1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
|
.. currentmodule:: numpy
.. _alignment:
****************
Memory alignment
****************
NumPy alignment goals
=====================
There are three use-cases related to memory alignment in NumPy (as of 1.14):
1. Creating :term:`structured datatypes <structured data type>` with
:term:`fields <field>` aligned like in a C-struct.
2. Speeding up copy operations by using :class:`uint` assignment in instead of
``memcpy``.
3. Guaranteeing safe aligned access for ufuncs/setitem/casting code.
NumPy uses two different forms of alignment to achieve these goals:
"True alignment" and "Uint alignment".
"True" alignment refers to the architecture-dependent alignment of an
equivalent C-type in C. For example, in x64 systems :attr:`float64` is
equivalent to ``double`` in C. On most systems, this has either an alignment of
4 or 8 bytes (and this can be controlled in GCC by the option
``malign-double``). A variable is aligned in memory if its memory offset is a
multiple of its alignment. On some systems (eg. sparc) memory alignment is
required; on others, it gives a speedup.
"Uint" alignment depends on the size of a datatype. It is defined to be the
"True alignment" of the uint used by NumPy's copy-code to copy the datatype, or
undefined/unaligned if there is no equivalent uint. Currently, NumPy uses
``uint8``, ``uint16``, ``uint32``, ``uint64``, and ``uint64`` to copy data of
size 1, 2, 4, 8, 16 bytes respectively, and all other sized datatypes cannot
be uint-aligned.
For example, on a (typical Linux x64 GCC) system, the NumPy :attr:`complex64`
datatype is implemented as ``struct { float real, imag; }``. This has "true"
alignment of 4 and "uint" alignment of 8 (equal to the true alignment of
``uint64``).
Some cases where uint and true alignment are different (default GCC Linux):
====== ========= ======== ========
arch type true-aln uint-aln
====== ========= ======== ========
x86_64 complex64 4 8
x86_64 float128 16 8
x86 float96 4 \-
====== ========= ======== ========
Variables in NumPy which control and describe alignment
=======================================================
There are 4 relevant uses of the word ``align`` used in NumPy:
* The :attr:`dtype.alignment` attribute (``descr->alignment`` in C). This is
meant to reflect the "true alignment" of the type. It has arch-dependent
default values for all datatypes, except for the structured types created
with ``align=True`` as described below.
* The ``ALIGNED`` flag of an ndarray, computed in ``IsAligned`` and checked
by :c:func:`PyArray_ISALIGNED`. This is computed from
:attr:`dtype.alignment`.
It is set to ``True`` if every item in the array is at a memory location
consistent with :attr:`dtype.alignment`, which is the case if the
``data ptr`` and all strides of the array are multiples of that alignment.
* The ``align`` keyword of the dtype constructor, which only affects
:ref:`structured_arrays`. If the structure's field offsets are not manually
provided, NumPy determines offsets automatically. In that case,
``align=True`` pads the structure so that each field is "true" aligned in
memory and sets :attr:`dtype.alignment` to be the largest of the field
"true" alignments. This is like what C-structs usually do. Otherwise if
offsets or itemsize were manually provided ``align=True`` simply checks that
all the fields are "true" aligned and that the total itemsize is a multiple
of the largest field alignment. In either case :attr:`dtype.isalignedstruct`
is also set to True.
* ``IsUintAligned`` is used to determine if an ndarray is "uint aligned" in
an analogous way to how ``IsAligned`` checks for true alignment.
Consequences of alignment
=========================
Here is how the variables above are used:
1. Creating aligned structs: To know how to offset a field when
``align=True``, NumPy looks up ``field.dtype.alignment``. This includes
fields that are nested structured arrays.
2. Ufuncs: If the ``ALIGNED`` flag of an array is False, ufuncs will
buffer/cast the array before evaluation. This is needed since ufunc inner
loops access raw elements directly, which might fail on some archs if the
elements are not true-aligned.
3. Getitem/setitem/copyswap function: Similar to ufuncs, these functions
generally have two code paths. If ``ALIGNED`` is False they will
use a code path that buffers the arguments so they are true-aligned.
4. Strided copy code: Here, "uint alignment" is used instead. If the itemsize
of an array is equal to 1, 2, 4, 8 or 16 bytes and the array is uint
aligned then instead NumPy will do ``*(uintN*)dst) = *(uintN*)src)`` for
appropriate N. Otherwise, NumPy copies by doing ``memcpy(dst, src, N)``.
5. Nditer code: Since this often calls the strided copy code, it must
check for "uint alignment".
6. Cast code: This checks for "true" alignment, as it does
``*dst = CASTFUNC(*src)`` if aligned. Otherwise, it does
``memmove(srcval, src); dstval = CASTFUNC(srcval); memmove(dst, dstval)``
where dstval/srcval are aligned.
Note that the strided-copy and strided-cast code are deeply intertwined and so
any arrays being processed by them must be both uint and true aligned, even
though the copy-code only needs uint alignment and the cast code only true
alignment. If there is ever a big rewrite of this code it would be good to
allow them to use different alignments.
|
