Resume: Design Verification Consultant (Specman)

Discussion in 'VHDL' started by Veritec, Oct 10, 2003.

  1. Veritec

    Veritec Guest

    Chris Starr


    Cadence, Verisity, Viewlogic, VCS, MTI, Verilog-XL, VHDL, Speedsim, Specman,
    Denali, SKILL, Composer, ViewDraw, Smartmodels, Hardware Models
    Sonet, STS-1, STS-3, STS-12, DS1, DS3, VT 1.5
    Fuel Measurement/Management Systems, Full Authority Digital Engine Controls,
    Electronic Flight Instrumentation Systems
    DO-178A, MIL-STD-2167A, OOD, OOP
    C++, Pascal, PLM, FORTRAN, FORTH, MAILSAIL, MIL-STD-1750A (Fairchild 9450),
    80x86, 68000 Assembler
    LEX, Perl, YACC, FLEX, BISON (specialist in application specific compilers)
    UNIX, Solaris, AIX, VAX/VMS, MS-DOS, HP/UX
    PC, VAX, Sun, HP-700, RS6000

    MAY03-SEP03 Intel, ICG Division, Austin, Texas. Responsible for the
    implementation of several functional coverage plans for the host interface
    of a LAN access chip. Functional coverage was implemented using Verisity's
    Specman language. The simulation platform was MTI using Verilog design
    components. Interfaces of interest include PCI, PCI-X 1.0b, PCI-Express and
    TCP/IP. The hardware platform was an X86 based PC running Redhat Linux.

    JAN03-MAY03 Inrange Technologies, Lumberton, New Jersey. Helped complete
    verification of a fibre channel switch platform. The test bench was written
    using Verisity's Specman language. The simulation platform was MTI using a
    mix of Verilog and VHDL design components. Duties included test bench
    enhancements and system-level verification of the design. The hardware
    platform was an X86 based PC running Redhat Linux.

    FEB01-JUL02 Agere Systems, Austin, Texas. Responsible for the design and
    implementation of simulation test bench drivers to configure a network
    processor chipset. The test bench was implemented using Specman e-code and
    Denali memory models. Some highlights of the configuration code include:

    * Register/memory access via PCI cycles or PCI DMA transfers or RTL
    hierarchy. Access via RTL hierarchy would allow configuration to be
    completed in zero-time. Configuration access mode could be switched with a
    run time flag (no test case changes).
    * High-level register/memory access interfaces to simplify test case
    writing. The test case writers would interface with abstract system-level
    data structures instead of low-level bits and bytes.
    * Automated configuration of complex memory architectures (FCRAM, QDRRAM).

    Also worked as a general resource for netlist bring-up and test case

    AUG99-FEB01 Analog Devices/Intel Joint Development Project, Austin, Texas.
    Responsible for the design and implementation of a simulation test bench for
    a new DSP core being developed jointly by ADI and Intel. The test bench was
    implemented in Verilog (XL/VCS/NCVLOG). Some highlights of the test bench

    * Ability to load program code into L2-memory and instruction-cache memory.
    * Ability to load program data into L2-memory and data-cache memory.
    * Integration of an architectural reference model for the DSP core into the
    test bench.
    * Automated checking of register value changes using data from the reference
    * Implementation of the entire 4GB L2 memory space using a sparse memory
    * Automated checking of program counter changes due to interrupts and
    * Very fast instruction execution (approximately 350 instruction per second
    using compiled simulation).
    * Scripting language to control core inputs and to test core outputs
    (resets, clocks, interrupts, exceptions, configuration pins, etc.).

    Also worked on a chip being developed solely by ADI using the DSP core.
    This chip included the DSP core, a bridge from the core to several Amba
    buses (high and low speed), L2-memory, SDRAM controller, SRAM controller and
    various on-chip Amba bus peripherals (UART, SPORT, SPI, DMA engines, etc.).
    My work involved real-time troubleshooting of the test bench and netlist to
    enable test case writing to continue.

    MAR99-AUG99 Compaq, Austin, Texas. Responsible for the block-level
    verification of a PCI-X megacell. The megacell converted PCI-X transactions
    to and from a proprietary on-chip interface bus. Developed test plans,
    coded a random PCI-X transaction generator for stress testing the megacell
    (using Verilog-XL), implemented test cases, and helped supervise a group of
    co-op students at Texas A&M also working on the project.

