From patchwork Mon Jul 1 11:24:45 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jan Beulich X-Patchwork-Id: 11025659 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 7B8E01510 for ; Mon, 1 Jul 2019 11:26:56 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 6E9ED205FD for ; Mon, 1 Jul 2019 11:26:56 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 631432866D; Mon, 1 Jul 2019 11:26:56 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.2 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_MED,UNPARSEABLE_RELAY autolearn=ham version=3.3.1 Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id BB4D3205FD for ; Mon, 1 Jul 2019 11:26:55 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1hhuRE-00029R-Iz; Mon, 01 Jul 2019 11:25:36 +0000 Received: from us1-rack-dfw2.inumbo.com ([104.130.134.6]) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1hhuRD-00029J-6A for xen-devel@lists.xenproject.org; Mon, 01 Jul 2019 11:25:35 +0000 X-Inumbo-ID: f0940c7f-9bf2-11e9-8980-bc764e045a96 Received: from m4a0039g.houston.softwaregrp.com (unknown [15.124.2.85]) by us1-rack-dfw2.inumbo.com (Halon) with ESMTPS id f0940c7f-9bf2-11e9-8980-bc764e045a96; Mon, 01 Jul 2019 11:25:33 +0000 (UTC) Received: FROM m4a0039g.houston.softwaregrp.com (15.120.17.146) BY m4a0039g.houston.softwaregrp.com WITH ESMTP; Mon, 1 Jul 2019 11:21:58 +0000 Received: from M9W0067.microfocus.com (2002:f79:be::f79:be) by M4W0334.microfocus.com (2002:f78:1192::f78:1192) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.1.1591.10; Mon, 1 Jul 2019 11:24:47 +0000 Received: from NAM05-BY2-obe.outbound.protection.outlook.com (15.124.72.13) by M9W0067.microfocus.com (15.121.0.190) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.1.1591.10 via Frontend Transport; Mon, 1 Jul 2019 11:24:46 +0000 Received: from BY5PR18MB3394.namprd18.prod.outlook.com (10.255.139.95) by BY5PR18MB3299.namprd18.prod.outlook.com (10.255.138.96) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.2032.18; Mon, 1 Jul 2019 11:24:45 +0000 Received: from BY5PR18MB3394.namprd18.prod.outlook.com ([fe80::2005:4b02:1d60:d1bc]) by BY5PR18MB3394.namprd18.prod.outlook.com ([fe80::2005:4b02:1d60:d1bc%3]) with mapi id 15.20.2008.020; Mon, 1 Jul 2019 11:24:45 +0000 From: Jan Beulich To: "xen-devel@lists.xenproject.org" Thread-Topic: [PATCH v9 16/23] x86emul: support AVX512_VNNI insns Thread-Index: AQHVL/+VpPCmg3T9Z0y65tMD3Dz8fw== Date: Mon, 1 Jul 2019 11:24:45 +0000 Message-ID: <2ee78b22-c58f-f393-e83d-e9be8e3e4a5c@suse.com> References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-clientproxiedby: DB8PR04CA0030.eurprd04.prod.outlook.com (2603:10a6:10:110::40) To BY5PR18MB3394.namprd18.prod.outlook.com (2603:10b6:a03:194::31) authentication-results: spf=none (sender IP is ) smtp.mailfrom=JBeulich@suse.com; x-ms-exchange-messagesentrepresentingtype: 1 x-originating-ip: [87.234.252.170] x-ms-publictraffictype: