summaryrefslogtreecommitdiff
path: root/numpy/polynomial/polyutils.py
diff options
context:
space:
mode:
authorSebastian Berg <sebastian@sipsolutions.net>2020-06-07 13:40:11 -0500
committerSebastian Berg <sebastian@sipsolutions.net>2020-06-08 09:21:17 -0500
commite32c3155cff28437593952caddfedbda565b64cf (patch)
tree42303cf159a65bba0bef0c53dfb1eec05084d08c /numpy/polynomial/polyutils.py
parent267621f07a902eceffc179fa01f984f3e2d4bdcd (diff)
downloadnumpy-e32c3155cff28437593952caddfedbda565b64cf.tar.gz
MAINT: User array alignment to speed up buffer export
Calculating alignments can be slow, but unless we have a structured dtype, we can simply use the arrays alignment. This code can probably re-use more of the array alignment code (or maybe alignment checks moved into the dtype) even in the structured case. The code using the % operator is very slow compared to the existing power-of-two-assuming alignment code.
Diffstat (limited to 'numpy/polynomial/polyutils.py')
0 files changed, 0 insertions, 0 deletions