100644 --- a/target/ppc/insn32.decode +++ b/target/ppc/insn32.decode @@ -440,6 +440,28 @@ VEXTRACTWM 000100 ..... 01010 ..... 11001000010 @VX_tb VEXTRACTDM 000100 ..... 01011 ..... 11001000010 @VX_tb VEXTRACTQM 000100 ..... 01100 ..... 11001000010 @VX_tb +## Vector Multiply Instruction + +VMULESB 000100 ..... ..... ..... 01100001000 @VX +VMULOSB 000100 ..... ..... ..... 00100001000 @VX +VMULEUB 000100 ..... ..... ..... 01000001000 @VX +VMULOUB 000100 ..... ..... ..... 00000001000 @VX + +VMULESH 000100 ..... ..... ..... 01101001000 @VX +VMULOSH 000100 ..... ..... ..... 00101001000 @VX +VMULEUH 000100 ..... ..... ..... 01001001000 @VX +VMULOUH 000100 ..... ..... ..... 00001001000 @VX + +VMULESW 000100 ..... ..... ..... 01110001000 @VX +VMULOSW 000100 ..... ..... ..... 00110001000 @VX +VMULEUW 000100 ..... ..... ..... 01010001000 @VX +VMULOUW 000100 ..... ..... ..... 00010001000 @VX + +VMULESD 000100 ..... ..... ..... 01111001000 @VX +VMULOSD 000100 ..... ..... ..... 00111001000 @VX +VMULEUD 000100 ..... ..... ..... 01011001000 @VX +VMULOUD 000100 ..... ..... ..... 00011001000 @VX + # VSX Load/Store Instructions LXV 111101 ..... ..... ............ . 001 @DQ_TSX diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c index d1b12788b2..7d925418d4 100644 --- a/target/ppc/int_helper.c +++ b/target/ppc/int_helper.c @@ -1063,7 +1063,7 @@ void helper_vmsumuhs(CPUPPCState *env, ppc_avr_t *r, ppc_avr_t *a, } #define VMUL_DO_EVN(name, mul_element, mul_access, prod_access, cast) \ - void helper_v##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) \ + void helper_V##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) \ { \ int i; \ \ @@ -1074,7 +1074,7 @@ void helper_vmsumuhs(CPUPPCState *env, ppc_avr_t *r, ppc_avr_t *a, } #define VMUL_DO_ODD(name, mul_element, mul_access, prod_access, cast) \ - void helper_v##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) \ + void helper_V##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) \ { \ int i; \ \ @@ -1085,17 +1085,33 @@ void helper_vmsumuhs(CPUPPCState *env, ppc_avr_t *r, ppc_avr_t *a, } #define VMUL(suffix, mul_element, mul_access, prod_access, cast) \ - VMUL_DO_EVN(mule##suffix, mul_element, mul_access, prod_access, cast) \ - VMUL_DO_ODD(mulo##suffix, mul_element, mul_access, prod_access, cast) -VMUL(sb, s8, VsrSB, VsrSH, int16_t) -VMUL(sh, s16, VsrSH, VsrSW, int32_t) -VMUL(sw, s32, VsrSW, VsrSD, int64_t) -VMUL(ub, u8, VsrB, VsrH, uint16_t) -VMUL(uh, u16, VsrH, VsrW, uint32_t) -VMUL(uw, u32, VsrW, VsrD, uint64_t) + VMUL_DO_EVN(MULE##suffix, mul_element, mul_access, prod_access, cast) \ + VMUL_DO_ODD(MULO##suffix, mul_element, mul_access, prod_access, cast) +VMUL(SB, s8, VsrSB, VsrSH, int16_t) +VMUL(SH, s16, VsrSH, VsrSW, int32_t) +VMUL(SW, s32, VsrSW, VsrSD, int64_t) +VMUL(UB, u8, VsrB, VsrH, uint16_t) +VMUL(UH, u16, VsrH, VsrW, uint32_t) +VMUL(UW, u32, VsrW, VsrD, uint64_t) #undef VMUL_DO_EVN #undef VMUL_DO_ODD #undef VMUL +void helper_VMULESD(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) +{ + muls64(&r->VsrD(1), &r->VsrD(0), a->VsrSD(0), b->VsrSD(0)); +} +void helper_VMULOSD(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) +{ + muls64(&r->VsrD(1), &r->VsrD(0), a->VsrSD(1), b->VsrSD(1)); +} +void helper_VMULEUD(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) +{ + mulu64(&r->VsrD(1), &r->VsrD(0), a->VsrD(0), b->VsrD(0)); +} +void helper_VMULOUD(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) +{ + mulu64(&r->VsrD(1), &r->VsrD(0), a->VsrD(1), b->VsrD(1)); +} void helper_vmulhsw(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) { diff --git a/target/ppc/translate/vmx-impl.c.inc b/target/ppc/translate/vmx-impl.c.inc index d5e02fd7f2..430579addd 100644 --- a/target/ppc/translate/vmx-impl.c.inc +++ b/target/ppc/translate/vmx-impl.c.inc @@ -798,29 +798,11 @@ 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); @@ -2104,6 +2086,40 @@ static bool trans_VPEXTD(DisasContext *ctx, arg_VX *a) 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; +} + +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_helper, gen_helper_VMULESD) +TRANS_FLAGS2(ISA310, VMULOSD, do_vx_helper, gen_helper_VMULOSD) +TRANS_FLAGS2(ISA310, VMULEUD, do_vx_helper, gen_helper_VMULEUD) +TRANS_FLAGS2(ISA310, VMULOUD, do_vx_helper, gen_helper_VMULOUD) + #undef GEN_VR_LDX #undef GEN_VR_STX #undef GEN_VR_LVE diff --git a/target/ppc/translate/vmx-ops.c.inc b/target/ppc/translate/vmx-ops.c.inc index 25ee715b43..f310b2fbde 100644 --- a/target/ppc/translate/vmx-ops.c.inc +++ b/target/ppc/translate/vmx-ops.c.inc @@ -101,20 +101,11 @@ GEN_VXFORM_DUAL(vmrgow, vextuwlx, 6, 26, PPC_NONE, PPC2_ALTIVEC_207), GEN_VXFORM_300(vextubrx, 6, 28), GEN_VXFORM_300(vextuhrx, 6, 29), GEN_VXFORM_DUAL(vmrgew, vextuwrx, 6, 30, PPC_NONE, PPC2_ALTIVEC_207), -GEN_VXFORM(vmuloub, 4, 0), -GEN_VXFORM(vmulouh, 4, 1), -GEN_VXFORM_DUAL(vmulouw, vmuluwm, 4, 2, PPC_ALTIVEC, PPC_NONE), -GEN_VXFORM(vmulosb, 4, 4), -GEN_VXFORM(vmulosh, 4, 5), -GEN_VXFORM_207(vmulosw, 4, 6), +GEN_VXFORM_207(vmuluwm, 4, 2), GEN_VXFORM_310(vmulld, 4, 7), -GEN_VXFORM(vmuleub, 4, 8), -GEN_VXFORM(vmuleuh, 4, 9), -GEN_VXFORM_DUAL(vmuleuw, vmulhuw, 4, 10, PPC_ALTIVEC, PPC_NONE), +GEN_VXFORM_310(vmulhuw, 4, 10), GEN_VXFORM_310(vmulhud, 4, 11), -GEN_VXFORM(vmulesb, 4, 12), -GEN_VXFORM(vmulesh, 4, 13), -GEN_VXFORM_DUAL(vmulesw, vmulhsw, 4, 14, PPC_ALTIVEC, PPC_NONE), +GEN_VXFORM_310(vmulhsw, 4, 14), GEN_VXFORM_310(vmulhsd, 4, 15), GEN_VXFORM(vslb, 2, 4), GEN_VXFORM(vslh, 2, 5), diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index dea24f23c4..69d22e08cb 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -3987,3 +3987,9 @@ void tcg_register_jit(const void *buf, size_t buf_size) tcg_register_jit_int(buf, buf_size, &debug_frame, sizeof(debug_frame)); } #endif /* __ELF__ */ +#undef VMULEUB +#undef VMULEUH +#undef VMULEUW +#undef VMULOUB +#undef VMULOUH +#undef VMULOUW

[v4,02/47] target/ppc: moved vector even and odd multiplication to decodetree

Commit Message

Comments

Patch