summaryrefslogtreecommitdiff
path: root/manual
diff options
context:
space:
mode:
Diffstat (limited to 'manual')
-rw-r--r--manual/CHAPTER_CellLib.tex71
-rw-r--r--manual/CHAPTER_Overview.tex21
2 files changed, 74 insertions, 18 deletions
diff --git a/manual/CHAPTER_CellLib.tex b/manual/CHAPTER_CellLib.tex
index 277e8932..cb1bcf1b 100644
--- a/manual/CHAPTER_CellLib.tex
+++ b/manual/CHAPTER_CellLib.tex
@@ -119,6 +119,12 @@ than one bit from \B{S} is set the output is undefined. Cells of this type are u
``parallel cases'' (defined by using the {\tt parallel\_case} attribute or detected by
an optimization).
+The {\tt \$tribuf} cell is used to implement tristate logic. Cells of this type have a \B{WIDTH}
+parameter and inputs \B{A} and \B{EN} and an output \B{Y}. The \B{A} input and \B{Y} output are
+\B{WIDTH} bits wide, and the \B{EN} input is one bit wide. When \B{EN} is 0, the output \B{Y}
+is not driven. When \B{EN} is 1, the value from \B{A} input is sent to the \B{Y} output. Therefore,
+the {\tt \$tribuf} cell implements the function \lstinline[language=Verilog]; Y = EN ? A : 'bz;.
+
Behavioural code with cascaded {\tt if-then-else}- and {\tt case}-statements
usually results in trees of multiplexer cells. Many passes (from various
optimizations to FSM extraction) heavily depend on these multiplexer trees to
@@ -211,14 +217,15 @@ Add information about {\tt \$sr} cells (set-reset flip-flops) and d-type latches
\subsection{Memories}
\label{sec:memcells}
-Memories are either represented using RTLIL::Memory objects and {\tt \$memrd} and {\tt \$memwr} cells
-or simply by using {\tt \$mem} cells.
+Memories are either represented using RTLIL::Memory objects, {\tt \$memrd}, {\tt \$memwr}, and {\tt \$meminit}
+cells, or by {\tt \$mem} cells alone.
In the first alternative the RTLIL::Memory objects hold the general metadata for the memory (bit width,
size in number of words, etc.) and for each port a {\tt \$memrd} (read port) or {\tt \$memwr} (write port)
cell is created. Having individual cells for read and write ports has the advantage that they can be
consolidated using resource sharing passes. In some cases this drastically reduces the number of required
-ports on the memory cell.
+ports on the memory cell. In this alternative, memory initialization data is represented by {\tt \$meminit} cells,
+which allow delaying constant folding for initialization addresses and data until after the frontend finishes.
The {\tt \$memrd} cells have a clock input \B{CLK}, an enable input \B{EN}, an
address input \B{ADDR}, and a data output \B{DATA}. They also have the
@@ -253,7 +260,7 @@ enable bit for each data bit), an address input \B{ADDR} and a data input
\begin{itemize}
\item \B{MEMID} \\
-The name of the RTLIL::Memory object that is associated with this read port.
+The name of the RTLIL::Memory object that is associated with this write port.
\item \B{ABITS} \\
The number of address bits (width of the \B{ADDR} input port).
@@ -262,7 +269,7 @@ The number of address bits (width of the \B{ADDR} input port).
The number of data bits (width of the \B{DATA} output port).
\item \B{CLK\_ENABLE} \\
-When this parameter is non-zero, the clock is used. Otherwise this read port is asynchronous and
+When this parameter is non-zero, the clock is used. Otherwise this write port is asynchronous and
the \B{CLK} input is not used.
\item \B{CLK\_POLARITY} \\
@@ -273,6 +280,27 @@ edge if this parameter is {\tt 1'b0}.
The cell with the higher integer value in this parameter wins a write conflict.
\end{itemize}
+The {\tt \$meminit} cells have an address input \B{ADDR} and a data input \B{DATA}, with the width
+of the \B{DATA} port equal to \B{WIDTH} parameter times \B{WORDS} parameter. Both of the inputs
+must resolve to a constant for synthesis to succeed.
+
+\begin{itemize}
+\item \B{MEMID} \\
+The name of the RTLIL::Memory object that is associated with this initialization cell.
+
+\item \B{ABITS} \\
+The number of address bits (width of the \B{ADDR} input port).
+
+\item \B{WIDTH} \\
+The number of data bits per memory location.
+
+\item \B{WORDS} \\
+The number of consecutive memory locations initialized by this cell.
+
+\item \B{PRIORITY} \\
+The cell with the higher integer value in this parameter wins an initialization conflict.
