Progress of CUDA kernelize (File Point of View)

sonots commented 6 years ago

ls ext/cumo/narray/gen/tmpl/

[x] accum_binary.c (mulsum)
[x] accum.c (sum, prod, min, max, ptp?) TODO: Implement NaN version
[x] accum_index.c (max_index, min_index)
[x] allocate.c
[x] alloc_func.c
[x] aref.c (view) <= maybe not needed
[x] aset.c (indexing set) <= maybe not needed
[x] binary2.c
[x] binary.c
[x] binary_s.c
[ ] bincount.c
[x] cast_array.c (store)
[x] cast.c (new_dim0, cast_array, store)
[x] class.c
[ ] clip.c
[x] coerce_cast.c
[x] cond_binary.c
[ ] cond_unary.c
[ ] cum.c (cumsum, sumprod)
[x] each.c (m_data_to_num) <= cudaSynchronize()
[x] each_with_index.c <= cudaSynchronize()
[x] extract.c Extract element value as Ruby Object if self is a dimensionless NArray,
[x] extract_data.c Convert a data value of obj (with a single element) to dtype <= no need
[x] eye.c
[x] fill.c
[x] format.c
[x] format_to_a.c
[ ] frexp.c
[x] init_class.c <= no need
[x] init_module.c <= no need
[x] inspect.c
[x] lib.c
[x] logseq.c
[x] map_with_index.c <= cudaSynchronize()
[ ] median.c
[ ] minmax.c (minmax) => fix accum's ptp, too
[x] module.c (no need)
[x] new_dim0.c
[ ] poly.c
[x] pow.c
[x] powint.c
[ ] qsort.c
[ ] rand.c
[ ] rand_norm.c
[x] seq.c
[ ] set2.c
[ ] sort.c
[ ] sort_index.c
[x] store_array.c (store from Ruby array)
[x] store_bit.c (store from Numo::Bit)
[x] store.c (store_numeric, store_from, store_bit, store_array)
[x] store_from.c (store from other dtype arrays)
[x] store_numeric.c (store from a numeric value) new_dim0, store
[x] to_a.c
[ ] unary2.c
[x] unary.c
[ ] unary_ret2.c
[x] unary_s.c (sqrt, etc)

ls ext/cumo/narray/gen/tmpl_bit/

sonots commented 6 years ago

narray.c

sonots commented 6 years ago

data.c