From patchwork Thu Oct 27 02:00:31 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695202 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=mSd9dlIF; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTVL1byCz23kp for ; Thu, 27 Oct 2022 13:03:19 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsCw-0001dW-A0; Wed, 26 Oct 2022 22:01:22 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCb-0001W4-AU for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:03 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCY-0004sz-Af for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:00 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836058; x=1698372058; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=UyZWp1kELLoxMAnHvZybWFAVmfkD9Fko8h5Vk+D9uTQ=; b=mSd9dlIFHIY/Pn77A98j71/kq77o3lsxb/v1PveQ/HfI6kovex8ycFEZ OLHjbWLGXLIE3yVtCcAAhovHMci/q6Gz2fe4lHLNsidM/vFHVDk6p+I51 ZF5+kP8+katiGJYwPTC1wdQN5CW56AQSiIcvqma+kQxOIZwWc936SaeRL bpa2m4F4RB6mJEB+sWX0ktrlRadmvJhzBZSPtbQrI2uz9lROvJwAj0VGy LE3GBg9xzPVrk8hV41JwRprxbhR3uYQgaP5kzh+WBGt9PjBPScd1jWp8e xEl0CZF7jwP5yrx+jRDHEgyUbC07LQdzkYV6IKyx6fW5T13kcca8rHRVR A==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174869" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174869" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407928" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407928" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:37 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 1/6] i386: Introduce FeatureWordInfo for AMX CPUID leaf 0x1D and 0x1E Date: Wed, 26 Oct 2022 19:00:31 -0700 Message-Id: <20221027020036.373140-2-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, UPPERCASE_50_75=0.008 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org CPUID leaf 0x1D and 0x1E enumerate tile and TMUL information for AMX. Introduce FeatureWord FEAT_1D_1_EAX, FEAT_1D_1_EBX, FEAT_1D_1_ECX and FEAT_1E_0_EBX. Thus these features of AMX can be expanded when "-cpu host/max" and can be configured in named CPU model. Signed-off-by: Wang, Lei --- target/i386/cpu.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++ target/i386/cpu.h | 12 +++++++++++ 2 files changed, 67 insertions(+) diff --git a/target/i386/cpu.c b/target/i386/cpu.c index 8a11470507..e98780773c 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -1002,6 +1002,45 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = { }, .tcg_features = ~0U, }, + [FEAT_1D_1_EAX] = { + .type = CPUID_FEATURE_WORD, + .cpuid = { + .eax = 0x1D, + .needs_ecx = true, .ecx = 1, + .reg = R_EAX, + }, + .migratable_flags = CPUID_AMX_PALETTE_1_TOTAL_TILE_BYTES_MASK | + CPUID_AMX_PALETTE_1_BYTES_PER_TILE_MASK, + }, + [FEAT_1D_1_EBX] = { + .type = CPUID_FEATURE_WORD, + .cpuid = { + .eax = 0x1D, + .needs_ecx = true, .ecx = 1, + .reg = R_EBX, + }, + .migratable_flags = CPUID_AMX_PALETTE_1_BYTES_PER_ROW_MASK | + CPUID_AMX_PALETTE_1_MAX_NAMES_MASK, + }, + [FEAT_1D_1_ECX] = { + .type = CPUID_FEATURE_WORD, + .cpuid = { + .eax = 0x1D, + .needs_ecx = true, .ecx = 1, + .reg = R_ECX, + }, + .migratable_flags = CPUID_AMX_PALETTE_1_MAX_ROWS_MASK, + }, + [FEAT_1E_0_EBX] = { + .type = CPUID_FEATURE_WORD, + .cpuid = { + .eax = 0x1E, + .needs_ecx = true, .ecx = 0, + .reg = R_EBX, + }, + .migratable_flags = CPUID_AMX_TMUL_MAX_K_MASK | + CPUID_AMX_TMUL_MAX_N_MASK, + }, /*Below are MSR exposed features*/ [FEAT_ARCH_CAPABILITIES] = { .type = MSR_FEATURE_WORD, @@ -1371,6 +1410,22 @@ static FeatureDep feature_dependencies[] = { .from = { FEAT_7_0_EBX, CPUID_7_0_EBX_INTEL_PT }, .to = { FEAT_14_0_ECX, ~0ull }, }, + { + .from = { FEAT_7_0_EDX, CPUID_7_0_EDX_AMX_TILE }, + .to = { FEAT_1D_1_EAX, ~0ull }, + }, + { + .from = { FEAT_7_0_EDX, CPUID_7_0_EDX_AMX_TILE }, + .to = { FEAT_1D_1_EBX, ~0ull }, + }, + { + .from = { FEAT_7_0_EDX, CPUID_7_0_EDX_AMX_TILE }, + .to = { FEAT_1D_1_ECX, ~0ull }, + }, + { + .from = { FEAT_7_0_EDX, CPUID_7_0_EDX_AMX_TILE }, + .to = { FEAT_1E_0_EBX, ~0ull }, + }, { .from = { FEAT_8000_0001_EDX, CPUID_EXT2_RDTSCP }, .to = { FEAT_VMX_SECONDARY_CTLS, VMX_SECONDARY_EXEC_RDTSCP }, diff --git a/target/i386/cpu.h b/target/i386/cpu.h index 7edf5dfac3..1c90fb6c9d 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -583,6 +583,14 @@ typedef enum X86Seg { XSTATE_Hi16_ZMM_MASK | XSTATE_PKRU_MASK | \ XSTATE_XTILE_CFG_MASK | XSTATE_XTILE_DATA_MASK) +#define CPUID_AMX_PALETTE_1_TOTAL_TILE_BYTES_MASK 0xffffU +#define