Message152404
I've uploaded a new patch that should address the remaining issues:
o In the documentation _testbuffer has been replaced by
m.cast() + the now multi-dimensional m.tolist().
o I restored the state of the limited API. If we want
to include Py_buffer again, I think this should be done
in a separate patch.
o Flags of the memoryview object are private.
Additionally, because NumPy allows non-aligned array accesses,
I changed the packing functions to use memcpy for multi-byte types.
On x86/amd64 gcc is smart enough to produce almost exactly the same
asm output as before, with a slowdown of 0-1%, depending on the
benchmark.
On other platforms the situation might be worse, but I don't have
access to real hardware where alignment actually matters. |
|
Date |
User |
Action |
Args |
2012-01-31 20:34:30 | skrah | set | recipients:
+ skrah, teoliphant, paul.moore, mark.dickinson, ncoghlan, rupole, pitrou, kristjan.jonsson, vstinner, pv, jcon, petri.lehtinen |
2012-01-31 20:34:30 | skrah | set | messageid: <1328042070.59.0.932149953197.issue10181@psf.upfronthosting.co.za> |
2012-01-31 20:34:30 | skrah | link | issue10181 messages |
2012-01-31 20:34:29 | skrah | create | |
|