What's the purpose of adding an extra stride when creating dynamic buffers?

[xubury]

bgfx/src/bgfx_p.h

Line 3595 in 4581f14

const uint32_t size = bx::strideAlign<16>(_num*_layout.m_stride, _layout.m_stride)+_layout.m_stride;

Is the code above add an extra stride on buffer size?
When using variable size ssbo like below, the buffer object size will mismatch with desired num * layout.m_stride.
So the data_SSBO.length() will be num + 1 instead of num?

layout(std430) buffer layoutName
{
    vec4 data_SSBO[];
};

[bkaradzic]

Actual usable size GL should see is here:

bgfx/src/bgfx_p.h

Line 3611 in 4581f14

dvb.m_size = _num * _layout.m_stride;

[xubury]

But isn't bx::strideAlign<16>(_num*_layout.m_stride, _layout.m_stride)+_layout.m_stride the one that pass down to glBufferData?

[xubury]

It make sense to me when creating dynamic index buffers and use them as std140 layout ssbo.

bgfx/src/bgfx_p.h

Line 3463 in 6be057b

const uint32_t size = bx::alignUp(_num*indexSize, 16);

fits what opengl wiki says The array stride (the bytes between array elements) is always rounded up to the size of a vec4 (ie: 16-bytes)