Intel Architecture Instruction Set Extensions Programming .

3y ago
28 Views
2 Downloads
1.34 MB
165 Pages
Last View : 1m ago
Last Download : 3m ago
Upload by : Oscar Steel
Transcription

Intel ArchitectureInstruction Set Extensions and Future FeaturesProgramming Reference319433-040JUNE 2020

Intel technologies features and benefits depend on system configuration and may require enabled hardware, software, or service activation. Learnmore at intel.com, or from the OEM or retailer.No computer system can be absolutely secure. Intel does not assume any liability for lost or stolen data or systems or any damages resultingfrom such losses.You may not use or facilitate the use of this document in connection with any infringement or other legal analysis concerning Intel productsdescribed herein. You agree to grant Intel a non-exclusive, royalty-free license to any patent claim thereafter drafted which includes subjectmatter disclosed herein.No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.This document contains information on products, services and/or processes in development. All information provided here is subject to changewithout notice. Intel does not guarantee the availability of these interfaces in any future product. Contact your Intel representative to obtain thelatest Intel product specifications and roadmaps.Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1800-548-4725, or by visiting http://www.intel.com/design/literature.htm.Intel, the Intel logo, Intel Deep Learning Boost, Intel DL Boost, Intel Atom, Intel Core, Intel SpeedStep, MMX, Pentium, VTune, Xeon Phi, andXeon are trademarks of Intel Corporation in the U.S. and/or other countries.*Other names and brands may be claimed as the property of others.Copyright 1997-2020, Intel Corporation. All Rights Reserved.iiRef. # 319433-040

Revision HistoryRevisionDescriptionDate Removed instructions that now reside in the Intel 64 andIA-32 Architectures Software Developer’s Manual. Minor updates to chapter 1. Updates to Table 2-1, Table 2-2 and Table 2-8 (leaf 07H) toindicate support for AVX512 4VNNIW and AVX512 4FMAPS.-025 Minor update to Table 2-8 (leaf 15H) regarding ECXdefinition.September 2016 Minor updates to Section 4.6.2 and Section 4.6.3 to clarifythe effects of “suppress all exceptions”. Footnote addition to CLWB instruction indicating operandencoding requirement. Removed PCOMMIT.-026 Removed CLWB instruction; it now resides in the Intel 64and IA-32 Architectures Software Developer’s Manual.October 2016 Added additional 512-bit instruction extensions in chapter 6. Added TLB CPUID leaf in chapter 2.-027 Added VPOPCNTD/Q instruction in chapter 6,and CPUIDdetails in chapter 2.December 2016-028 Updated intrinsics for VPOPCNTD/Q instruction in chapter 6.December 2016 Corrected typo in CPUID leaf 18H.-029 Updated operand encoding table format; extracted tupleinformation from operand encoding. Added VPERMB back into chapter 5; inadvertently removed.April 2017 Moved all instructions from chapter 6 to chapter 5. Updated operation section of VPMULTISHIFTQB. Removed unnecessary information from document (chapters2, 3 and 4). Added table listing recent instruction set extensionsintroduction in Intel 64 and IA-32 Processors. Updated CPUID instruction with additional details.-030 Added the following instructions: GF2P8AFFINEINVQB,GF2P8AFFINEQB, GF2P8MULB, VAESDEC, VAESDECLAST,VAESENC, VAESENCLAST, VPCLMULQDQ, VPCOMPRESS,VPDPBUSD, VPDPBUSDS, VPDPWSSD, VPDPWSSDS,VPEXPAND, VPOPCNT, VPSHLD, VPSHLDV, VPSHRD,VPSHRDV, VPSHUFBITQMB.October 2017 Removed the following instructions: VPMADD52HUQ,VPMADD52LUQ, VPERMB, VPERMI2B, VPERMT2B, andVPMULTISHIFTQB. They can be found in the Intel 64 andIA-32 Architectures Software Developer’s Manual, Volumes2A, 2B, 2C & 2D. Moved instructions unique to processors based on theKnights Mill microarchitecture to chapter 3. Added chapter 4: EPT-Based Sub-Page Permissions. Added chapter 5: Intel Processor Trace: VMXImprovements.Ref. # 319433-040iii