    NOV97-MAR99 AMD, Austin, Texas. Responsible for the system-level
    verification of a system-on-a-chip consisting of an x86 microprocessor core,
    PCI host bridge, SDRAM controller, ROM controller and various PC/AT
    compatible devices (8237 DMA, UARTs, RTC, PIC, etc.). The system test bench
    environment was constructed using Verisity's e-language based verification
    tool (Specman) integrated with Verilog-XL based bus function models (PCI
    bus, x86 bus, SDRAM devices, ROM devices, etc.)

    NOV96-OCT97 IBM, Microelectronics Division, Research Triangle Park, North
    Carolina. Responsible for the system-level verification of a 1100K gate
    set-top-box ASIC. Developed bus functional models for several internal
    interconnect buses using C++. These models would interface to SpeedSims'
    cycle-based simulator through the standard PLI interface. Developed and
    implemented test suites to verify the connectivity of the microprocessor
    core, cross-bar switch, MPEG-2 video and audio decoders, DRAM, SRAM, SDRAM
    and other off-chip interfaces.

    JUN 95-NOV 96 Cisco Systems, Research Triangle Park, North Carolina.
    Responsible for the system-level verification of the Cisco 7200 series
    router. Developed verification tools and turnkey regression test suites for
    three high-density printed circuit cards. Verification was done using the
    Cadence Verilog-XL simulator, Synopsys SmartModels and hardware models.
    System-level regression test suite was written primarily in Synopsys PCL
    code. A Verilog behavioral model was created to simulate the bus activity
    generated by various port adapter cards which plug into the router chassis
    via a PCI bus interface.

    Also developed a chip-level test bench for an ASIC that bridges token-ring
    frames from a proprietary switching bus to a PCI bus. The test bench was
    developed using Specman (a next-generation verification tool from Verisity).
    This tool was used to automatically generate token-ring frame stimulus and
    verify that the frame data was passed correctly through the chip based on
    the destination address, RIF, SNAP and DSAP fields.

    FEB94-JUN95 Alcatel Network Systems, Raleigh, North Carolina. Developed
    drivers and monitors (Verilog-XL) to generate and analyze STS-1, STS-3 and
    STS-12 stimulus patterns for an ASIC chipset being used in Alcatels' next
    generation SONET switches. These drivers/monitors were integrated into a
    user-friendly test bench to expedite system- level verification of the

    FEB92-FEB94 IBM, Network Systems, Research Triangle Park, North Carolina.
    Developed behavioral models (Verilog-XL) to simulate the system-level
    environment of a full motion video co-processor for an IBM-PC graphics
    adapter card. Responsible for the implementation of full chip simulation
    scenarios to test capabilities of the co-processor which included scaling,
    cropping, dithering and YUV-RGB conversion. Behavioral models generated for
    system level functions included VRAMs, NTSC and compressed video data
    sources, host computer data source, memory bus arbitration module and
    assorted analysis modules to track data flow inside and outside of the chip.

    NOV88-FEB92 Honeywell, Commercial Flight Systems, Phoenix, Arizona.
    Designed, coded and tested software to perform branch and code coverage
    analysis on an electronic flight instrumentation system for the MD-11
    commercial aircraft. This package consisted of: a pre-processor (VAX C) to
    determine the possible branches taken by the flight software; a
    mid-processor (80386 Assembler) to record the actual branches taken; a
    post-processor (VAX C) to compare the possible with the actual branches and
    produce a coverage report detailing any unexecuted code or branches. Also,
    developed software to provide symbology for 80386 emulation on an HP9000
    workstation. Also, developed software to facilitate functional testing of
    the flight software.

    Developed a unique testing tool which allowed component integration-level
    tests to be generated automatically from individual component-level tests
    resulting in a dramatic reduction in man-hours required to certify the
    flight software.

    Developed a code-scanning compiler to verify the integrity of manually
    entered data-dictionary information with procedure/variable references from
    the source code.

    FEB88-AUG88 Hamilton Standard, Windsor Locks, Connecticut. Designed, coded
    and tested software to perform diagnostic testing on Full Authority Digital
    Engine Controls (FADEC) for large military jet engines. Software was
    written in Pascal for an HP9000.

    Oct87-Feb88 AT&T Information Systems, Denver Colorado. Designed, coded and
    tested software to perform diagnostic testing on ISDN BRI compatible
    terminals for the System 85 PBX.