+\end{itemize}
+
The HDL frontend models a memory using RTLIL::Memory objects and asynchronous
{\tt \$memrd} and {\tt \$memwr} cells. The {\tt memory} pass (i.e.~its various sub-passes) migrates
{\tt \$dff} cells into the {\tt \$memrd} and {\tt \$memwr} cells making them synchronous, then
@@ -295,6 +323,9 @@ The number of address bits.
\item \B{WIDTH} \\
The number of data bits per word.
+\item \B{INIT} \\
+The initial memory contents.
+
\item \B{RD\_PORTS} \\
The number of read ports on this memory cell.
@@ -345,9 +376,11 @@ This input is \B{WR\_PORTS}*\B{ABITS} bits wide, containing all address signals
This input is \B{WR\_PORTS}*\B{WIDTH} bits wide, containing all data signals for the write ports.
\end{itemize}
-The {\tt techmap} pass can be used to manually map {\tt \$mem} cells to
-specialized memory cells on the target architecture, such as block ram resources
-on an FPGA.
+The {\tt memory\_collect} pass can be used to convert discrete {\tt \$memrd}, {\tt \$memwr}, and {\tt \$meminit} cells
+belonging to the same memory to a single {\tt \$mem} cell, whereas the {\tt memory\_unpack} pass performs the inverse operation.
+The {\tt memory\_dff} pass can combine asynchronous memory ports that are fed by or feeding registers into synchronous memory ports.
+The {\tt memory\_bram} pass can be used to recognize {\tt \$mem} cells that can be implemented with a block RAM resource on an FPGA.
+The {\tt memory\_map} pass can be used to implement {\tt \$mem} cells as basic logic: word-wide DFFs and address decoders.
\subsection{Finite State Machines}
@@ -371,9 +404,15 @@ Verilog & Cell Type \\
\hline
\lstinline[language=Verilog]; Y = ~A; & {\tt \$\_NOT\_} \\
\lstinline[language=Verilog]; Y = A & B; & {\tt \$\_AND\_} \\
+\lstinline[language=Verilog]; Y = ~(A & B); & {\tt \$\_NAND\_} \\
+\lstinline[language=Verilog]; Y = A & ~B; & {\tt \$\_ANDNOT\_} \\
\lstinline[language=Verilog]; Y = A | B; & {\tt \$\_OR\_} \\
+\lstinline[language=Verilog]; Y = ~(A | B); & {\tt \$\_NOR\_} \\
+\lstinline[language=Verilog]; Y = A | ~B; & {\tt \$\_ORNOT\_} \\
\lstinline[language=Verilog]; Y = A ^ B; & {\tt \$\_XOR\_} \\
+\lstinline[language=Verilog]; Y = ~(A ^ B); & {\tt \$\_XNOR\_} \\
\lstinline[language=Verilog]; Y = S ? B : A; & {\tt \$\_MUX\_} \\
+\lstinline[language=Verilog]; Y = EN ? A : 'bz; & {\tt \$\_TBUF\_} \\
\hline
\lstinline[language=Verilog]; always @(negedge C) Q <= D; & {\tt \$\_DFF\_N\_} \\
\lstinline[language=Verilog]; always @(posedge C) Q <= D; & {\tt \$\_DFF\_P\_} \\
@@ -396,9 +435,10 @@ $ClkEdge$ & $RstLvl$ & $RstVal$ & Cell Type \\
\end{table}
Table~\ref{tab:CellLib_gates} lists all cell types used for gate level logic. The cell types
-{\tt \$\_NOT\_}, {\tt \$\_AND\_}, {\tt \$\_OR\_}, {\tt \$\_XOR\_} and {\tt \$\_MUX\_}
-are used to model combinatorial logic. The cell types {\tt \$\_DFF\_N\_} and {\tt \$\_DFF\_P\_}
-represent d-type flip-flops.
+{\tt \$\_NOT\_}, {\tt \$\_AND\_}, {\tt \$\_NAND\_}, {\tt \$\_ANDNOT\_}, {\tt \$\_OR\_}, {\tt \$\_NOR\_},
+{\tt \$\_ORNOT\_}, {\tt \$\_XOR\_}, {\tt \$\_XNOR\_} and {\tt \$\_MUX\_} are used to model combinatorial logic.
+The cell type {\tt \$\_TBUF\_} is used to model tristate logic.
+The cell types {\tt \$\_DFF\_N\_} and {\tt \$\_DFF\_P\_} represent d-type flip-flops.
