( constant x : U32 , 0 )A definition is a unit of syntax that introduces a name N and associates some code with N.
+In FPP, there are two kinds of definitions:
+Global definitions that introduce globally visible +qualified names.
+Bound parameter definitions introduced by binding +template arguments to +template parameters within +the scope of an expanded template. +A bound parameter definition is a pair \$(p, A)\$ consisting of +a template parameter p and a template argument A. +The name associated with a bound parameter definition \$(p, A)\$ +is the identifier specified in p. +Bound parameter definitions come into scope +during +template expansion.
+Within an expanded template, a bound parameter definition may represent +a global definition, either directly or indirectly. +A bound parameter definition \$B = (p, A)\$ represents +a global definition D if (1) A +refers to D or (2) +A refers to a bound parameter definition B' that represents D.
+Not all bound parameter definitions represent global definitions. +For example, the bound parameter definition
+( constant x : U32 , 0 )does not represent any global definition.
+The rest of this section discusses the global definitions.
+A module template definition associates a name +with a module template. +A module template is a set of syntactic elements, each of which +becomes a member of a translation unit or a +module definition +at the point where the template is expanded. +Each template may be expanded any number of times, and +each expansion creates a distinct set of translation unit +or module members. +Templates may have parameters that become bound +to arguments +at the point of expansion.
+module template
+identifier
+[
+( template-param-list )
+]
+{
+module-member-sequence
+}
module-member-sequence is identical to the module member sequence +of a module definition.
+The identifier is the name of the template.
+No module template definition may be the ancestor in the +abstract syntax tree of another module template definition +or of a template expansion specifier. +This rule ensures that expanding a template +never causes another template to be defined or another +template expansion to occur.
+Each module template definition M that appears in a model +must pass the following analysis, regardless of whether it +is expanded.
+Expand M at the point of the definition of M, providing the following +arguments for the template +parameters:
+For each constant template parameter, a literal value of the correct type.
+For each port interface instance template argument, a component +instance definition that minimally satisfies the interface.
+For each type template argument, an abstract type definition.
+In the expansion E, do the following:
+Construct the use-def +graph, checking for cycles.
+Compute the types associated with type definition symbols, enumerated +constant symbols, and type names, except that array size expressions +and default value expressions are still unevaluated.
+Perform type checking, +except that array size expressions and default value expressions in type +definitions are still unevaluated.
+template MT(
+ constant baseId: FwIdType
+ instance comDriver: Drv.ByteStreamDriver
+) {
+
+ instance comStub: Svc.ComStub base id \
+ baseId + 0x0800
+
+ topology T {
+ instance comDriver
+ instance comStub
+
+ # Connect to the driver from the parameter list
+ connections Downlink {
+ comStub.drvSendOut -> comDriver.$send
+ comDriver.sendReturnOut -> comStub.drvSendReturnIn
+ comDriver.ready -> comStub.drvConnected
+ }
+ }
+
+}An port definition defines an F Prime port. A port is the endpoint of a connection between F Prime components.
@ Port 1
@@ -3018,14 +3237,14 @@ 5.10.3. Examples
A port interface definition defines a port interface and associates a name with it.
interface identifier
{ port-interface-member-sequence }
The identifier is the name of the interface.
@ A command interface
@@ -3149,7 +3368,7 @@ 5.11.3. Examples
A state machine definition defines a state machine.
state machine identifier
[ { state-machine-member-sequence } ]
An internal state machine M has the following runtime behavior:
state machine MonitorSm {
@@ -3678,13 +3897,13 @@ 5.12.4. Examples
A struct definition defines a new structure type and associates a name with it.
[
dictionary
@@ -3711,7 +3930,7 @@
The identifier is the name N of the type. The definition associates the name N with a @@ -3763,7 +3982,7 @@
# Defines a struct type A with members x and y
@@ -3798,14 +4017,14 @@ 5.13.3. Examples
A system definition defines an embedded software system. When a system definition appears in the model, it causes code and a ground dictionary to be generated.