CPUID_AMX_PALETTE_1_BYTES_PER_TILE_MASK (0xffffU << 16) +#define CPUID_AMX_PALETTE_1_BYTES_PER_ROW_MASK 0xffffU +#define CPUID_AMX_PALETTE_1_MAX_NAMES_MASK (0xffffU << 16) +#define CPUID_AMX_PALETTE_1_MAX_ROWS_MASK 0xffffU +#define CPUID_AMX_TMUL_MAX_K_MASK 0xffU +#define CPUID_AMX_TMUL_MAX_N_MASK (0xffffU << 8) + /* CPUID feature words */ typedef enum FeatureWord { FEAT_1_EDX, /* CPUID[1].EDX */ @@ -603,6 +611,10 @@ typedef enum FeatureWord { FEAT_6_EAX, /* CPUID[6].EAX */ FEAT_XSAVE_XCR0_LO, /* CPUID[EAX=0xd,ECX=0].EAX */ FEAT_XSAVE_XCR0_HI, /* CPUID[EAX=0xd,ECX=0].EDX */ + FEAT_1D_1_EAX, /* CPUID[EAX=0x1d,ECX=1].EAX */ + FEAT_1D_1_EBX, /* CPUID[EAX=0x1d,ECX=1].EBX */ + FEAT_1D_1_ECX, /* CPUID[EAX=0x1d,ECX=1].ECX */ + FEAT_1E_0_EBX, /* CPUID[EAX=0x1e,ECX=0].EBX */ FEAT_ARCH_CAPABILITIES, FEAT_CORE_CAPABILITY, FEAT_PERF_CAPABILITIES, From patchwork Thu Oct 27 02:00:32 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695206 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=em3OX3i+; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTdG4xSlz1ygr for ; Thu, 27 Oct 2022 13:09:22 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsCx-0001dr-3Y; Wed, 26 Oct 2022 22:01:23 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCc-0001X1-OS for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:03 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCZ-0004u8-M0 for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:02 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836059; x=1698372059; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=lXnOrnrHZe3OJ6BeC65+VIIy20/ZTCijpfch6HkmVvo=; b=em3OX3i+xIM5vjfMxN0AOqZ20FCJ5xg6BaFFXCkfg9q4CWhUi533A81q hQfDN7P+A4zxRdKX39763Ngaab2tL496HGk3Qm6fb0td2wrS604MfKq/c RFBBL1SU4XGJaLuawsm34R8Ja996rfZYynLifPz1EQD1RSsRW1Kqdycq4 +4m2aaijnFs1Qmgm9714iXNskHM8xZqntAglXTsW9SdxUXQNSjaCaPTbR qah3yjsfMfsDuZE6SIGBLyWLn7EYLG/OCe6Pk1YEQL7J6oW3Plz/mKTgN Lk5AjiO2NQfaWDGydsLHPGkm1k+iQ6fTNF804i01i2oDpKMUWScZ10SvG Q==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174870" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174870" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407931" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407931" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:37 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 2/6] i386: Remove unused parameter "uint32_t bit" in feature_word_description() Date: Wed, 26 Oct 2022 19:00:32 -0700 Message-Id: <20221027020036.373140-3-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Parameter "uint32_t bit" is not used in function feature_word_description(), so remove it. Signed-off-by: Wang, Lei --- target/i386/cpu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/target/i386/cpu.c b/target/i386/cpu.c index e98780773c..0083a2a7f7 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -4290,7 +4290,7 @@ static const TypeInfo max_x86_cpu_type_info = { .class_init = max_x86_cpu_class_init, }; -static char *feature_word_description(FeatureWordInfo *f, uint32_t bit) +static char *feature_word_description(FeatureWordInfo *f) { assert(f->type == CPUID_FEATURE_WORD || f->type == MSR_FEATURE_WORD); @@ -4329,6 +4329,7 @@ static void mark_unavailable_features(X86CPU *cpu, FeatureWord w, uint64_t mask, CPUX86State *env = &cpu->env; FeatureWordInfo *f = &feature_word_info[w]; int i; + g_autofree char *feat_word_str = feature_word_description(f); if (!cpu->force_features) { env->features[w] &= ~mask; @@ -4341,7 +4342,6 @@ static void mark_unavailable_features(X86CPU *cpu, FeatureWord w, uint64_t mask, for (i = 0; i < 64; ++i) { if ((1ULL << i) & mask) { - g_autofree char *feat_word_str = feature_word_description(f, i); warn_report("%s: %s%s%s [bit %d]", verbose_prefix, feat_word_str, From patchwork Thu Oct 27 02:00:33 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695204 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=hNQknoUv; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTcC2fDkz1ygr for ; Thu, 27 Oct 2022 13:08:25 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsCz-0001ej-35; Wed, 26 Oct 2022 22:01:25 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCd-0001Xv-Q3 for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:07 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCb-0004sz-MT for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:03 