RevisionDescriptionDate Updated change log to correct typo in changes from previousrelease. Updated instructions with imm8 operand missing in operandencoding table. Replaced “VLMAX” with “MAXVL” to align terminology usedacross documentation. Added back information on detection of Intel AVX-512instructions.-031 Added Intel Memory Encryption Technologies instructionsPCONFIG and WBNOINVD. These instructions are also addedto Table 1-1 “Recent Instruction Set Extensions Introductionin Intel 64 and IA-32 Processors”. Added Section 1.5“Detection of Intel Memory Encryption Technologies (Intel MKTME) Instructions”. CPUID instruction updated with PCONFIG and WBNOINVDdetails.January 2018 CPUID instruction updated with additional details on leaf07H: Intel Xeon Phi only features identified and listed. CPUID instruction updated with new Intel SGX features inleaf 12H. CPUID instruction updated with new PCONFIG informationsub-leaf 1BH. Updated short descriptions in the following instructions:VPDPBUSD, VPDPBUSDS, VPDPWSSD and VPDPWSSDS. Corrections and clarifications in Chapter 4 “EPT-Based SubPage Permissions”. Corrections and clarifications in Chapter 5 “Intel ProcessorTrace: VMX Improvements”. Corrected PCONFIG CPUID feature flag on instruction page.-032 Minor updates to PCONFIG instruction pages: Changed Table2-2 to use Hex notation; changed “RSVD, MBZ” to“Reserved, must be zero” in two places in Table 2-3.January 2018 Minor typo correction in WBNOINVD instruction description. Updated Table 1-2 “Recent Instruction Set Extensions /Features Introduction in Intel 64 and IA-32 Processors” . Added Section 1.4, “Detection of Future Instructions andFeatures”. Added CLDEMOTE, MOVDIRI, MOVDIR64B, TPAUSE,UMONITOR and UMWAIT instructions.-033 Updated the CPUID instruction with details on newinstructions/features added, as well as new powermanagement details and information on hardware feedbackinterface ISA extensions.March 2018 Corrections to PCONFIG instruction. Moved instructions unique to processors based on theKnights Mill microarchitecture to the Intel 64 and IA-32Architectures Software Developer’s Manual. Added Chapter 5 “Hardware Feedback Interface ISAExtensions”. Added Chapter 6 “AC Split Lock Detection”. Added clarification to leaf 07H in the CPUID instruction.-034 Added MSR index for IA32 UMWAIT CONTROL MSR. Updated registers in TPAUSE and UMWAIT instructions.May 2018 Updated TPAUSE and UMWAIT intrinsics.ivRef. # 319433-040

RevisionDescriptionDate Updated Table 1-2 “Recent Instruction Set Extensions /Features Introduction in Intel 64 and IA-32 Processors” tolist the AVX512 VNNI instruction set architecture on aseparate line due to presence on future processors availablesooner than previously listed. Updated CPUID instruction in various places.-035 Removal of NDD/DDS/NDS terms from instructions. Note:Previously, the terms NDS, NDD and DDS were used ininstructions with an EVEX (or VEX) prefix. These termsindicated that the vvvv field was valid for encoding, andspecified register usage. These terms are no longernecessary and are redundant with the instruction operandencoding tables provided with each instruction. Theinstruction operand encoding tables give explicit details on alloperands, indicating where every operand is stored and ifthey are read or written. If vvvv is not listed as an operand inthe instruction operand encoding table, then EVEX (or VEX)vvvv must be 0b1111.October 2018 Added additional #GP exception condition to TPAUSE andUMWAIT. Updated Chapter 5 “Hardware Feedback Interface ISAExtensions” as follows: changed scheduler/software tooperating system or OS, changed LP0 Scheduler Feedback toLP0 Capability Values, various description updates, clarifiedthat capability updates are independent, and added anupdate to clarify that bits 0 and 1 will always be set togetherin Section 5.1.4. Added IA32 CORE CAPABILITY MSR to Chapter 6 “AC SplitLock Detection”. Added AVX512 BF16 instructions in chapter 2; relatedCPUID information updated in chapter 1. Added new section to chapter 1 describing bfloat16 format.-036 CPUID leaf updates to align with the Intel 64 and IA-32Architectures Software Developer’s Manual. Removed CLDEMOTE, TPAUSE, UMONITOR, and UMWAITinstructions; they now reside in the Intel 64 and IA-32Architectures Software Developer’s Manual.April 2019 Changes now marked by green change bars and green fontin order to view changes at a text level. Removed chapter 3, “EPT-Based Sub-Page Permissions”,chapter 4, “Intel Processor Trace: VMX Improvements”, andchapter 6, “Split Lock Detection”; this information is in theIntel 64 and IA-32 Architectures Software Developer’sManual. Removed MOVDIRI and MOVDIR64B instructions; they nowreside in the Intel 64 and IA-32 Architectures SoftwareDeveloper’s Manual.-037 Updated Table 1-2 with new features in future processors.May 2019 Updated Table 1-3 with support for AVX512 VP2INTERSECT. Updated Table 1-5 with support for ENQCMD: EnqueueStores. Added ENQCMD/ENQCMDS andVP2INTERSECTD/VP2INTERSECTQ instructions, and updatedCPUID accordingly. Added new chapter: Chapter 4, Enqueue Stores and ProcessAddress Space Identifiers (PASIDs).Ref. # 319433-040v