    OCT86-OCT87 Simmonds Precision, Vergennes, Vermont. Designed, coded and
    tested software to perform diagnostic testing on a fuel measurement and
    management system for military aircraft. This package consisted of software
    running in the flight computer (MIL-STD-1750A or Fairchild 9450) to perform
    functional testing of the hardware and software running on an IBM-PC
    (Microsoft C & 8088 Assembler) to control test execution and error logging.
    Challenges of this assignment: a wide variety of communications interfaces
    (RS-232, IEEE 488, MIL-STD-1553); customized interrupt handlers (UART,
    keyboard, timer); a wide variety of test equipment (HP6032A Power Supply,
    LORAL SBA 100A, Control Systems Research MICRISTAR). Extensive
    documentation (MIL-STD-2167A) was produced to satisfy the standards and
    procedures of military contracts. Also, developed a Software Source Control
    System to help maintain customized releases of this package.

    FEB86-MAY86 General Electric, Co., Somersworth, New Hampshire. Made bug
    fixes/enhancements to EPROM burning software (FORTRAN 77) used by GE to
    program their line of demand recording meters. These enhancements included
    the ability to handle 256K EPROMs, validation testing features and changing
    file allocation from static to dynamic.

    FEB85-FEB86 Versatile Systems, Inc., Belmont, California. Designed, coded
    and tested software to translate the database of one PC-based architectural
    drawing program to another. Also, developed a customized PILOT language

    MAR82-FEB85 VLSI Technology, Inc., San Jose, California. Major contributor
    to the cell compiler project. This is a library of customized hardware
    components generated by a collection of software elements called cell
    compilers. These cells, by allowing a logic designer to create flexible
    building blocks (i.e., registers, counters, ALUs), can significantly reduce
    the time required to develop an ASIC. Consultant for several custom chip
    projects: Macintosh Computer for Apple, Atari home video game, Cable TV
    descrambler, Display Controller for a portable PC. These projects
    encompassed all aspects of VLSI chip design from schematic entry/logic
    simulation to cell layout/chip composition.

    JUN81-MAR82 Hewlett Packard, Disc Memory Division, Boise, Idaho. Designed,
    coded and tested software to automatically generate PLAs for VLSI chips.
    This package consisted of: a compiler to convert a high-level language
    description of the finite state machine behavior to truth tables; a logic
    minimizer to reduce the truth tables; a PLA generator to convert the truth
    tables to mask geometry.

    SEP73-JUN75 South Burlington High School, South Burlington, Vermont.
    Designed, coded and tested software to perform class scheduling, produce
    class lists and to maintain attendance records for the school.


    JUNE 1981 Master of Science in Electrical Engineering program, Dartmouth
    College, Thayer School of Engineering, Hanover, New Hampshire. One year
    credit toward degree.

    JUNE 1980 Bachelor of Science in Computer Science with a minor in electrical
    engineering, Michigan State University, East Lansing, Michigan. Graduated
    cum laude.

    PATENTS United States Patent #6,052,745, System for asserting burst
    termination signal and burst complete signal one cycle prior to and during
    last cycle in fixed length burst transfers.


    [1] Cell Compilers vs. Standard Cell Design, Custom Integrated
    Circuits Conference, Rochester, NY, 1984.

    [2] Cell Layout Compilers Simplify Custom IC Design, EDN, September
    15, 1983.

    [3] Automatic PLA Generation, Hewlett Packard VLSI Conference,
    Corvalis, Oregon, 1981.
    Veritec, Oct 10, 2003
    1. Advertisements

  2. Veritec


    Nov 15, 2010
    Likes Received:
    DB601 engines

    Dear Mr. Starr,

    I wonder whether you are the same Chris Starr that wrote an excellent article on the Daimler Benz DB engines on Aeroplane Monthly in May 2005.

    If so, I would ask you to help me in my research on the WW2 Italioan aircraft. As you certainly know, the DB601A engine was built in Italy as the Alfa Romeo RA.1000 RC.41, rated at 1,175 cv (cv is the same as PS in Germany).
    According to the very limited info contained in the Macchi C.202 flight manual, the engine was rated for a normal TO power at 2400 rpm and 1.35 kg/cm2 MP, and an emergency power at 2400 rpm and 1,45 kg/cm2 (limited to 1 minute use): according to your article, this is more than the standard DB601A-1 and corresponds with the DB601N rating. Furthermore the Italian manual requires fuel of 89-91 octane.
    So, in reality which one was the version on which the Italian engine was based?
    Finally, would you have a complete set of power xcurves for this engine (or for the DB601A-1 and DB605A?

    Thank you in advance for your help

    Giulio Valdonio
    Milano, Italy
    valdonio, Nov 15, 2010
    1. Advertisements

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.