Email x-ms-office365-filtering-correlation-id: da9ee704-81c9-444f-e739-08d6fe16b806 x-microsoft-antispam: BCL:0; PCL:0; RULEID:(2390118)(7020095)(4652040)(8989299)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(5600148)(711020)(4605104)(1401327)(2017052603328)(7193020); SRVR:BY5PR18MB3299; x-ms-traffictypediagnostic: BY5PR18MB3299: x-microsoft-antispam-prvs: x-ms-oob-tlc-oobclassifiers: OLM:1850; x-forefront-prvs: 00851CA28B x-forefront-antispam-report: SFV:NSPM; SFS:(10019020)(4636009)(346002)(136003)(396003)(366004)(376002)(39860400002)(189003)(199004)(6916009)(478600001)(5660300002)(256004)(14444005)(76176011)(486006)(7736002)(80792005)(8676002)(4326008)(2501003)(66556008)(64756008)(66946007)(2906002)(66476007)(71190400001)(8936002)(71200400001)(66446008)(25786009)(73956011)(3846002)(446003)(68736007)(36756003)(81166006)(81156014)(66066001)(14454004)(6116002)(31686004)(11346002)(99286004)(305945005)(86362001)(31696002)(6436002)(386003)(6506007)(53936002)(102836004)(2616005)(54906003)(52116002)(6512007)(72206003)(316002)(476003)(186003)(6486002)(5640700003)(26005)(2351001); DIR:OUT; SFP:1102; SCL:1; SRVR:BY5PR18MB3299; H:BY5PR18MB3394.namprd18.prod.outlook.com; FPR:; SPF:None; LANG:en; PTR:InfoNoRecords; MX:1; A:1; received-spf: None (protection.outlook.com: suse.com does not designate permitted sender hosts) x-ms-exchange-senderadcheck: 1 x-microsoft-antispam-message-info: 6zX2nu+ga1Z6rTu2wMIed1E3jgMFsWDC5/ImTFJTXk2RZwc3Ys4OjejXa0fgRvUPFfeXj3d8Q0p2Yg5KxVeviGN8XW6dBmPbnTh7pdv6BW6DFOAjukZgxI+Hsahdsjx0Qp3dDPmyDtBfrvsr+WaJJzF9hQ3TPIu3oHYlNk9FEUMPP3Oy0kqSM8A2Ou2ECezwGd0aXCuRoD2AUKPX0KMctazO+KTRsU74xalAA+Wm56//hukaiZFIpJXUjFMNvRjyDELuFBF3orgTBZSb4l/cNQCXYrKxlUT0OvSsL3c3401D2hqHJL53Y3fMYeXx24UpCJhyIqxoRJOgBRJ0YLdNAv4SIShOHzBfukJ4Q5jSJWcWt+bC3oo/ojmjs6JtZE10ajKH3ky1Mbf+Zp3C32hq7FNhzgLyVjRHky+Ja5bGE5Q= Content-ID: <25F4A4BBE34A3E4CBF21F6865E24AEF3@namprd18.prod.outlook.com> MIME-Version: 1.0 X-MS-Exchange-CrossTenant-Network-Message-Id: da9ee704-81c9-444f-e739-08d6fe16b806 X-MS-Exchange-CrossTenant-originalarrivaltime: 01 Jul 2019 11:24:45.1878 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: 856b813c-16e5-49a5-85ec-6f081e13b527 X-MS-Exchange-CrossTenant-mailboxtype: HOSTED X-MS-Exchange-CrossTenant-userprincipalname: JBeulich@suse.com X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY5PR18MB3299 X-OriginatorOrg: suse.com Subject: [Xen-devel] [PATCH v9 16/23] x86emul: support AVX512_VNNI insns X-BeenThere: xen-devel@lists.xenproject.