HBM2E PHY V2 (Hard 1) in TSMC (N7, N6, N5)

Overview

The Synopsys HBM2/HBM2E PHY is a complete physical layer IP interface (PHY) solution for high-performance computing (HPC), graphics,
and networking ASIC, ASSP, and system-on-chip (SoC) applications requiring high-bandwidth HBM2/HBM2E SDRAM interfaces operating at up to 3600 Mbps. The Synopsys IP HBM2/HBM2E PHY is ideal for systems with low to modest memory capacity that require higher bandwidth than is attainable with practical DDR4-based systems.
The Synopsys IP HBM2/HBM2E PHY is provided as a set of hard macrocells delivered as GDSII. These hard macrocells include integrated application- specific HBM2/HBM2E I/Os required for HBM2/HBM2E signaling. The design is optimized for high performance, low latency, low area, low power, and ease of integration. The hard macrocells are easily assembled into a complete 1024-bit HBM2/HBM2E PHY. The RTL-based PHY Utility Block (PUB) supports the GDSII-based PHY components and includes the PHY training circuitry, configurations registers and BIST control. The HBM2/HBM2E PHY includes a DFI 4.0-compatible interface to the memory controller, supporting 1:1 and 1:2 clock ratios. The design is compatible with both metal-insulator-metal (MIM) and non-MIM power decoupling strategies.

Key Features

  • Low latency, small area, low power
  • Compatible with JEDEC standard HBM2/HBM2E SDRAMs
  • Data rates up to 2400 Mbps for HBM2 and 3600 Mbps for HBM2E
  • 4H and 8H HBM2/HBM2E SDRAM stacks supported
  • Support for fast switching between 4 frequencies
  • Product subcomponents designed to precisely control timing critical delay and skews
  • Controller DFI-compatible interface (DFI v4.0 Addendum 2)
  • PHY supports 1:1 and 1:2 modes (PHY:SDRAM clock ratios)
  • Includes 1 phase-locked loop (PLL) per PHY and digital delay lines necessary to meet timing specifications
  • PLL allows bypass mode that enables PHY to be run at low speeds
  • Separate transmit DQ and transmit DQS as well as transmit Row/Col and transmit CK delay lines
  • Allows de-skew of transmit DQS/CK vs. transmit DQ/Row-Col, respectively
  • Includes separate receive DQS delay lines for both rising and falling edge of DQS
  • Supports 8-channel HBM2/HBM2E DRAM systems
  • Supports pseudo-channels as defined in the JEDEC HBM2/HBM2E standard
  • Boot time impedance calibration
  • Programmable I/O drive strength matching the HBM SDRAM
  • Delay line VT compensation
  • Time axis data eye training
  • Internal finite state machine (FSM) allows automated training to optimum setting
  • Alternative software hooks included to allow external software to program optimum setting
  • PHY VREF can either be supplied externally through a bump or use an internal VREF generator that can provide a programmable VREF for use internal to the PHY
  • Internal VREF generator is programmable from 0 to VDDQ
  • Physical implementation of the top-level HBM PHY hard macro is designed to be compatible with a face centered rectangular (FCR) micro bump pattern similar to the JEDEC standard HBM2 SDRAM micro bump pattern with a minimum of 55um center-to- center spacing
  • Permits shortest 2.5D routes between PHY and HBM2/HBM2E SDRAMs for highest signal integrity
  • Enhanced power savings support that includes:
  • Per-channel DFI_LP
  • PHY leakage mode with fast exit (<5us)
  • PHY controls to support DRAM retention mode
  • ARM® AMBA ® APB interface for configuration register access
  • Test data register (TDR) interface for configuration register access
  • Multiple test modes
  • Delay line oscillator test mode
  • Mux-scan ATPG
  • At-speed loopback testing on both the address and data channels
  • Pre-hardened HBM PHYs are available in select process technologies
  • Services available for custom PHY hardening requirements
  • 2.5D Interposer reference designs available

Benefits

  • Supports 2.5D-based JEDEC standard HBM2/HBM2E SDRAMs with data rates up to 3600 Mbps
  • 8 independent memory channels (e.g., 1024 bits)
  • Pseudo-channel operation supported to enable up to 16 channels with 1024 bit PHY
  • Supports up to 4 trained frequencies with <5us switching time
  • DFI 4.0-compatible controller interface
  • PHY independent training capability
  • Comprehensive set of design-for-test (DFT) features

Applications

  • High-performance computing
  • Graphics
  • Networking
  • Artificial intelligence
  • Machine learning

Deliverables

  • Executable .run installation file, including GDSII, LEF Files, LVS netlists, .lib/.db timing models, Verilog model, DRC/LVS log files, I/O IBIS model, I/O HSPICE netlist, parameterized Verilog top-level PHY netlist files, sample verification environment, PHY data book, physical implementation guide, application notes, verification guide, installation guide, implementation checklist
  • The PHY Utility Block includes Verilog code, synthesis/STA constraints and scripts, sample verification environment, data book

Technical Specifications

Foundry, Node
TSMC N7, N6, N5 - FF
Availability
Contact the Vendor
TSMC
Pre-Silicon: 5nm , 6nm , 7nm
×
Semiconductor IP