@ The spacecraft system
@@ -3844,7 +4063,7 @@ 5.14.3. Examples
A topology definition defines an F Prime topology, that is, a set of objects with ports (component instances and subtopologies) @@ -3865,7 +4084,7 @@
[ deployment ]
topology
@@ -3907,7 +4126,7 @@
If present, the optional keyword deployment indicates that the topology
is a deployment topology.
@@ -3965,9 +4184,9 @@
Let S be the set of all topologies +represented by port +interface instance specifiers in D. Recursively resolve all the topologies in S.
For each connection C discovered in step 3, check that each @@ -4018,22 +4237,27 @@
. I, and do the following:
If E is a component instance endpoint, -then do nothing.
+If E is a component +instance endpoint, check that the associated +port interface +of the component instance contains the port instance I.
Otherwise the port instance identifier
-P appearing in E has the form Q
-. I, where Q refers to a topology T', and
-I names a port of T'.
Otherwise if E is a topology +instance endpoint, where Q refers to a topology T', then
Check that the associated port interface of T' contains the +port instance I,
+Find the topology port instance specifier port I = P' in T'.
Check that the port interface of p contains the port instance I.
+Replace B with N in the endpoint E. +Retain the port number assignment, if any, at E computed in step 1.
+Reapply steps a, b, and c to the endpoint E.
+When a topology T has a port interface instance specifier instance Q
as a member, and Q refers to a topology
@@ -4389,7 +4633,7 @@
For purposes of port numbering, FPP orders connections between component instances as follows.
@@ -4478,7 +4722,7 @@When generating a dictionary from a deployment topology T, the analyzer treats the definition of T as if it contained @@ -4491,7 +4735,7 @@
Example 1.
Example 6 (Topology Ports).
+topology A {
+
+ instance b
+
+ port p = b.p
+
+}
+
+topology B {
+
+ instance a
+
+ instance B
+
+ connections C {
+ a.p -> B.p
+ }
+
+}Here we have imported topology B with the instance keyword, emphasizing
+that B is to be treated as an interface instance with ports.
+We could also have used import as in the examples above.
+See the section on port
+interface instance specifiers.
Import resolution produces this topology:
+topology A {
+
+ instance a
+ instance b
+
+ connections C {
+ a.p -> b.p
+ }
+
+}Example 7 (Topology Templates).
+port P
+
+interface IOut {
+ output port p: P
+}
+
+passive component COut {
+ import IOut
+ output port pOther: P
+}
+
+passive component CIn {
+ sync input port p: P
+}
+
+instance cOut: COut base id 0x100
+instance cIn: CIn base id 0x200
+
+module template T(
+ # Introduce an instance by template parameter.
+ # At this point we don't know what this instance is.
+ # All we know is that it has the port interface IOut.
+ instance iOut: IOut
+) {
+ topology T {
+ # On expansion, the topology T will have two instances: iOut and cIn.
+ # iOut will refer to the bound parameter definition with name iOut.
+ # cIn will refer to the component instance cIn.
+ instance iOut
+ instance cIn
+ connections C {
+ # Connection goes from the port interface of iOut to
+ # the port interface of cIn
+ iOut.p -> cIn.p
+ # We must connect ports in the interfaces of the instances named.
+ # For example, a connection iOut.pOther -> cIn.p is not allowed here,
+ # because pOther is not in the port interface of iOut.
+ # This is true even if iOut is eventually bound to cOut, which has the
+ # output port pOther.
+ }
+ }
+}
+
+# In the expansion, the bound parameter (iOut: IOut, cOut) is in scope.
+# After resolution, the connection is cOut.p -> cIn.p.
+expand T(cOut)A type definition is one of the following:
Several definitions are used specially by the F Prime framework. These are called framework definitions.