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Cc: Andrew Cooper , Wei Liu , RogerPau Monne Errors-To: xen-devel-bounces@lists.xenproject.org Sender: "Xen-devel" X-Virus-Scanned: ClamAV using ClamSMTP Along the lines of the 4FMAPS case, convert the 4VNNIW-based table entries to a decoder adjustment. Because of the current sharing of table entries between different (implied) opcode prefixes and with the same major opcodes being used for vp4dpwssd{,s}, which have a different memory operand size and different Disp8 scaling, the pre-existing table entries get converted to a decoder override. The table entries will now represent the insns here, in line with other table entries preferably representing the prefix-66 insns. As in a few cases before, since the insns here and in particular their memory access patterns follow the usual scheme, I didn't think it was necessary to add a contrived test specifically for them, beyond the Disp8 scaling one. Signed-off-by: Jan Beulich Acked-by: Andrew Cooper --- v9: Re-base. Explain need for decoder special case. v8: Re-base. v7: New. --- a/tools/tests/x86_emulator/evex-disp8.c +++ b/tools/tests/x86_emulator/evex-disp8.c @@ -580,6 +580,13 @@ static const struct test avx512_vbmi2_al INSN(pshrdw, 66, 0f3a, 72, vl, w, vl), }; +static const struct test avx512_vnni_all[] = { + INSN(pdpbusd, 66, 0f38, 50, vl, d, vl), + INSN(pdpbusds, 66, 0f38, 51, vl, d, vl), + INSN(pdpwssd, 66, 0f38, 52, vl, d, vl), + INSN(pdpwssds, 66, 0f38, 53, vl, d, vl), +}; + static const struct test avx512_vpopcntdq_all[] = { INSN(popcnt, 66, 0f38, 55, vl, dq, vl) }; @@ -959,5 +966,6 @@ void evex_disp8_test(void *instr, struct RUN(avx512_ifma, all); RUN(avx512_vbmi, all); RUN(avx512_vbmi2, all); + RUN(avx512_vnni, all); RUN(avx512_vpopcntdq, all); } --- a/tools/tests/x86_emulator/x86-emulate.h +++ b/tools/tests/x86_emulator/x86-emulate.h @@ -144,6 +144,7 @@ static inline bool xcr0_mask(uint64_t ma #define cpu_has_avx512vl (cp.feat.avx512vl && xcr0_mask(0xe6)) #define cpu_has_avx512_vbmi (cp.feat.avx512_vbmi && xcr0_mask(0xe6)) #define cpu_has_avx512_vbmi2 (cp.feat.avx512_vbmi2 && xcr0_mask(0xe6)) +#define cpu_has_avx512_vnni (cp.feat.avx512_vnni && xcr0_mask(0xe6)) #define cpu_has_avx512_bitalg (cp.feat.avx512_bitalg && xcr0_mask(0xe6)) #define cpu_has_avx512_vpopcntdq (cp.feat.avx512_vpopcntdq && xcr0_mask(0xe6)) #define cpu_has_avx512_4vnniw (cp.feat.avx512_4vnniw && xcr0_mask(0xe6)) --- a/xen/arch/x86/x86_emulate/x86_emulate.c +++ b/xen/arch/x86/x86_emulate/x86_emulate.c @@ -479,7 +479,7 @@ static const struct ext0f38_table { [0x4d] = { .simd_size = simd_scalar_vexw, .d8s = d8s_dq }, [0x4e] = { .simd_size = simd_packed_fp, .two_op = 1, .d8s = d8s_vl }, [0x4f] = { .simd_size = simd_scalar_vexw, .d8s = d8s_dq }, - [0x52 ... 0x53] = { .simd_size = simd_128, .d8s = 4 }, + [0x50 ... 0x53] = { .simd_size = simd_packed_int, .d8s = d8s_vl }, [0x54 ... 