As GPUs have their own memory, the first step consists of allocating memory on
-the GPU. A call to \texttt{cudaMalloc}\index{CUDA~functions!cudaMalloc}
-allocates memory on the GPU. The second parameter represents the size of the
-allocated variables, this size is expressed in bits.
-
+the GPU. A call to \texttt{cudaMalloc}\index{CUDA functions!cudaMalloc}
+allocates memory on the GPU. {\bf REREAD The first parameter of this function is a pointer
+on a memory on the device, i.e. the GPU.} The second parameter represents the
+size of the allocated variables, this size is expressed in bits.
+\pagebreak
\lstinputlisting[label=ch2:lst:ex1,caption=simple example]{Chapters/chapter2/ex1.cu}
In this example, we want to compare the execution time of the additions of two
arrays in CPU and GPU. So for both these operations, a timer is created to
-measure the time. CUDA proposes to manipulate timers quite easily. The first
-step is to create the timer\index{CUDA~functions!timer}, then to start it, and at
+measure the time. CUDA manipulates timers quite easily. The first
+step is to create the timer\index{CUDA functions!timer}, then to start it, and at
the end to stop it. For each of these operations a dedicated function is used.
In order to compute the same sum with a GPU, the first step consists of
transferring the data from the CPU (considered as the host with CUDA) to the GPU
(considered as the device with CUDA). A call to \texttt{cudaMemcpy} copies the content of an array allocated in the host to the device when the fourth
parameter is set
-to \texttt{cudaMemcpyHostToDevice}\index{CUDA~functions!cudaMemcpy}. The first
+to \texttt{cudaMemcpyHostToDevice}\index{CUDA functions!cudaMemcpy}. The first
parameter of the function is the destination array, the second is the
source array, and the third is the number of elements to copy (expressed in
bytes).
sufficient). In Listing~\ref{ch2:lst:ex1} at the beginning, a simple kernel,
called \texttt{addition} is defined to compute in parallel the summation of the
two arrays. With CUDA, a kernel starts with the
-keyword \texttt{\_\_global\_\_} \index{CUDA~keywords!\_\_shared\_\_} which
+keyword \texttt{\_\_global\_\_} \index{CUDA keywords!\_\_shared\_\_} which
indicates that this kernel can be called from the C code. The first instruction
in this kernel is used to compute the variable \texttt{tid} which represents the
-thread index. This thread index\index{thread index} is computed according to
+thread index. This thread index\index{CUDA keywords!thread index} is computed according to
the values of the block index
-(called \texttt{blockIdx} \index{CUDA~keywords!blockIdx} in CUDA) and of the
-thread index (called \texttt{threadIdx}\index{CUDA~keywords!threadIdx} in
+(called \texttt{blockIdx} \index{CUDA keywords!blockIdx} in CUDA) and of the
+thread index (called \texttt{threadIdx}\index{CUDA keywords!threadIdx} in
CUDA). Blocks of threads and thread indexes can be decomposed into 1 dimension,
-2 dimensions, or 3 dimensions. {\bf A REGARDER} According to the dimension of manipulated data,
-the appropriate dimension can be useful. In our example, only one dimension is
+2 dimensions, or 3 dimensions. According to the dimension of manipulated data,
+the dimension of blocks of threads must be chosen carefully. In our example, only one dimension is
used. Then using the notation \texttt{.x}, we can access the first dimension
(\texttt{.y} and \texttt{.z}, respectively allow access to the second and
-third dimension). The variable \texttt{blockDim}\index{CUDA~keywords!blockDim}
+third dimensions). The variable \texttt{blockDim}\index{CUDA keywords!blockDim}
gives the size of each block.
-\section{Second example: using CUBLAS}
+\section{Second example: using CUBLAS \index{CUBLAS}}
\label{ch2:2ex}
-The Basic Linear Algebra Subprograms (BLAS) allows programmers to use efficient
+The Basic Linear Algebra Subprograms (BLAS) allow programmers to use efficient
routines for basic linear operations. Those routines are heavily used in many
scientific applications and are optimized for vector operations, matrix-vector
operations, and matrix-matrix
operations~\cite{ch2:journals/ijhpca/Dongarra02}. Some of those operations seem
-to be easy to implement with CUDA. Nevertheless, as soon as a reduction is
+to be easy to implement with CUDA; however, as soon as a reduction is
needed, implementing an efficient reduction routine with CUDA is far from being
-simple. Roughly speaking, a reduction operation\index{reduction~operation} is an
+simple. Roughly speaking, a reduction operation\index{reduction operation} is an
operation which combines all the elements of an array and extracts a number
computed from all the elements. For example, a sum, a maximum, or a dot product
are reduction operations.
result of the multiplication of $A \times B$ is $C$. The
element \texttt{C[i*size+j]} is computed as follows:
\begin{equation}
-C[size*i+j]=\sum_{k=0}^{size-1} A[size*i+k]*B[size*k+j];
+C[size*i+j]=\sum_{k=0}^{size-1} A[size*i+k]*B[size*k+j].
\end{equation}
In Listing~\ref{ch2:lst:ex3}, the CPU computation is performed using 3 loops,