RevisionDescriptionDate Removed instruction extensions/features from Table 1-2“Recent Instruction Set Extensions / Features Introduction inIntel 64 and IA-32 Processors” that are available inprocessors covered in the Intel 64 and IA-32 ArchitecturesSoftware Developer’s Manual. This information can be foundin Chapter 5 “Instruction Set Summary”, of Volume 1. In Section 1.7, “Detection of Future Instructions”, removedinstructions from Table 1-5 “Future Instructions” that areavailable in processors covered in the Intel 64 and IA-32Architectures Software Developer’s Manual. Removed instructions with the following CPUID feature flags:AVX512 VNNI, VAES, GFNI (AVX/AVX512), AVX512 VBMI2,VPCLMULQDQ, AVX512 BITALG; they now reside in theIntel 64 and IA-32 Architectures Software Developer’sManual.-038 CPUID instruction updated with Hybrid information sub-leaf1AH, SERIALIZE and TSXLDTRK support, updates to the L3Cache Intel RDT Monitoring Capability Enumeration Sub-leaf,and updates to the Memory Bandwidth AllocationEnumeration Sub-leaf.March 2020 Replaced with : notation in operation sections ofinstructions. These changes are not marked with changebars. Added the following instructions: SERIALIZE, XRESLDTRK,XSUSLDTRK. Update to the VDPBF16PS instruction. Updates to Chapter 4, “Hardware Feedback Interface ISAExtensions”. Added Chapter 5, “TSX Suspend Load Address Tracking”. Added Chapter 6, “Hypervisor-managed Linear AddressTranslation”. Added Chapter 7, “Architectural Last Branch Records (LBRs)”. Added Chapter 8, “Non-Write-Back Lock DisableArchitecture”. Added Chapter 9, “Intel Resource Director TechnologyFeature Updates”. Updated Section 1.1 “About this Document” to reflectchapter changes in this release. Added Section 1.2 “DisplayFamily and DisplayModel forFuture Processors”. Updated Table 1-2 “Recent Instruction Set Extensions /Features Introduction in Intel 64 and IA-32 Processors”.-039 CPUID instruction updated. Removed Chapter 4 “Hardware Feedback Interface”. Thisinformation is now in the Intel 64 and IA-32 ArchitecturesSoftware Developer’s Manual.June 2020 Updated Figure 5-1 “Example HLAT Software Usage”. Added Table 6-5 “Encodings for 64-Bit Guest-State Fields(0010 10xx xxxx xxxAb)” to Chapter 6. Added Chapter 8 “Bus Lock and VM Notify”.viRef. # 319433-040