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836061; x=1698372061; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=r90dV8iF2ORg/n2SN9spjUCoFBK8wGi5ozVritK4IwE=; b=hNQknoUvHuV6BbCZpG6ffqXITX9HqrDI53x8FyePXt2AVp8Zm4Go+BUx B1x/4FAwVg04QCRkX2qmyd9obb+bRxRbABVdF3g/hRqHX8Y9688/DikGS i4Y8XzzCfO5zewlGXNxDVl0wnQDFF4ihTDuFTVwvHi5DVFO64nMukXBSA x2inF7ag+Z1shtrGFsAxUbCgeoQXnu5o+08u3c1ae2u/qOEOl97ytRvty FOmGWsbLvcmB0QTJZiS2YdBk1aUvswlIZ1U2NDxO90x7jpdI8cTVjOpaw csRJOYK5bc8U0pPUC8ztgL+EahRPxfAP8Jok+RIrmpjuGXff8eusI59WC A==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174871" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174871" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407935" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407935" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:37 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 3/6] i386: Introduce new struct "MultiBitFeatureInfo" for multi-bit features Date: Wed, 26 Oct 2022 19:00:33 -0700 Message-Id: <20221027020036.373140-4-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Some features use multiple CPUID bits to form a value to be used, e.g., CPUID(0x1E,0):EBX[23:08] is regarded as the tmul_maxn value for AMX. Introduce a new struct "MultiBitFeatureInfo" to hold the information for those features and create a corresponding member in struct FeatureWordInfo, so that the infomation can be assigned for each item in feature_word_info array and used in the future. Signed-off-by: Wang, Lei --- target/i386/cpu-internal.h | 9 +++++++ target/i386/cpu.c | 54 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 63 insertions(+) diff --git a/target/i386/cpu-internal.h b/target/i386/cpu-internal.h index 9baac5c0b4..66b3d66cb4 100644 --- a/target/i386/cpu-internal.h +++ b/target/i386/cpu-internal.h @@ -25,6 +25,13 @@ typedef enum FeatureWordType { MSR_FEATURE_WORD, } FeatureWordType; +typedef struct MultiBitFeatureInfo { + const char *feat_name; + uint64_t mask; + unsigned high_bit_position; + unsigned low_bit_position; +} MultiBitFeatureInfo; + typedef struct FeatureWordInfo { FeatureWordType type; /* feature flags names are taken from "Intel Processor Identification and @@ -51,6 +58,8 @@ typedef struct FeatureWordInfo { uint64_t migratable_flags; /* Feature flags known to be migratable */ /* Features that shouldn't be auto-enabled by "-cpu host" */ uint64_t no_autoenable_flags; + unsigned num_multi_bit_features; + MultiBitFeatureInfo *multi_bit_features; } FeatureWordInfo; extern FeatureWordInfo feature_word_info[]; diff --git a/target/i386/cpu.c b/target/i386/cpu.c index 0083a2a7f7..7ae232ab18 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -1011,6 +1011,21 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = { }, .migratable_flags = CPUID_AMX_PALETTE_1_TOTAL_TILE_BYTES_MASK | CPUID_AMX_PALETTE_1_BYTES_PER_TILE_MASK, + .num_multi_bit_features = 2, + .multi_bit_features = (MultiBitFeatureInfo[]){ + { + .feat_name = "total_tile_bytes", + .mask = CPUID_AMX_PALETTE_1_TOTAL_TILE_BYTES_MASK, + .high_bit_position = 15, + .low_bit_position = 0, + }, + { + .feat_name = "bytes_per_tile", + .mask = CPUID_AMX_PALETTE_1_BYTES_PER_TILE_MASK, + .high_bit_position = 31, + .low_bit_position = 16, + }, + }, }, [FEAT_1D_1_EBX] = { .type = CPUID_FEATURE_WORD, @@ -1021,6 +1036,21 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = { }, .migratable_flags = CPUID_AMX_PALETTE_1_BYTES_PER_ROW_MASK | CPUID_AMX_PALETTE_1_MAX_NAMES_MASK, + .num_multi_bit_features = 2, + .multi_bit_features = (MultiBitFeatureInfo[]){ + { + .feat_name = "bytes_per_row", + .mask = CPUID_AMX_PALETTE_1_BYTES_PER_ROW_MASK, + .high_bit_position = 15, + .low_bit_position = 0, + }, + { + .feat_name = "max_names", + .mask = CPUID_AMX_PALETTE_1_MAX_NAMES_MASK, + .high_bit_position = 31, + .low_bit_position = 16, + }, + }, }, [FEAT_1D_1_ECX] = { .type = CPUID_FEATURE_WORD, @@ -1030,6 +1060,15 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = { .reg = R_ECX, }, .migratable_flags = CPUID_AMX_PALETTE_1_MAX_ROWS_MASK, + .num_multi_bit_features = 1, + .multi_bit_features = (MultiBitFeatureInfo[]){ + { + .feat_name = "max_rows", + .mask = CPUID_AMX_PALETTE_1_MAX_ROWS_MASK, + .high_bit_position = 15, + .low_bit_position = 0, + }, + }, }, [FEAT_1E_0_EBX] = { .type = CPUID_FEATURE_WORD, @@ -1040,6 +1079,21 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = { }, .migratable_flags = CPUID_AMX_TMUL_MAX_K_MASK | CPUID_AMX_TMUL_MAX_N_MASK, + .num_multi_bit_features = 2, + .multi_bit_features = (MultiBitFeatureInfo[]){ + { + .feat_name = "tmul_maxk", + .mask = CPUID_AMX_TMUL_MAX_K_MASK, + .high_bit_position = 7, + .low_bit_position = 0, + }, + { + .feat_name = "tmul_maxn", + .mask = CPUID_AMX_TMUL_MAX_N_MASK, + .high_bit_position = 23, + .low_bit_position = 8, + }, + }, }, /*Below are MSR exposed features*/ [FEAT_ARCH_CAPABILITIES] = { From patchwork Thu Oct 27 02:00:34 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695207 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=ijf2/mRE; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTdt44Zvz1ygr for ; Thu, 27 Oct 2022 13:09:54 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsCx-0001eO-M5; Wed, 26 Oct 2022 22:01:23 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCh-0001Zc-37 for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:09 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCc-0004uO-AY for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:05 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836062; x=1698372062; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=F9Rn29zR/Z9XakMroorv4VOXbXgVF+MyPK2naYDp0yY=; b=ijf2/mREEGpo4Zm2FCf6jbE0An2lXRgiW8eZ78USoESal5R7ogETPOAl q/2hMPArsl5yDjfHxrRl3kUaQjwdCOjQ7BSO/UxFLYP55UcrQkmpilsKo lbRkAW41RhsVefGMgRyrZwZmIAuZd0o3jMAKcc4pMvLk2mF6ozE0ljZ3i QXmh8g98SNpt9l/9oCqh7N39GLS7NaXFNBKezN8jJaxD2sPU83uCaVE5T s1yNoUzIaQw2d1NGAZFDqD9YZSXgnNe3JQZRs32ivW/b08gjT2n+Rojzs PhBTLTZC2Skg9CBqQmbuboUo6ni3xdjNV7riKIkIcKmHlZzhlzpHYp4yv A==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174872" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174872" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407939" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407939" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 4/6] i386: Mask and report unavailable multi-bit feature values Date: Wed, 26 Oct 2022 19:00:34 -0700 Message-Id: <20221027020036.373140-5-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Some feature words, e.g., feature words in AMX-related CPUID leaf 0x1D and 0x1E are not bit-wise but multiple bits represents one value. Handle this situation when the values specified are not the same as which are reported by KVM. The handling includes: - The responsibility of masking bits and giving warnings are delegated to the feature enabler. A framwork is also provided to enable this. - To simplify the initialization, a default function is provided if the the function is not specified. The reason why delegating this responsibility rather than just marking them as zeros when they are not same is because different multi-bit features may have different logic, which is case by case, for example: 1. CPUID.0x14_0x1:EBX[15:0]. Even though it's multi-bits field, it's a bitmap and each bit represents a separate capability. 2. CPUID.0x14_0x1:EAX[2:0] represents the number of configurable Address Ranges. 3 bits as a whole to represent a integer value. It means the maximum capability of HW. If KVM reports M, then M to 0 is legal value to configure (because KVM can emulate each value correctly). 3. CPUID.0x1D_0x1:EAX[31:16] represents palette 1 bytes_per_tile. 16 bits as a whole represent an integer value. It's not like case 2 and SW needs to configure the same value as reported. Because it's not possible for SW to configure to a different value and KVM cannot emulate it. So marking them blindly as zeros is incorrect, and delegating this responsibility can let each multi-bit feature have its own way to mask bits. Signed-off-by: Wang, Lei --- target/i386/cpu-internal.h | 2 ++ target/i386/cpu.c | 39 ++++++++++++++++++++++++++++++++++---- target/i386/cpu.h | 2 ++ 3 files changed, 39 insertions(+), 4 deletions(-) diff --git a/target/i386/cpu-internal.h b/target/i386/cpu-internal.h index 66b3d66cb4..f973046b4e 100644 --- a/target/i386/cpu-internal.h +++ b/target/i386/cpu-internal.h @@ -30,6 +30,8 @@ typedef struct MultiBitFeatureInfo { uint64_t mask; unsigned high_bit_position; unsigned low_bit_position; + void (*mark_unavailable_bits)(X86CPU *cpu, FeatureWord w, int index, + const char *verbose_prefix); } MultiBitFeatureInfo; typedef struct FeatureWordInfo { diff --git a/target/i386/cpu.c b/target/i386/cpu.c index 7ae232ab18..fc120c0694 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -4377,6 +4377,26 @@ static bool x86_cpu_have_filtered_features(X86CPU *cpu) return false; } +void mark_unavailable_bits(X86CPU *cpu, FeatureWord w, int index, + const char *verbose_prefix) +{ + FeatureWordInfo *f = &feature_word_info[w]; + g_autofree char *feat_word_str = feature_word_description(f); + uint64_t host_feat = x86_cpu_get_supported_feature_word(w, false); + MultiBitFeatureInfo mf = f->multi_bit_features[index]; + + if ((cpu->env.features[w] ^ host_feat) & mf.mask) { + if (!cpu->force_features) { + cpu->env.features[w] &= ~mf.mask; + } + cpu->filtered_features[w] |= mf.mask; + if (verbose_prefix) + warn_report("%s: %s.