The table below lists the framework constant definitions. Each row in the table has the following meaning:
@@ -4968,7 +5315,7 @@The table below lists the framework type definitions. The rows have the same meaning as stated for @@ -5094,7 +5441,7 @@
The F Prime framework provides port definitions with the following qualified names:
@@ -5145,7 +5492,7 @@If the optional keyword dictionary appears in a
type definition or
@@ -5156,7 +5503,7 @@
If a constant definition D is a dictionary definition, then the expression appearing in D must have one of the following types:
@@ -5183,7 +5530,7 @@dictionary type T = U32
@@ -6519,12 +6866,16 @@ 7.2.2. Semantics
identifier appearing in E.
-The port interface instance Q must exist in the enclosing
-topology,
+
The port interface instance Q must exist in the enclosing topology,
after
resolution to a partially numbered topology.
-If Q is a component instance, then E is a component instance endpoint.
-Otherwise Q is a topology, and E is a topology endpoint.
+If Q refers to a component
+instance definition, then E is a component instance endpoint.
+Otherwise if Q refers to a topology
+definition, then E is a topology endpoint.
+Otherwise Q refers to a port
+interface instance parameter definition, and E is a bound parameter
+endpoint.
The optional expression e following I, if it is present,
@@ -7030,7 +7381,8 @@
7.5.2. Semantics
location of the include specifier.
During parsing, the translator does the following:
+During parsing, the translator does the +following:
All include specifiers are resolved this way, even ones that occur inside or +transitively inside a template +definition.
+The suffix .fppi is conventional for included files, to distinguish them
from files presented directly for analysis or translation.
A template specifier locate template
+qual-ident at
+string-literal
A system location specifier locate system
qual-ident at
string-literal
A template specifier refers to a +template definition.
+A type location specifier refers to a type definition.
A port interface instance specifier specifies that a -component instance -I or a -topology T' +port interface instance I is part of a topology T. -I or T' provides a port interface +I provides a port interface that can be used in connections in T.
The qualified identifier must refer to a -port interface instance, -i.e., a component instance definition or a topology definition.
+port interface instance. +Every port interface instance specifier represents a component instance +definition or topology definition D. +A port interface instance specifier S represents D +if the port interface instance I to which S refers +represents D.
If the qualified identifier refers to a topology definition D, @@ -8809,83 +9179,224 @@
A topology port instance specifier specifies a -port instance in a topology definition -that is analogous to a port instance in a -component definition. -A port instance specifier in a topology T maps a name to a port instance -of a component instance that is part of T. -Such a specifier can be the endpoint of a -connection.
+A template expansion specifier names a +template +and specifies arguments for the parameters of the template. +During template expansion, it generates the definitions specified +by the template and the arguments.
port
-identifier
-=
-port-instance-id
expand
+qual-ident
+[
+( template-arg-list )
+]
template-arg-list is an +element sequence in +which the elements are template arguments, and +the terminating punctuation is a comma.
+Each template argument is one of the following:
The identifier specifies the name N of the port instance.
+A constant template argument constant expression
The port instance identifier must be valid and refer to a port instance I.
+A port interface instance template argument
+instance qual-ident
The whole specifier makes I part of the - interface of the enclosing -topology, after renaming it with name N.
+A type template argument type type-name
topology T {
- instance c1
- instance c2
-
- connections {
- c1.portInternal -> c2.portInternal
- }
-
- port portOut = c1.portOut
- port customPortName = c2.portIn
-}
-
-topology Y {
- instance T
-
- instance c3
-
- connections {
- T.portOut -> c3.portIn
- c3.portOut -> T.customPortName
- }
-}See also the examples for topology -definitions.
-The qualified identifier must +refer to +the definition of a +module template T.
+The number of template arguments must match the number of template +parameters in T. +If there are no parameters, then the argument list may be omitted.
+For each template parameter p, let A be the corresponding argument.
+If p is a constant template parameter, then A must be a constant +template argument. The type of the expression in A must be +convertible to the type specified in p.
