© 2012 Elsevier, Inc. All rights reserved. Chapter 08 © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.1 A 1D convolution example, inside elements. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.2 1D convolution, calculation of P[3]. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.3 1D convolution boundary condition. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.4 A 2D convolution example. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.5 A 2D convolution boundary condition. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.6 A 1D convolution kernel with boundary condition handling. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.7 A review of the CUDA memory model. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.8 A 1D convolution kernel using constant memory for M. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.9 A simplified view of the cache hierarchy of modern processors. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.10 A 1D tiled convolution example. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.11 A tiled 1D convolution kernel using constant memory for M. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.12 A small example of accessing N elements and ghost elements. © 2012 Elsevier, Inc. All rights reserved.

© 2012 Elsevier, Inc. All rights reserved. Figure 8.13 A simpler tiled 1D convolution kernel using constant memory and general caching. © 2012 Elsevier, Inc. All rights reserved.