%s [%u:%u]", verbose_prefix, feat_word_str, + mf.feat_name, mf.high_bit_position, + mf.low_bit_position); + } +} + static void mark_unavailable_features(X86CPU *cpu, FeatureWord w, uint64_t mask, const char *verbose_prefix) { @@ -6424,10 +6444,21 @@ static void x86_cpu_filter_features(X86CPU *cpu, bool verbose) } for (w = 0; w < FEATURE_WORDS; w++) { - uint64_t host_feat = - x86_cpu_get_supported_feature_word(w, false); - uint64_t requested_features = env->features[w]; - uint64_t unavailable_features = requested_features & ~host_feat; + uint64_t host_feat = x86_cpu_get_supported_feature_word(w, false); + FeatureWordInfo f = feature_word_info[w]; + uint64_t unavailable_features = env->features[w] & ~host_feat; + int i; + + for (i = 0; i < f.num_multi_bit_features; i++) { + MultiBitFeatureInfo mf = f.multi_bit_features[i]; + if (!mf.mark_unavailable_bits) { + mf.mark_unavailable_bits = mark_unavailable_bits; + } + mf.mark_unavailable_bits(cpu, w, i, prefix); + + unavailable_features &= ~mf.mask; + } + mark_unavailable_features(cpu, w, unavailable_features, prefix); } diff --git a/target/i386/cpu.h b/target/i386/cpu.h index 1c90fb6c9d..824a2b0f85 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -2103,6 +2103,8 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, void cpu_clear_apic_feature(CPUX86State *env); void host_cpuid(uint32_t function, uint32_t count, uint32_t *eax, uint32_t *ebx, uint32_t *ecx, uint32_t *edx); +void mark_unavailable_bits(X86CPU *cpu, FeatureWord w, int index, + const char *verbose_prefix); /* helper.c */ void x86_cpu_set_a20(X86CPU *cpu, int a20_state); From patchwork Thu Oct 27 02:00:35 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695203 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=ZnjruUs7; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTWn6v1Vz20KC for ; Thu, 27 Oct 2022 13:04:37 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsCz-0001fj-KH; Wed, 26 Oct 2022 22:01:25 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCh-0001Zd-3y for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:09 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCd-0004u8-Jx for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:05 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836063; x=1698372063; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=JWUViv/RcU5IlWP3WUKFML58YY1O8/EHiGTrPfzZm3c=; b=ZnjruUs7xzfALTs7Lapzt36aH/FQf+OV7YlmACTEnsNeZUGYoJLGc8l2 UDGjb2R3GjWmU7nf6oVkRHQRV/v5qEgfSqCZnx++MPjyBVzgbnGbqcETy rjJLDwkV8317MysA8wGpuFzZXZI2a4SkrO51IaK5SJZ/XQzC7m//kUIBG gHNh1dyGNI35Giil6P9OHHs9yCoQnrG3B+cQa7H4DgSIswFNs6W0Db2+S m+FBIOKCFO1SNPFnOCyn4j9p2m75X79QAwXoJ8lVR/GBgK2810BmUX0sa S877Hl1q0xazoOrRW6ZCg/XQlTxaa0F8pXPI6GHsaaYnen1wOg1P77VBn Q==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174873" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174873" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407943" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407943" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 5/6] i386: Initialize AMX CPUID leaves with corresponding env->features[] leaves Date: Wed, 26 Oct 2022 19:00:35 -0700 Message-Id: <20221027020036.373140-6-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org The AMX-related CPUID value, i.e., CPUID(0x1D,1):EAX, CPUID(0x1D,1):EBX, CPUID(0x1D,1):ECX and CPUID(0x1E,0):EBX are hard-coded to Sapphire Rapids without considering future platforms. Replace these hard-coded values with env->features[], so QEMU can pass the right value to KVM. Signed-off-by: Wang, Lei --- target/i386/cpu.