The cell types {\tt \$\_DFF\_NN0\_}, {\tt \$\_DFF\_NN1\_}, {\tt \$\_DFF\_NP0\_}, {\tt \$\_DFF\_NP1\_},
{\tt \$\_DFF\_PN0\_}, {\tt \$\_DFF\_PN1\_}, {\tt \$\_DFF\_PP0\_} and {\tt \$\_DFF\_PP1\_} implement
@@ -410,7 +450,7 @@ otherwise.
\begin{lstlisting}[mathescape,language=Verilog]
always @($ClkEdge$ C, $RstEdge$ R)
if (R == $RstLvl$)
- Q <= $RstVa$l;
+ Q <= $RstVal$;
else
Q <= D;
\end{lstlisting}
@@ -426,6 +466,10 @@ Add information about {\tt \$assert}, {\tt \$assume}, {\tt \$live}, {\tt \$fair}
\end{fixme}
\begin{fixme}
+Add information about {\tt \$specify2}, {\tt \$specify3}, and {\tt \$specrule} cells.
+\end{fixme}
+
+\begin{fixme}
Add information about {\tt \$slice} and {\tt \$concat} cells.
\end{fixme}
@@ -450,7 +494,6 @@ Add information about {\tt \$\_DFFE\_??\_}, {\tt \$\_DFFSR\_???\_}, {\tt \$\_DLA
\end{fixme}
\begin{fixme}
-Add information about {\tt \$\_NAND\_}, {\tt \$\_NOR\_}, {\tt \$\_XNOR\_}, {\tt \$\_ANDNOT\_}, {\tt \$\_ORNOT\_},
-{\tt \$\_AOI3\_}, {\tt \$\_OAI3\_}, {\tt \$\_AOI4\_}, and {\tt \$\_OAI4\_} cells.
+Add information about {\tt \$\_AOI3\_}, {\tt \$\_OAI3\_}, {\tt \$\_AOI4\_}, and {\tt \$\_OAI4\_} cells.
\end{fixme}
diff --git a/manual/CHAPTER_Overview.tex b/manual/CHAPTER_Overview.tex
index ae5cf094..4136efed 100644
--- a/manual/CHAPTER_Overview.tex
+++ b/manual/CHAPTER_Overview.tex
@@ -331,8 +331,9 @@ to update {\tt \textbackslash{}q}.
An RTLIL::Process is a container for zero or more RTLIL::SyncRule objects and
exactly one RTLIL::CaseRule object, which is called the {\it root case}.
-An RTLIL::SyncRule object contains an (optional) synchronization condition
-(signal and edge-type) and zero or more assignments (RTLIL::SigSig).
+An RTLIL::SyncRule object contains an (optional) synchronization condition (signal and edge-type) and zero or
+more assignments (RTLIL::SigSig). The {\tt always} synchronization condition is used to break combinatorial
+loops when a latch should be inferred instead.
An RTLIL::CaseRule is a container for zero or more assignments (RTLIL::SigSig)
and zero or more RTLIL::SwitchRule objects. An RTLIL::SwitchRule objects is a
@@ -350,6 +351,18 @@ and this bit is a one (the second ``1'').} for {\tt \textbackslash{}reset == 1}
sets {\tt \$0\textbackslash{}q[0:0]} to the value of {\tt \textbackslash{}d} if {\tt
\textbackslash{}enable} is active (lines $6 \dots 11$).
+A case can specify zero or more compare values that will determine whether it matches. Each of the compare values
+must be the exact same width as the control signal. When more than one compare value is specified, the case matches
+if any of them matches the control signal; when zero compare values are specified, the case always matches (i.e.
+it is the default case).
+
+A switch prioritizes cases from first to last: multiple cases can match, but only the first matched case becomes
+active. This normally synthesizes to a priority encoder. The {\tt parallel\_case} attribute allows passes to assume
+that no more than one case will match, and {\tt full\_case} attribute allows passes to assume that exactly one
+case will match; if these invariants are ever dynamically violated, the behavior is undefined. These attributes
+are useful when an invariant invisible to the synthesizer causes the control signal to never take certain
+bit patterns.
+
The lines $13 \dots 16$ then cause {\tt \textbackslash{}q} to be updated whenever there is
a positive clock edge on {\tt \textbackslash{}clock} or {\tt \textbackslash{}reset}.
@@ -428,8 +441,8 @@ memory object has the following properties:
All read accesses to the memory are transformed to {\tt \$memrd} cells and all write accesses to
{\tt \$memwr} cells by the language frontend. These cells consist of independent read- and write-ports
-to the memory. The \B{MEMID} parameter on these cells is used to link them together and to the
-RTLIL::Memory object they belong to.
+to the memory. Memory initialization is transformed to {\tt \$meminit} cells by the language frontend.
+The \B{MEMID} parameter on these cells is used to link them together and to the RTLIL::Memory object they belong to.
The rationale behind using separate cells for the individual ports versus
creating a large multiport memory cell right in the language frontend is that