CS6023 GPU Programming

January 2022
Important Links Slides
  1. Intro + Logistics
  2. Computation
  3. Memory
  4. Synchronization
  5. Functions
  6. Support
  7. Streams
  8. Topics
  9. Case Study -- Graphs

TCF rating:

  • course = 0.89 (institute mean 0.81)
  • instructor = 0.93 (institute mean 0.83)
  • Full report
  •   Evaluation
    Eval. ItemMarksDeadlines
    StudentTAInstructor
    A110Feb 13+10 days--
    A215Mar 6+10--
    A315Mar 31 -- Apr 3+10--
    A4/CP20Apr 24+10--
    MidSem20Mar 11--+10
    EndSem20Apr 28--+10

    Attendance
    Standard institute rules apply. WebEx tracks attendance. See a sample report (from the first lecture).

    Other details


    Schedule
    MonthDatesTopicComments
     January   17, 18, 20, 21  Introduction, Computation  
  • Hello World, One, Two, Three
  • Grid, Blocks, Threads
  • Kernel Launch: 1D, 1D-General, 2D
  •    24, 25, 27, 28  Computation  
  • CPU-GPU Communication (cudaMalloc, cudaMemcpy)
  • Global variables
  • Matrix mult.: CPU, Outer parallel, Outer+Inner parallel
  •    31  Computation  
  • Thread Divergence
  • Divergence due to switch
  • Problem Set 1
  •  February  1, 3, 4  Memory  
  • Memory Coalescing
  • AoS versus SoA
  • Barrier
  •    7, 8, 10, 11  Memory  
  • Linked List Copying
  • Shared Memory
  • Shared Memory with Barrier
  • String Permutation
  • Dynamic Shared Memory
  • Dynamic Shared Memory with Multiple Arrays

    CUDA GDB
  • Error Handling
  • Dangling Pointer
  •    14, 15, 17, 18  Memory  
  • Texture Memory (via CUDA SDK)
  • Constant Memory
  • Bank Conflicts
  • Problem Set 2

    NvProf
  • Original Code
  • Loop Fusion
  • Kernel Fusion
  • Converting Loop to Blocks
  •     21, 22, 24, 25  Synchronization  
  • Convolution
  • Worklist Insertion
  • Task Donation
  •     28  Synchronization  
  • Reduction: i + N/2, N - i, i + 1
  • Prefix Sum / Scan
  •  March   1, 3, 4  Synchronization  
  • No Global Barrier
  • Global Barrier using Atomics
  • Hierarchical Global Barrier
  •     7, 8, 10, 11  Synchronization   MidSem on 11.
  • Linked List Insertion
  • CPU-GPU Shared Pinned Memory
  • Persistent Kernel
  • Problem Set 3
  •     14, 15, 17, 18  Synchronization  
  • Array increment: Sequential, Parallel
  • Thrust basics
  • Thrust Reduction
  • Thrust Prefix Sum
  • Thrust-like device vector implementation
  •    21, 22, 24, 25  Synchronization  
  • Basic Stream Program
  • with Asynchronous memcpy
  • with cudaHostAlloc
  • Cooperative Kernels
  •    28, 29, 31  Support, Functions  
  • Dynamic Parallelism
  • Conditional Child Kernels
  • using Global Device Memory
  • with Non-Blocking Streams
  •  April  1  Functions  
  • MultiGPU: Number of Devices
  • Cross-Device Synchronization
  •    4, 5, 7, 8  Topics  
  • PTX: CUDA Code, Assembly Code
  • Basic Warp Voting
  • Converting Mask to Count (popc)
  • Use of ffs
  • Conditional Participation in ballot
  •     11, 12  Topics  
  • Loop Unrolling, Unrolled Assembly
  • Heterogeneous Computation
  •     18, 19, 21, 22  Topics  
  • with Shared Variable
  • Task Distribution
  • OpenMP Reduction
  • with HostAlloc'ed Memory
  • Dynamic Scheduling
  • OpenCL: Driver, Kernel
  •     25, 26  Case Study  
        28    EndSem from 10:00 -- 12:00


    GPU Programming Crossword Puzzle (click and type)

    Courtesy: crosswordlabs.com