Fixed imageAtomicAverageRGBA8


Randall Rauwendaal


May 2, 2013

So I fixed some issues I had in my previous implementation of imageAtomicAverageRGBA8, see the previous post (“GLSL Snippet: emulating running atomic average of colors using imageAtomicCompSwap”) for an explanation of what I got wrong.  Reposting the corrected code here, and sorry to anyone who was trying to use the broken version.

Anyway, original credit for this technique should go to Cyril Crassin, whose implementation in [Crassin & Greene] deftly avoided the mistakes I made by implementing his own pack/unpack functions. Still not sure why his implementation doesn't work for me though. Note: I tried to debug these in the Nsight shader debugger and got the message "Not a debuggable shader", so either it doesn't support atomics (unverified), or these "spinlock" type shaders are too clever for the debugger somehow (for now).

void imageAtomicAverageRGBA8(layout(r32ui) coherent volatile uimage3D voxels, ivec3 coord, vec3 nextVec3)
    uint nextUint = packUnorm4x8(vec4(nextVec3,1.0f/255.0f));
    uint prevUint = 0;
    uint currUint;
    vec4 currVec4;
    vec3 average;
    uint count;
    //"Spin" while threads are trying to change the voxel
    while((currUint = imageAtomicCompSwap(voxels, coord, prevUint, nextUint)) != prevUint)
        prevUint = currUint;                    //store packed rgb average and count
        currVec4 = unpackUnorm4x8(currUint);    //unpack stored rgb average and count
        average =      currVec4.rgb;            //extract rgb average
        count   = uint(currVec4.a*255.0f);      //extract count
        //Compute the running average
        average = (average*count + nextVec3) / (count+1);
        //Pack new average and incremented count back into a uint
        nextUint = packUnorm4x8(vec4(average, (count+1)/255.0f));


[Crassin & Greene] Octree-Based Sparse Voxelization Using the GPU Hardware Rasterizer