Skip to content
Stan Seibert edited this page Aug 2, 2018 · 1 revision

Numba Meeting: 2018-08-02

Attendees: Siu, Stuart, Ehsan, Todd

1. New issues

  • #3187 - bug report (lowering list comprehension in sum, but without reproducer...)
    • objectmode
  • #3185 - Support for numpy.partition feature_request
  • #3184 - Equivalent to functools.partial or caching compiled closures
  • #3182 - numpy structured arrays are not working with numpy arrays as elements in nopython mode.
  • #3175 - Support scalars in np.sum bug
  • #3173 - Unable to use jit python function to numba with roctools
  • #3170 - Typing of np.empty_like is not extensible

2. Open PRs

New

  • 3193 Fix setitem/getitem resolvers
    • can merge
  • 3191 CUDA: fix log size
    • can merge
  • 3190 Support passing closures as literal values in typing
    • can merge
  • 3189 Support literal constants in typing of object methods
    • can merge
  • 3186 Support Records in CUDA Const Memory
    • needs review
  • 3181 Support function extension in alias analysis
    • Todd to check
  • 3172 Use float64 add Atomics, Where Available
    • needs review by Siu
  • 3171 Fix #3146. Fix CFUNCTYPE void* return-type handling

Old

  • 3166 [WIP] Objmode with-block
    • Stuart needs to look at dependent PR
  • 3162 Support constant dtype string in nopython mode in functions like numpy.empty.
  • 3160 First attempt at parallel diagnostics
    • Looking for feedback on utility of output, impl is a bit hacky
  • 3153 Fix canonicalize_array_math typing for calls with kw args
  • 3145 [WIP] support for np.fill_diagonal
  • 3142 Issue3139
  • 3137 Fix for issue3103
  • 3134 [WIP] Cfunc x86 abi
    • Contributor updated patch and tests, need rereview?
  • 3132 Adds an ~5 minute guide to Numba.
    • Merged
  • 3128 WIP: Fix recipe for jetson tx2/ARM
    • Will merge when ready
  • 3127 Support for reductions on arrays.
    • First pass review, waiting for changes
  • 3124 Fix 3119, raise for 0d arrays in reductions
    • _
  • 3093 [WIP] Singledispatch overload support for cuda array interface.
    • Needs review
  • 3046 Pairwise sum implementation.
  • 3017 Add facility to support with-contexts
  • #2999 Support LowLevelCallable
  • #2983 [WIP] invert mapping b/w binop operators and the operator module
  • #2950 Fix dispatcher to only consider contiguous-ness.
  • #2942 Fix linkage nature (declspec(dllexport)) of some test functions
  • #2894: [WIP] Implement jitclass default constructor arguments.
  • #2817: [WIP] Emit LLVM optimization remarks

===========================

3. Feature Discussion

  • Adding CI on Numba+HPAT integration? (Yes)
    • Ehsan: One configuration is fine.
    • Just on TravisCI

4. Next Release: Version 0.40, RC=Sept 3, 2018, Final=Sept 10, 2018

  • Experimental python mode blocks
  • Refactored threadpool interface
  • AMD GPU backend
  • Parallel diagnostics
  • Usual collection of bug fixes
Clone this wiki locally