devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

we were already aligning to 16 byte boundary fixed-size objects that are multiple of 16 bytes;

Browse Source 
now we also align to 8byte boundary fixed-size objects that are multiple of 8 bytes.
That's only useful for now for double, not e.g. for Vector2f, but that didn't seem to hurt. Am I missing something? Do you prefer that we don't align Vector2f at all?
Also, improvements in test_unalignedassert.
This commit is contained in:
Benoit Jacob
2009-10-05 10:11:11 -04:00
parent a9a9ba8453
commit d41577819b
6 changed files with 108 additions and 80 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
test/packetmath.cpp
View File

@@ -99,10 +99,10 @@ template<typename Scalar> void packetmath()
const int PacketSize = ei_packet_traits<Scalar>::size;
const int size = PacketSize*4;
EIGEN_ALIGN_128 Scalar data1[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN_128 Scalar data2[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN_128 Packet packets[PacketSize*2];
EIGEN_ALIGN_128 Scalar ref[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Scalar data1[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Scalar data2[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Packet packets[PacketSize*2];
EIGEN_ALIGN16 Scalar ref[ei_packet_traits<Scalar>::size*4];
for (int i=0; i<size; ++i)
{
data1[i] = ei_random<Scalar>();
@@ -202,9 +202,9 @@ template<typename Scalar> void packetmath_real()
const int PacketSize = ei_packet_traits<Scalar>::size;
const int size = PacketSize*4;
EIGEN_ALIGN_128 Scalar data1[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN_128 Scalar data2[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN_128 Scalar ref[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Scalar data1[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Scalar data2[ei_packet_traits<Scalar>::size*4];
EIGEN_ALIGN16 Scalar ref[ei_packet_traits<Scalar>::size*4];
for (int i=0; i<size; ++i)
{