+If p is an interface instance template parameter, then A +must be an interface instance template argument. +The qualified identifier in A must +refer to +a port interface instance N. +The port interface of referred to in p must be a +sub-interface of the port +interface of N.
+If p is a type template parameter, then A must be a type +template argument.
+Within the expansion generated by the +template expansion specifier, the pair \$B = (p, A)\$ is +in +scope as a definition and is called a bound parameter definition. +There are three kinds of bound parameter definitions:
+When p is a constant template parameter, B is called a +constant parameter definition.
+When p is a port interface instance template parameter, B +is called a port interface instance parameter definition.
+When p is a type template parameter, B +is called a type parameter definition.
+port P
+
+@ A basic template parameterizing an interface port count
+module template T(constant numPorts: U32) {
+
+ interface I {
+ output port p: [numPorts] P
+ }
+
+ passive component C {
+ import I
+ }
+
+}
+
+@ Instantiate the template giving the parameters concrete values
+module M {
+
+ expand T(constant 30)
+ # Bound parameter definition (constant numPorts: U32, 30)
+ # Global definition interface I { output port p: [numPorts] P }
+ # Global definition passive component C { output port p: [numPorts] P }
+
+}A topology port instance specifier specifies a +port instance in a topology definition +that is analogous to a port instance in a +component definition. +A port instance specifier in a topology T maps a name to a port instance +of a component instance that is part of T. +Such a specifier can be the endpoint of a +connection.
+port
+identifier
+=
+port-instance-id
The identifier specifies the name N of the port instance.
+The port instance identifier must be valid and refer to a port instance I.
+The whole specifier makes I part of the + interface of the enclosing +topology, after renaming it with name N.
+topology T {
+ instance c1
+ instance c2
+
+ connections {
+ c1.portInternal -> c2.portInternal
+ }
+
+ port portOut = c1.portOut
+ port customPortName = c2.portIn
+}
+
+topology Y {
+ instance T
+
+ instance c3
+
+ connections {
+ T.portOut -> c3.portIn
+ c3.portOut -> T.customPortName
+ }
+}See also the examples for topology +definitions.
+A instance member identifier is a unit of syntax that refers to a member of a component instance or a topology. @@ -9155,13 +9666,30 @@
If Q refers to a global definition D, +then Q names the +type defined in D.
+Otherwise Q refers to a bound parameter definition B. +In this case, B must be a +type template parameter +definition, and Q names the type defined in the global definition +represented by B.
+For example:
@@ -9623,19 +10151,48 @@An identifier expression is an -identifier -that refers to a -constant definition -or +
+The identifier must refer to a +constant definition, enumerated constant definition, +or bound parameter definition according to the -rules for resolving identifiers.
+rules for resolving identifiers. +If the identifier refers to a bound parameter definition B, +then B must be a constant template +parameter definition.Example:
+An identifier expression is evaluated as follows:
If the identifier refers to a constant definition or enumerated constant +definition D, then the value of the expression is the value provided +by D.
+Otherwise the identifier refers to a bound parameter definition
+\$B = (p, A)\$.
+By the rules for binding arguments
+to template parameters,
+A is constant e for some expression e.
+In this case the value is provided by evaluating the expression e.
constant a = 42
@@ -9643,6 +10200,7 @@ 10.7. Identifier Expressions
A template parameter list is an +element sequence +in which the elements are template parameters, +and the terminating punctuation is a comma.
+A template parameter is a name to which a +template argument +becomes bound when a template is expanded. +Within the scope of an expanded template, a the bound parameter is a +definition.
+Each template parameter is one of the following:
+A constant template parameter constant
+identifier : type-name
A port interface instance template parameter instance
+ identifier :
+qual-ident
A type template parameter type
+identifier
The type name in a constant template parameter must be +a valid type name.
+The qualified identifier in a port interface instance template parameter +must refer to +a port interface definition.
