Issues

• Sorry, cudaMallocManaged() is not implemented in the current version.

  data61/cuda-fixnum#72 · Filter by author guiming-shi opened on Aug 1, 2021
  1

• Implement faster Newton-Raphson

  data61/cuda-fixnum#63 · Filter by author unzvfu opened on Oct 11, 2018
  1

• Investigate specialised mulmod for base*cuml in modexp

  data61/cuda-fixnum#62 · Filter by author unzvfu opened on Sep 24, 2018
  1

• Rewrite test suite to run via Python interface

  data61/cuda-fixnum#61 · Filter by author unzvfu opened on Sep 11, 2018 · Release 0.3
  1

• Re-jigger the API to ease writing HLL interfaces

  data61/cuda-fixnum#60 · Filter by author unzvfu opened on Sep 11, 2018 · Release 0.3
  1

• Implement Python interface

  data61/cuda-fixnum#59 · Filter by author unzvfu opened on Sep 11, 2018 · Release 0.3
  1

• Allow checking for overflow without penalising fast path

  data61/cuda-fixnum#51 · Filter by author unzvfu opened on Aug 27, 2018
  1

• Overhaul benchmarking system

  data61/cuda-fixnum#50 · Filter by author unzvfu opened on Aug 27, 2018
  1

• Use warp votes to branch on argument size to select fastest algo

  data61/cuda-fixnum#49 · Filter by author unzvfu opened on Aug 27, 2018
  1

• Work out why 32-bit digits is slower than 64-bit digits with same fixnum size

  data61/cuda-fixnum#48 · Filter by author unzvfu opened on Aug 27, 2018
  1

• Review consequences of Cuda 7 Independent Thread Scheduling on warp-synchronicity assumptions

  data61/cuda-fixnum#45 · Filter by author unzvfu opened on Aug 17, 2018
  1

• Benchmark against AccelerateHS "bignum" implementation.

  data61/cuda-fixnum#44 · Filter by author unzvfu opened on Aug 16, 2018
  1

• Understand why CLNW sliding-window is faster than k-ary in the tests

  data61/cuda-fixnum#43 · Filter by author unzvfu opened on Aug 13, 2018
  1

• Implement VLNW and compare to the current CLNW sliding-window method

  data61/cuda-fixnum#42 · Filter by author unzvfu opened on Aug 13, 2018
  1

• Enforce exponent sharing in sliding-window modexp function

  data61/cuda-fixnum#41 · Filter by author unzvfu opened on Aug 13, 2018
  1

• Assess potential utility of CUDA warp matrix functions

  data61/cuda-fixnum#39 · Filter by author unzvfu opened on Aug 8, 2018
  1

• Consider reëxpressing fixnum layout in terms of CUDA's "coöperative groups"

  data61/cuda-fixnum#38 · Filter by author unzvfu opened on Aug 7, 2018
  1

• Investigate, document and incorporate "secure" arithmetic implementations

  data61/cuda-fixnum#37 · Filter by author unzvfu opened on Aug 1, 2018
  1

• Clean up warp_fixnum division code

  data61/cuda-fixnum#36 · Filter by author unzvfu opened on Jul 31, 2018
  1

• Ensure arguments from user functions are always read into registers

  data61/cuda-fixnum#35 · Filter by author unzvfu opened on Jul 30, 2018
  1

• Implement bignum arithmetic in terms of fixnum arithmetic

  data61/cuda-fixnum#34 · Filter by author unzvfu opened on Jul 28, 2018
  1

• Speed up test case generation using multiprocessing

  data61/cuda-fixnum#33 · Filter by author unzvfu opened on Jul 23, 2018
  2

• Store test cases in compressed format

  data61/cuda-fixnum#31 · Filter by author unzvfu opened on Jul 23, 2018
  1

• Investigate using Montgomery multiplication in the division algorithms

  data61/cuda-fixnum#29 · Filter by author unzvfu opened on Jul 17, 2018
  1

• Consider storing precomputed values in shared memory

  data61/cuda-fixnum#28 · Filter by author unzvfu opened on Jul 13, 2018
  1