vulkan: improve im2col performance

Author: daniandtheweb

ggml/src/ggml-vulkan/vulkan-shaders/im2col.comp

@@ -37,51 +37,33 @@ void main() {
     const uint gidx = gl_GlobalInvocationID.x;
 
     const uint oh = gl_GlobalInvocationID.y;
-    const uint batch = gl_GlobalInvocationID.z / p.IC;
-    const uint ic = gl_GlobalInvocationID.z % p.IC;
+    const uint batch_ic = gl_GlobalInvocationID.z;
 
-    A_TYPE values[NUM_ITER];
-    uint offset_dst[NUM_ITER];
-    [[unroll]] for (uint idx = 0; idx < NUM_ITER; ++idx) {
-        values[idx] = A_TYPE(0);
-    }
+    const uint batch = batch_ic / p.IC;
+    const uint ic = batch_ic % p.IC;
+
+    const uint ksize = p.OW * ((p.KH > 1) ? p.KW : 1);
+    const uint src_base = ic * p.offset_delta + batch * p.batch_offset;
+    const uint dst_base = ((batch * p.OH + oh) * p.OW) * p.CHW + ic * p.KW * p.KH;
 
     [[unroll]] for (uint idx = 0; idx < NUM_ITER; ++idx) {
 
         const uint i = gidx * NUM_ITER + idx;
+        if (i >= p.pelements) continue;
 
-        const uint ksize = p.OW * (p.KH > 1 ? p.KW : 1);
         const uint kx = i / ksize;
-        const uint kd = kx * ksize;
-        const uint ky = (i - kd) / p.OW;
+        const uint ky = (i % ksize) / p.OW;
         const uint ix = i % p.OW;
 
-        const uint iiw = ix * p.s0 + kx * p.d0 - p.p0;
-        const uint iih = oh * p.s1 + ky * p.d1 - p.p1;
-
-        offset_dst[idx] =
-            ((batch * p.OH + oh) * p.OW + ix) * p.CHW +
-            (ic * (p.KW * p.KH) + ky * p.KW + kx);
-
-        if (i >= p.pelements) {
-            continue;
-        }
+        const int iiw = int(ix * uint(p.s0)) + int(kx * uint(p.d0)) - p.p0;
+        const int iih = int(oh * uint(p.s1)) + int(ky * uint(p.d1)) - p.p1;
 
-        if (iih < p.IH && iiw < p.IW) {
-            const uint offset_src = ic * p.offset_delta + batch * p.batch_offset;
-            values[idx] = data_a[offset_src + iih * p.IW + iiw];
-        }
-    }
-
-    [[unroll]] for (uint idx = 0; idx < NUM_ITER; ++idx) {
-
-        const uint i = gidx * NUM_ITER + idx;
+        const uint dst_offset = dst_base + ix * p.CHW + ky * p.KW + kx;
 
-        if (i >= p.pelements) {
-            continue;
-        }
+        const bool valid = iih >= 0 && iih < int(p.IH) && iiw >= 0 && iiw < int(p.IW);
+        const uint src_offset = src_base + uint(iih) * p.IW + uint(iiw);
 
-        data_d[offset_dst[idx]] = D_TYPE(values[idx]);
+        data_d[dst_offset] = D_TYPE(valid ? data_a[src_offset] : 0.0);
     }
 
 }