+constant c: U32
+instance cmdDisp: Svc.CmdDispatcher
+type TA format string is a string that specifies the display format @@ -10216,10 +10837,10 @@
A comment is model text that is ignored by the translator. It provides information to human readers of the source model.
@@ -10244,7 +10865,7 @@An annotation is similar to a comment, but it is attached to a @@ -10254,7 +10875,7 @@
Annotations can occur at the following syntax elements:
There are two kinds of annotations: pre-annotations and post-annotations.
@@ -10329,7 +10950,7 @@You can write several pre-annotations in a row before an annotatable element e. @@ -10370,10 +10991,10 @@
A translation unit forms part of a model at the top level. @@ -10382,7 +11003,7 @@
A translation unit is an element sequence in which each @@ -10393,7 +11014,7 @@
Here is a translation unit:
A model is a collection of one or more
translation
@@ -10460,7 +11081,7 @@ 14.2. Models
According to the -semantics +semantics of module definitions, this is also equivalent:
Every syntactic element E in a source model has an associated location L. @@ -10509,27 +11130,29 @@
A qualified identifier is one of the following:
+A qualified identifier qual-ident is one of the following:
An -identifier.
+Q . I, where Q is a qualified identifier and I is an
-identifier.
qual-ident . identifier
Examples:
+Therefore every qualified identifier consists of +a nonempty list of identifiers separated by dots.
+Examples:
Every -definition +
Global definitions: +Every global definition D appearing in an FPP model has a unique qualified name. The qualified name is a qualified identifier formed as follows:
@@ -10577,9 +11200,14 @@Bound parameter definitions: +Every bound parameter definition is referred to by an +identifier in the scope of an expanded template.
+The qualified names of definitions reside in the following name groups:
@@ -10602,6 +11230,9 @@The state machine name group
The template name group
+The system name group
Constant definitions reside in the value name group.
Constant parameter definitions reside in the value name group.
+Enum definitions reside in both the value name group and the type name group.
Module definitions reside in all the name groups.
Module template definitions reside in the template name group.
+Port definitions reside in the port name group.
Port interface definitions reside in the port interface name group.
Port interface instance parameter definitions reside in the +port interface instance name group.
+State machine definitions reside in the state machine name group, the value name group, and the type name group.
Struct type definitions reside in the type name group.
Topology definitions reside in the port interface instance name group.
+System definitions reside in the system name group.
Topology definitions reside in the topology name group.
+Type parameter definitions reside in the type name group.
Notice that elements with scopes reside in the name groups of items -that they may enclose. -For example, enums reside in the value name group because they -enclose constant definitions.
-Two definitions with the same qualified name are allowed if they are in +
Two global definitions with the same qualified name are allowed if they are in different name groups. For example:
Multiple syntactic module definitions with the same qualified name are allowed. The semantic analysis combines all such definitions into a @@ -10732,7 +11370,7 @@
Within the same name group, two definitions with the same qualified name are not allowed, unless they are both module definitions @@ -10761,36 +11399,89 @@
Within the scope of a single expanded template T, +the following rules apply:
+No two template parameters of T +in the same name group may have the same name.
+No template parameter of T may have the same name as +the unqualified name of any definition in the same name group +appearing in the expansion of T.
+The purpose of these rules is to ensure that all the template parameters +and all the definitions in the body of the template can coexist +in the same scope.
+For example, the following code violates these rules:
+module template T(constant x: U32, constant x: I32) { } # Reuse of the parameter name x
+module template T(constant x: U32) {
+ constant x = 0 # constant x conflicts with parameter x
+}On the other hand, this code is allowed:
+module template T(constant x: U32, type x) { } # OK: constant x, type x
+module template T(constant x: U32) {
+ type x = U32 # OK: constant x, type x
+}The following rules govern the resolution of identifiers, i.e., associating identifiers with definitions:
To resolve an identifier I outside of any +expanded template, do the following:
+Use the context to determine which name group -S to use. For example, if we are expecting a type name, then use the +G to use. For example, if we are expecting a type name, then use the type name group.