0x55] = { .simd_size = simd_packed_int, .two_op = 1, .d8s = d8s_vl }, [0x58] = { .simd_size = simd_other, .two_op = 1, .d8s = 2 }, [0x59] = { .simd_size = simd_other, .two_op = 1, .d8s = 3 }, @@ -1890,6 +1890,7 @@ in_protmode( #define vcpu_has_avx512vl() (ctxt->cpuid->feat.avx512vl) #define vcpu_has_avx512_vbmi() (ctxt->cpuid->feat.avx512_vbmi) #define vcpu_has_avx512_vbmi2() (ctxt->cpuid->feat.avx512_vbmi2) +#define vcpu_has_avx512_vnni() (ctxt->cpuid->feat.avx512_vnni) #define vcpu_has_avx512_bitalg() (ctxt->cpuid->feat.avx512_bitalg) #define vcpu_has_avx512_vpopcntdq() (ctxt->cpuid->feat.avx512_vpopcntdq) #define vcpu_has_rdpid() (ctxt->cpuid->feat.rdpid) @@ -3179,6 +3180,8 @@ x86_decode( switch ( b ) { + /* vp4dpwssd{,s} need special casing */ + case 0x52: case 0x53: /* v4f{,n}madd{p,s}s need special casing */ case 0x9a: case 0x9b: case 0xaa: case 0xab: if ( evex.pfx == vex_f2 ) @@ -9394,6 +9397,14 @@ x86_emulate( avx512_vlen_check(true); goto simd_zmm; + case X86EMUL_OPC_EVEX_66(0x0f38, 0x50): /* vpdpbusd [xyz]mm/mem,[xyz]mm,[xyz]mm{k} */ + case X86EMUL_OPC_EVEX_66(0x0f38, 0x51): /* vpdpbusds [xyz]mm/mem,[xyz]mm,[xyz]mm{k} */ + case X86EMUL_OPC_EVEX_66(0x0f38, 0x52): /* vpdpwssd [xyz]mm/mem,[xyz]mm,[xyz]mm{k} */ + case X86EMUL_OPC_EVEX_66(0x0f38, 0x53): /* vpdpwssds [xyz]mm/mem,[xyz]mm,[xyz]mm{k} */ + host_and_vcpu_must_have(avx512_vnni); + generate_exception_if(evex.w, EXC_UD); + goto avx512f_no_sae; + case X86EMUL_OPC_EVEX_F2(0x0f38, 0x9a): /* v4fmaddps m128,zmm+3,zmm{k} */ case X86EMUL_OPC_EVEX_F2(0x0f38, 0xaa): /* v4fnmaddps m128,zmm+3,zmm{k} */ host_and_vcpu_must_have(avx512_4fmaps); --- a/xen/include/asm-x86/cpufeature.h +++ b/xen/include/asm-x86/cpufeature.h @@ -111,6 +111,7 @@ /* CPUID level 0x00000007:0.ecx */ #define cpu_has_avx512_vbmi boot_cpu_has(X86_FEATURE_AVX512_VBMI) #define cpu_has_avx512_vbmi2 boot_cpu_has(X86_FEATURE_AVX512_VBMI2) +#define cpu_has_avx512_vnni boot_cpu_has(X86_FEATURE_AVX512_VNNI) #define cpu_has_avx512_bitalg boot_cpu_has(X86_FEATURE_AVX512_BITALG) #define cpu_has_avx512_vpopcntdq boot_cpu_has(X86_FEATURE_AVX512_VPOPCNTDQ) #define cpu_has_rdpid boot_cpu_has(X86_FEATURE_RDPID) --- a/xen/include/public/arch-x86/cpufeatureset.h +++ b/xen/include/public/arch-x86/cpufeatureset.h @@ -229,6 +229,7 @@ XEN_CPUFEATURE(UMIP, 6*32+ 2) / XEN_CPUFEATURE(PKU, 6*32+ 3) /*H Protection Keys for Userspace */ XEN_CPUFEATURE(OSPKE, 6*32+ 4) /*! OS Protection Keys Enable */ XEN_CPUFEATURE(AVX512_VBMI2, 6*32+ 6) /*A Additional AVX-512 Vector Byte Manipulation Instrs */ +XEN_CPUFEATURE(AVX512_VNNI, 6*32+11) /*A Vector Neural Network Instrs */ XEN_CPUFEATURE(AVX512_BITALG, 6*32+12) /*A Support for VPOPCNT[B,W] and VPSHUFBITQMB */ XEN_CPUFEATURE(AVX512_VPOPCNTDQ, 6*32+14) /*A POPCNT for vectors of DW/QW */ XEN_CPUFEATURE(RDPID, 6*32+22) /*A RDPID instruction */ --- a/xen/tools/gen-cpuid.py +++ b/xen/tools/gen-cpuid.py @@ -263,7 +263,7 @@ def crunch_numbers(state): # AVX512 features are built on top of AVX512F AVX512F: [AVX512DQ, AVX512_IFMA, AVX512PF, AVX512ER, AVX512CD, AVX512BW, AVX512VL, AVX512_4VNNIW, AVX512_4FMAPS, - AVX512_VPOPCNTDQ], + AVX512_VNNI, AVX512_VPOPCNTDQ], # AVX512 extensions acting on vectors of bytes/words are made # dependents of AVX512BW (as to requiring wider than 16-bit mask