RevisionDescriptionDate Updated Section 1.1 “About this Document” to reflectchapter changes in this release. Updated Table 1-2 “Recent Instruction Set Extensions /Features Introduction in Intel 64 and IA-32 Processors”. CPUID instruction updated.-040 Added notation updates to the beginning of Chapter 2.Updated ENQCMD and ENQCMDS instructions to use thisnotation.June 2020 Added Chapter 3, “Intel AMX Instruction Set Reference, AZ”. Minor updates to Chapter 6, “Hypervisor-managed LinearAddress Translation”.Ref. # 319433-040vii

viiiRef. # 319433-040

REVISION HISTORYCHAPTER 1FUTURE INTEL ARCHITECTURE INSTRUCTION EXTENSIONS AND FEATURES1.11.21.31.41.51.61.7About This Document. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1DisplayFamily and DisplayModel for Future Processors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1Instruction Set Extensions and Feature Introduction in Intel 64 and IA-32 Processors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2Detection of Future Instructions and Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2CPUID Instruction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3CPUID—CPU Identification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1-3Compressed Displacement (disp8*N) Support in EVEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-44bfloat16 Floating-Point Format. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-45CHAPTER 2INSTRUC

Features Introduction in Intel 64 and IA-32 Processors” to list the AVX512_VNNI instruction set architecture on a separate line due to presence on future processors available sooner than previously listed. Updated CPUID instruction in various places. Removal of NDD/DDS/NDS terms from instructions. Note:

Related Documents:

Intel C Compiler Intel Fortran Compiler Intel Distribution for Python* Intel Math Kernel Library Intel Integrated Performance Primitives Intel Threading Building Blocks Intel Data Analytics Acceleration Library Included in Composer Edition SCALE Intel MPI Library Intel Trace Analyze

Document Number: 337029 -009 Intel RealSenseTM Product Family D400 Series Datasheet Intel RealSense Vision Processor D4, Intel RealSense Vision Processor D4 Board, Intel RealSense Vision Processor D4 Board V2, Intel RealSense Vision Processor D4 Board V3, Intel RealSense Depth Module D400, Intel RealSense Depth Module D410, Intel

Lenovo recommends Windows 8 Pro. SPECIFICATIONS PrOCESSOr OPErATING SySTEM I/O (INPUT/OUTPUT) POrTS Mini-Tower / Small Form Factor (SFF) Intel Core i7-4770S 65W Intel Core i7-4770 84W Intel Core i5-4430S 65W Intel Core i5-4430 84W Intel Core i5-4570S 65W Intel Core i5-4570 84W Intel Core i5-4670S 65W Intel Core i5-4670 84W Intel Core i3-4330 65W

HP recommends Windows 10 Pro. FormFactor Mini AvailableOperatingSystem AvailableProcessors Intel Core i5-6500 with Intel HD Graphics 530 (3.2 GHz, up to 3.6 GHz with Intel Turbo Boost, 6 MB cache, 4 cores); Intel Core i5-6500T with Intel HD Graphics 530 (2.5 GHz, up to 3.1 GHz with Intel Turbo Boost, 6 MB cache, 4 cores); Intel Core i7-6700 with Intel HD Graphics 530 (3.4

Byung-Gon Chun Intel Labs Berkeley byung-gon.chun@intel.com Sunghwan Ihm Princeton University sihm@cs.princeton.edu Petros Maniatis Intel Labs Berkeley petros.maniatis@intel.com Mayur Naik Intel Labs Berkeley mayur.naik@intel.com Ashwin Patti Intel Labs Berkeley ashwin.patti@intel.com Abstract Mobile applications are becoming increasingly .

Intel QAT Intel C620 series chipset with Intel QAT integrated on motherboard Tested with Intel C627 chipset. Ships with Intel C627 Chipset. Intel C627 and C626 chipset PCIe adapters are also available from Advantech (PCIE-3030NP and PCIE-3031NP) Storage Minimum 2 x 480 GB Intel SSD Dat

Intel Core Duo Processor for Intel Centrino Duo Processor Technology Based on Mobile Intel 945 Express Chipset Family Datasheet Intel Core Duo Processor and Intel Core Solo Processor on ñ nm Process Datasheet Intel Pentium Dual-Core Mobile Processor Datasheet Intel

performance and thus better support for deep learning algorithms. In 2017, Intel released Intel Xeon Scalable processors, which includes Intel Advance Vector Extension 512 (Intel AVX-512) instruction set and Intel Math Kernel Library for Deep Neural Networks (Intel MKL-DNN) [10]. The Intel AVX-512 and MKL-DNN accelerate deep