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/target/i386/cpu.c b/target/i386/cpu.c index fc120c0694..21d9529d38 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -576,16 +576,16 @@ static CPUCacheInfo legacy_l3_cache = { #define INTEL_PT_PSB_BITMAP (0x003f << 16) /* Support 2K,4K,8K,16K,32K,64K */ /* CPUID Leaf 0x1D constants: */ -#define INTEL_AMX_TILE_MAX_SUBLEAF 0x1 -#define INTEL_AMX_TOTAL_TILE_BYTES 0x2000 -#define INTEL_AMX_BYTES_PER_TILE 0x400 -#define INTEL_AMX_BYTES_PER_ROW 0x40 -#define INTEL_AMX_TILE_MAX_NAMES 0x8 -#define INTEL_AMX_TILE_MAX_ROWS 0x10 +#define INTEL_SPR_AMX_TILE_MAX_SUBLEAF 0x1 +#define INTEL_SPR_AMX_TOTAL_TILE_BYTES 0x2000 +#define INTEL_SPR_AMX_BYTES_PER_TILE 0x400 +#define INTEL_SPR_AMX_BYTES_PER_ROW 0x40 +#define INTEL_SPR_AMX_TILE_MAX_NAMES 0x8 +#define INTEL_SPR_AMX_TILE_MAX_ROWS 0x10 /* CPUID Leaf 0x1E constants: */ -#define INTEL_AMX_TMUL_MAX_K 0x10 -#define INTEL_AMX_TMUL_MAX_N 0x40 +#define INTEL_SPR_AMX_TMUL_MAX_K 0x10 +#define INTEL_SPR_AMX_TMUL_MAX_N 0x40 void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1, uint32_t vendor2, uint32_t vendor3) @@ -5763,12 +5763,11 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, if (count == 0) { /* Highest numbered palette subleaf */ - *eax = INTEL_AMX_TILE_MAX_SUBLEAF; + *eax = INTEL_SPR_AMX_TILE_MAX_SUBLEAF; } else if (count == 1) { - *eax = INTEL_AMX_TOTAL_TILE_BYTES | - (INTEL_AMX_BYTES_PER_TILE << 16); - *ebx = INTEL_AMX_BYTES_PER_ROW | (INTEL_AMX_TILE_MAX_NAMES << 16); - *ecx = INTEL_AMX_TILE_MAX_ROWS; + *eax = env->features[FEAT_1D_1_EAX]; + *ebx = env->features[FEAT_1D_1_EBX]; + *ecx = env->features[FEAT_1D_1_ECX]; } break; } @@ -5784,7 +5783,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count, if (count == 0) { /* Highest numbered palette subleaf */ - *ebx = INTEL_AMX_TMUL_MAX_K | (INTEL_AMX_TMUL_MAX_N << 8); + *ebx = env->features[FEAT_1E_0_EBX]; } break; } From patchwork Thu Oct 27 02:00:36 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Lei Wang X-Patchwork-Id: 1695201 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.a=rsa-sha256 header.s=Intel header.b=Ym/6nYjJ; dkim-atps=neutral Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4MyTVL0Yw9z20S2 for ; Thu, 27 Oct 2022 13:03:19 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1onsD0-0001fn-63; Wed, 26 Oct 2022 22:01:26 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCh-0001Ze-3y for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:09 -0400 Received: from mga06b.intel.com ([134.134.136.31] helo=mga06.intel.com) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1onsCe-0004sz-3z for qemu-devel@nongnu.org; Wed, 26 Oct 2022 22:01:06 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1666836064; x=1698372064; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=ufJ22/DGUXT1N+ZEEl+g8+XvfZpk+SlemHq9qNRodlI=; b=Ym/6nYjJ8PfmHyN7AM+8r2NbeJLRWoSdJkwR3g9KmDxXUCKScGVc0YUz OllkemNFNMz/hVCpDb2vW5eq73dpMM/L4fM34Vn/7A4VuDWTZregUJDNZ ANEgh8YDhZkR2i5i3n25xsnWsO0qGJt96EpmRIAoiJpyP64py13pJcOLj P58iJFeFUMwv+fMgi7VmXlGKdsFybSBBw1TXDSvjEoJ4EyAOFfGDsbwGY OiZm3bpHMesG6dRtqe28o6JCCjZJnydwk5EqNvmgh5TM562jCyA9yRit5 ggIRkKMYvLdQ4w7cXLGD5kjXphS5lpdvkzqEQAP8uD96i876Ff02XTLj0 A==; X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="370174874" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="370174874" Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 X-IronPort-AV: E=McAfee;i="6500,9779,10512"; a="877407949" X-IronPort-AV: E=Sophos;i="5.95,215,1661842800"; d="scan'208";a="877407949" Received: from b49691a74b20.jf.intel.com ([10.45.76.123]) by fmsmga006-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 26 Oct 2022 19:00:38 -0700 From: "Wang, Lei" To: pbonzini@redhat.com Cc: qemu-devel@nongnu.org, dgilbert@redhat.com, berrange@redhat.com, xiaoyao.li@intel.com, yang.zhong@linux.intel.com Subject: [PATCH 6/6] i386: Add new CPU model SapphireRapids Date: Wed, 26 Oct 2022 19:00:36 -0700 Message-Id: <20221027020036.373140-7-lei4.wang@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221027020036.373140-1-lei4.wang@intel.com> References: <20221027020036.373140-1-lei4.wang@intel.com> MIME-Version: 1.0 Received-SPF: pass client-ip=134.134.136.31; envelope-from=lei4.wang@intel.com; helo=mga06.intel.com X-Spam_score_int: -48 X-Spam_score: -4.9 X-Spam_bar: ---- X-Spam_report: (-4.9 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.515, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, UPPERCASE_50_75=0.008 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "Qemu-devel" Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org The new CPU model mostly inherits features from