Rather than in the HLSL writer. This way we output vsir consistent with the vsir
we read, and the vsir that the backends expect [bringing us one step closer to
being able to feed the HLSL frontend directly into the individual backends.]
point-sprite.shader_test is not technically well formed since, in SM4,
the vertex output signature should be:
// Output signature:
// Name Index Mask Register SysValue Format Used
// -------------------- ----- ------ -------- -------- ------- ------
// sv_position 0 xyzw 0 POS float xyzw
// texcoord 0 xy 1 NONE float xy
and the pixel input signature should be:
// Input signature:
// Name Index Mask Register SysValue Format Used
// -------------------- ----- ------ -------- -------- ------- ------
// texcoord 0 xy 0 NONE float xy
so we are not passing "texcoord" properly to the pixel shader, even on
Windows.
The following table shows how each intrinsic maps to d3d assembly and the
flags that appear in the tpf bytecode, in binary.
GroupMemoryBarrier() sync_g 0010
GroupMemoryBarrierWithGroupSync() sync_g_t 0011
DeviceMemoryBarrier() sync_uglobal 1000
DeviceMemoryBarrierWithGroupSync() sync_uglobal_t 1001
AllMemoryBarrier() sync_uglobal_g 1010
AllMemoryBarrierWithGroupSync() sync_uglobal_g_t 1011
