diff options
Diffstat (limited to 'algorithm.tex')
| -rw-r--r-- | algorithm.tex | 16 |
1 files changed, 12 insertions, 4 deletions
diff --git a/algorithm.tex b/algorithm.tex index 684975c..5144bb9 100644 --- a/algorithm.tex +++ b/algorithm.tex @@ -31,6 +31,10 @@ This section covers the main architecture of the HLS tool, and the way in which The main work flow of \vericert{} is shown in Figure~\ref{fig:rtlbranch}, which shows the parts of the translation that are performed in \compcert{}, and which have been added with \vericert{}. +\compcert{} is made up of 11 intermediate languages in between the Clight input and the assembly output. These intermediate languages each serve a different purpose and contain various different optimisations. When designing a new backend for \compcert{}, it is crucial to know where to branch off and start the hardware generation. Many optimisations that the CompCert backend performs are not necessary when generating custom hardware, meaning no CPU architecture is being targeted. These optimisations include register allocation, as there is no more fixed number of registers that need to be targeted. It is therefore important to find the right intermediate language so that the HLS tool still benefits from many of the generic optimisations that CompCert performs, but does not receive the code transformations that are specific to CPU architectures. + +Existing HLS compilers usually use LLVM IR as an intermediate representation when performing HLS specific optimisations, as each instruction can be mapped quite well to hardware which performs the same behaviour. CompCert's three address code (3AC)\footnote{Three address code (3AC) is also know as register transfer language (RTL) in the CompCert literature, however, 3AC is used in this paper so as not to confuse it with register transfer level (RTL), which is another name for the final hardware target of the HLS tool.} is the intermediate language that resembles LLVM IR the most, as it also has an infinite number of pseudo-registers and each instruction maps well to hardware. However, one difference between the two is that 3AC uses operations of the target architecture and performs architecture specific optimisations as well, which is not the case in LLVM IR where all the instructions are quite abstract. This can be mitigated by making CompCert target a specific architecture such as x86\_32, where most operations translate quite well into hardware. In addition to that, many optimisations that are also useful for HLS are performed in 3AC, which means that if it is supported as the input language, the HLS algorithm benefits from the same optimisations. It is therefore a good candidate to be chosen as the input language to the HLS backend. The complete flow that Vericert takes is show in figure~\ref{fig:rtlbranch}. + % - Explain main differences between translating C to software and to hardware. % + This can be done by going through the simple example. @@ -43,7 +47,7 @@ The main work flow of \vericert{} is shown in Figure~\ref{fig:rtlbranch}, which \end{subfigure}% \begin{subfigure}[b]{0.5\linewidth} \inputminted[fontsize=\footnotesize]{c}{data/accumulator.rtl} - \caption{Accumulator RTL code.}\label{fig:accumulator_rtl} + \caption{Accumulator 3AC code.}\label{fig:accumulator_rtl} \end{subfigure} \caption{Accumulator example using \compcert{} to translate from C to three address code (3AC).}\label{fig:accumulator_c_rtl} \end{figure} @@ -58,10 +62,14 @@ The main work flow of \vericert{} is shown in Figure~\ref{fig:rtlbranch}, which \inputminted[fontsize=\tiny]{systemverilog}{data/accumulator2.v} \caption{Accumulator Verilog code.}\label{fig:accumulator_v_2} \end{subfigure} - \caption{Accumulator example using \compcert{} to translate from HTL to Verilog.\YH{I feel like these examples take up too much space, but don't really know of a different way to show off a complete example without the long code.} \JW{Ok, what about drawing the FSMD in some sort of pictorial way? I think you tried drawing it as a schematic previously, but that was too big and clumsy. What about drawing the FSMD as a state machine, with sixteen states and labelled edges between them? Or might that be too abstract?}}\label{fig:accumulator_v} + \caption{Accumulator example using \vericert{} to translate the 3AC to a state machine expressed in Verilog.}\label{fig:accumulator_v} \end{figure} -Taking the simple accumulator program shown in Figure~\ref{fig:accumulator_c}, we can describe the main translation that is performed in Vericert to go from the behaviour description into a hardware design. +\subsection{Example C to Verilog translation} + +Taking the simple accumulator program shown in Figure~\ref{fig:accumulator_c}, we can describe the main translation that is performed in Vericert to go from the behavioural description into a hardware design. + +The first step of the translation is to use \compcert{} to transform the C code into the 3AC shown in Figure~\ref{fig:accumulator_rtl}, including many optimisations that are performed in the 3AC language. This includes optimisations such as constant propagation and dead-code elimination. Function inlining is also performed, and it is used as a way to support function calls instead of having to support the \texttt{Icall} 3AC instruction. The duplication of the function bodies caused by inlining does affect the total area of the hardware, \begin{figure*} \centering @@ -71,7 +79,7 @@ Taking the simple accumulator program shown in Figure~\ref{fig:accumulator_c}, w % + TODO Explain the main mapping in a short simple way -% - Explain why we chose 3AC (RTL) as the branching off point. +% - Explain why we chose 3AC (3AC) as the branching off point. % + TODO Clarify connection between CFG and FSMD |