Icelake-Server, while adding new features: - AMX (Advance Matrix eXtensions) - Bus Lock Debug Exception and new instructions: - AVX VNNI (Vector Neural Network Instruction): - VPDPBUS: Multiply and Add Unsigned and Signed Bytes - VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation - VPDPWSSD: Multiply and Add Signed Word Integers - VPDPWSSDS: Multiply and Add Signed Integers with Saturation - FP16: Replicates existing AVX512 computational SP (FP32) instructions using FP16 instead of FP32 for ~2X performance gain - SERIALIZE: Provide software with a simple way to force the processor to complete all modifications, faster, allowed in all privilege levels and not causing an unconditional VM exit - TSX Suspend Load Address Tracking: Allows programmers to choose which memory accesses do not need to be tracked in the TSX read set - AVX512_BF16: Vector Neural Network Instructions supporting BFLOAT16 inputs and conversion instructions from IEEE single precision Features may be added in future versions: - CET (virtualization support hasn't been merged) Instructions may be added in future versions: - fast zero-length MOVSB (KVM doesn't support yet) - fast short STOSB (KVM doesn't support yet) - fast short CMPSB, SCASB (KVM doesn't support yet) Signed-off-by: Wang, Lei Reviewed-by: Robert Hoo --- target/i386/cpu.c | 135 ++++++++++++++++++++++++++++++++++++++++++++++ target/i386/cpu.h | 4 ++ 2 files changed, 139 insertions(+) diff --git a/target/i386/cpu.c b/target/i386/cpu.c index 21d9529d38..6bbca600e0 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -3576,6 +3576,141 @@ static const X86CPUDefinition builtin_x86_defs[] = { { /* end of list */ } } }, + { + .name = "SapphireRapids", + .level = 0x20, + .vendor = CPUID_VENDOR_INTEL, + .family = 6, + .model = 143, + .stepping = 4, + /* + * please keep the ascending order so that we can have a clear view of + * bit position of each feature. + */ + .features[FEAT_1_EDX] = + CPUID_FP87 | CPUID_VME | CPUID_DE | CPUID_PSE | CPUID_TSC | + CPUID_MSR | CPUID_PAE | CPUID_MCE | CPUID_CX8 | CPUID_APIC | + CPUID_SEP | CPUID_MTRR | CPUID_PGE | CPUID_MCA | CPUID_CMOV | + CPUID_PAT | CPUID_PSE36 | CPUID_CLFLUSH | CPUID_MMX | CPUID_FXSR | + CPUID_SSE | CPUID_SSE2, + .features[FEAT_1_ECX] = + CPUID_EXT_SSE3 | CPUID_EXT_PCLMULQDQ | CPUID_EXT_SSSE3 | + CPUID_EXT_FMA | CPUID_EXT_CX16 | CPUID_EXT_PCID | CPUID_EXT_SSE41 | + CPUID_EXT_SSE42 | CPUID_EXT_X2APIC | CPUID_EXT_MOVBE | + CPUID_EXT_POPCNT | CPUID_EXT_TSC_DEADLINE_TIMER | CPUID_EXT_AES | + CPUID_EXT_XSAVE | CPUID_EXT_AVX | CPUID_EXT_F16C | CPUID_EXT_RDRAND, + .features[FEAT_8000_0001_EDX] = + CPUID_EXT2_SYSCALL | CPUID_EXT2_NX | CPUID_EXT2_PDPE1GB | + CPUID_EXT2_RDTSCP | CPUID_EXT2_LM, + .features[FEAT_8000_0001_ECX] = + CPUID_EXT3_LAHF_LM | CPUID_EXT3_ABM | CPUID_EXT3_3DNOWPREFETCH, + .features[FEAT_8000_0008_EBX] = + CPUID_8000_0008_EBX_WBNOINVD, + .features[FEAT_7_0_EBX] = + CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 | CPUID_7_0_EBX_HLE | + CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | + CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | + CPUID_7_0_EBX_AVX512F | CPUID_7_0_EBX_AVX512DQ | + CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | CPUID_7_0_EBX_SMAP | + CPUID_7_0_EBX_AVX512IFMA | CPUID_7_0_EBX_CLFLUSHOPT | + CPUID_7_0_EBX_CLWB | CPUID_7_0_EBX_AVX512CD | CPUID_7_0_EBX_SHA_NI | + CPUID_7_0_EBX_AVX512BW | CPUID_7_0_EBX_AVX512VL, + .features[FEAT_7_0_ECX] = + CPUID_7_0_ECX_AVX512_VBMI | CPUID_7_0_ECX_UMIP | CPUID_7_0_ECX_PKU | + CPUID_7_0_ECX_AVX512_VBMI2 | CPUID_7_0_ECX_GFNI | + CPUID_7_0_ECX_VAES | CPUID_7_0_ECX_VPCLMULQDQ | + CPUID_7_0_ECX_AVX512VNNI | CPUID_7_0_ECX_AVX512BITALG | + CPUID_7_0_ECX_AVX512_VPOPCNTDQ | CPUID_7_0_ECX_LA57 | + CPUID_7_0_ECX_RDPID | CPUID_7_0_ECX_BUS_LOCK_DETECT, + .features[FEAT_7_0_EDX] = + CPUID_7_0_EDX_FSRM | CPUID_7_0_EDX_SERIALIZE | + CPUID_7_0_EDX_TSX_LDTRK | CPUID_7_0_EDX_AMX_BF16 | + CPUID_7_0_EDX_AVX512_FP16 | CPUID_7_0_EDX_AMX_TILE | + CPUID_7_0_EDX_AMX_INT8 | CPUID_7_0_EDX_SPEC_CTRL | + CPUID_7_0_EDX_ARCH_CAPABILITIES | CPUID_7_0_EDX_SPEC_CTRL_SSBD, + .features[FEAT_ARCH_CAPABILITIES] = + MSR_ARCH_CAP_RDCL_NO | MSR_ARCH_CAP_IBRS_ALL | + MSR_ARCH_CAP_SKIP_L1DFL_VMENTRY | MSR_ARCH_CAP_MDS_NO | + MSR_ARCH_CAP_PSCHANGE_MC_NO | MSR_ARCH_CAP_TAA_NO, + .features[FEAT_XSAVE] = + CPUID_XSAVE_XSAVEOPT | CPUID_XSAVE_XSAVEC | + CPUID_XSAVE_XGETBV1 | CPUID_XSAVE_XSAVES | CPUID_D_1_EAX_XFD, + .features[FEAT_6_EAX] = + CPUID_6_EAX_ARAT, + .features[FEAT_7_1_EAX] = + CPUID_7_1_EAX_AVX_VNNI | CPUID_7_1_EAX_AVX512_BF16, + .features[FEAT_1D_1_EAX] = INTEL_SPR_AMX_TOTAL_TILE_BYTES | + (INTEL_SPR_AMX_BYTES_PER_TILE << 16), + .features[FEAT_1D_1_EBX] = INTEL_SPR_AMX_BYTES_PER_ROW | + (INTEL_SPR_AMX_TILE_MAX_NAMES << 16), + .features[FEAT_1D_1_ECX] = INTEL_SPR_AMX_TILE_MAX_ROWS, + .features[FEAT_1E_0_EBX] = INTEL_SPR_AMX_TMUL_MAX_K | + (INTEL_SPR_AMX_TMUL_MAX_N << 8), + .features[FEAT_VMX_BASIC] = + MSR_VMX_BASIC_INS_OUTS | MSR_VMX_BASIC_TRUE_CTLS, + .features[FEAT_VMX_ENTRY_CTLS] = + VMX_VM_ENTRY_LOAD_DEBUG_CONTROLS | VMX_VM_ENTRY_IA32E_MODE | + VMX_VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL | + VMX_VM_ENTRY_LOAD_IA32_PAT | VMX_VM_ENTRY_LOAD_IA32_EFER, + .features[FEAT_VMX_EPT_VPID_CAPS] = + MSR_VMX_EPT_EXECONLY | + MSR_VMX_EPT_PAGE_WALK_LENGTH_4 | MSR_VMX_EPT_PAGE_WALK_LENGTH_5 | + MSR_VMX_EPT_WB | MSR_VMX_EPT_2MB | MSR_VMX_EPT_1GB | + MSR_VMX_EPT_INVEPT | MSR_VMX_EPT_AD_BITS | + MSR_VMX_EPT_INVEPT_SINGLE_CONTEXT | MSR_VMX_EPT_INVEPT_ALL_CONTEXT | + MSR_VMX_EPT_INVVPID | MSR_VMX_EPT_INVVPID_SINGLE_ADDR | + MSR_VMX_EPT_INVVPID_SINGLE_CONTEXT | + MSR_VMX_EPT_INVVPID_ALL_CONTEXT | + MSR_VMX_EPT_INVVPID_SINGLE_CONTEXT_NOGLOBALS, + .features[FEAT_VMX_EXIT_CTLS] = + VMX_VM_EXIT_SAVE_DEBUG_CONTROLS | + VMX_VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL | + VMX_VM_EXIT_ACK_INTR_ON_EXIT | VMX_VM_EXIT_SAVE_IA32_PAT | + VMX_VM_EXIT_LOAD_IA32_PAT | VMX_VM_EXIT_SAVE_IA32_EFER | + VMX_VM_EXIT_LOAD_IA32_EFER | VMX_VM_EXIT_SAVE_VMX_PREEMPTION_TIMER, + .features[FEAT_VMX_MISC] = + MSR_VMX_MISC_STORE_LMA | MSR_VMX_MISC_ACTIVITY_HLT | + MSR_VMX_MISC_VMWRITE_VMEXIT, + .features[FEAT_VMX_PINBASED_CTLS] = + VMX_PIN_BASED_EXT_INTR_MASK | VMX_PIN_BASED_NMI_EXITING | + VMX_PIN_BASED_VIRTUAL_NMIS | VMX_PIN_BASED_VMX_PREEMPTION_TIMER | + VMX_PIN_BASED_POSTED_INTR, + .features[FEAT_VMX_PROCBASED_CTLS] = + VMX_CPU_BASED_VIRTUAL_INTR_PENDING | + VMX_CPU_BASED_USE_TSC_OFFSETING | VMX_CPU_BASED_HLT_EXITING | + VMX_CPU_BASED_INVLPG_EXITING | VMX_CPU_BASED_MWAIT_EXITING | + VMX_CPU_BASED_RDPMC_EXITING | VMX_CPU_BASED_RDTSC_EXITING | + VMX_CPU_BASED_CR3_LOAD_EXITING | VMX_CPU_BASED_CR3_STORE_EXITING | + VMX_CPU_BASED_CR8_LOAD_EXITING | VMX_CPU_BASED_CR8_STORE_EXITING | + VMX_CPU_BASED_TPR_SHADOW | VMX_CPU_BASED_VIRTUAL_NMI_PENDING | + VMX_CPU_BASED_MOV_DR_EXITING | VMX_CPU_BASED_UNCOND_IO_EXITING | + VMX_CPU_BASED_USE_IO_BITMAPS | VMX_CPU_BASED_MONITOR_TRAP_FLAG | + VMX_CPU_BASED_USE_MSR_BITMAPS | VMX_CPU_BASED_MONITOR_EXITING | + VMX_CPU_BASED_PAUSE_EXITING | + VMX_CPU_BASED_ACTIVATE_SECONDARY_CONTROLS, + .features[FEAT_VMX_SECONDARY_CTLS] = + VMX_SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES | + VMX_SECONDARY_EXEC_ENABLE_EPT | VMX_SECONDARY_EXEC_DESC | + VMX_SECONDARY_EXEC_RDTSCP | + VMX_SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE | + VMX_SECONDARY_EXEC_ENABLE_VPID | VMX_SECONDARY_EXEC_WBINVD_EXITING | + VMX_SECONDARY_EXEC_UNRESTRICTED_GUEST | + VMX_SECONDARY_EXEC_APIC_REGISTER_VIRT | + VMX_SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY | + VMX_SECONDARY_EXEC_RDRAND_EXITING | + VMX_SECONDARY_EXEC_ENABLE_INVPCID | + VMX_SECONDARY_EXEC_ENABLE_VMFUNC | VMX_SECONDARY_EXEC_SHADOW_VMCS | + VMX_SECONDARY_EXEC_RDSEED_EXITING | VMX_SECONDARY_EXEC_ENABLE_PML | + VMX_SECONDARY_EXEC_XSAVES, + .features[FEAT_VMX_VMFUNC] = + MSR_VMX_VMFUNC_EPT_SWITCHING, + .xlevel = 0x80000008, + .model_id = "Intel Xeon Processor (SapphireRapids)", + .versions = (X86CPUVersionDefinition[]) { + { .version = 1 }, + { /* end of list */ }, + }, + }, { .name = "Denverton", .level = 21, diff --git a/target/i386/cpu.h b/target/i386/cpu.h index 824a2b0f85..4a837cdb4f 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -891,10 +891,14 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w, #define CPUID_7_0_EDX_TSX_LDTRK (1U << 16) /* Architectural LBRs */ #define CPUID_7_0_EDX_ARCH_LBR (1U << 19) +/* AMX_BF16 instruction */ +#define CPUID_7_0_EDX_AMX_BF16 (1U << 22) /* AVX512_FP16 instruction */ #define CPUID_7_0_EDX_AVX512_FP16 (1U << 23) /* AMX tile (two-dimensional register) */ #define CPUID_7_0_EDX_AMX_TILE (1U << 24) +/* AMX_INT8 instruction */ +#define CPUID_7_0_EDX_AMX_INT8 (1U << 25) /* Speculation Control */ #define CPUID_7_0_EDX_SPEC_CTRL (1U << 26) /* Single Thread Indirect Branch Predictors */