At the top level (outside the brace-delimited body of any definition), the +
At the top level scope (outside the brace-delimited body of any definition), the identifier I refers to the unique definition with qualified name I if it -exists in name group S. Otherwise an error results.
+exists in name group G. Otherwise an error results.Inside the brace-delimited body of a definition with +
Inside the brace-delimited scope of a definition with qualified name N appearing at the top level:
The identifier I refers to the definition with qualified name
-N . I if it exists in name group S.
. I if it exists in name group G.
Otherwise I refers to the definition with qualified name I if it -exists in name group S.
+exists in name group G.Otherwise an error results.
@@ -10799,13 +11490,13 @@Inside the brace-delimited body of a definition with qualified name +
Inside the brace-delimited scope of a definition with qualified name N appearing inside the body of a definition D:
The identifier I refers to the definition with qualified name
-N . I if it exists in name group S.
. I if it exists in name group G.
Otherwise proceed as if I were appearing inside D.
@@ -10819,78 +11510,220 @@Example:
S refers to the value name group.
+G refers to the value name group.
# Identifier M is in scope in S and refers to the qualified name M
-# Identifier a is in scope in S and refers to qualified name a
+# Identifier M is in scope in G and refers to the qualified name M
+# Identifier a is in scope in G and refers to qualified name a
-constant a = 1 # Unique definition in S with qualified name a
+constant a = 1 # Unique definition in G with qualified name a
module M {
- # Identifier M is in scope in S and refers to the qualified name M
- # Identifier N is in scope in S and refers to the qualified name N
- # Identifier a is in scope in S and refers to qualified name a
- # Identifier b is in scope in S and refers to qualified name M.b
- constant b = 2 # Unique definition in S with qualified name M.b
+ # Identifier M is in scope in G and refers to the qualified name M
+ # Identifier N is in scope in G and refers to the qualified name N
+ # Identifier a is in scope in G and refers to qualified name a
+ # Identifier b is in scope in G and refers to qualified name M.b
+ constant b = 2 # Unique definition in G with qualified name M.b
}
-# Identifier M is in scope in S and refers to the qualified name M
-# Identifier a is in scope in S and refers to qualified name a
+# Identifier M is in scope in G and refers to the qualified name M
+# Identifier a is in scope in G and refers to qualified name a
module M {
- # Identifier M is in scope in S and refers to the qualified name M
- # Identifier N is in scope in S and refers to the qualified name M.N
+ # Identifier M is in scope in G and refers to the qualified name M
+ # Identifier N is in scope in G and refers to the qualified name M.N
# Identifier a is in scope and refers to qualified name a
# Identifier b is in scope and refers to qualified name M.b
module N {
- # Identifier M is in scope in S and refers to the qualified name M
- # Identifier N is in scope in S and refers to the qualified name M.N
- # Identifier a is in scope in S and refers to qualified name a
- # Identifier b is in scope in S and refers to qualified name M.N.b
- constant b = 3 # Unique definition in S with qualified name M.N.b
+ # Identifier M is in scope in G and refers to the qualified name M
+ # Identifier N is in scope in G and refers to the qualified name M.N
+ # Identifier a is in scope in G and refers to qualified name a
+ # Identifier b is in scope in G and refers to qualified name M.N.b
+ constant b = 3 # Unique definition in G with qualified name M.N.b
}
}
-# Identifier M is in scope in S and refers to the qualified name M
-# Identifier a is in scope in S and refers to qualified name a
+# Identifier M is in scope in G and refers to the qualified name M
+# Identifier a is in scope in G and refers to qualified name aTo resolve an identifier I inside an +expanded template, do the following:
+Let E be the expanded template containing I. +By the rules +for template definitions, E is guaranteed to be unique.
+Let T be the template definition referred to by E.
+At the point of T, temporarily insert a definition D +with a brace-delimited scope and with the following properties:
+The scope of D contains the following definitions: (a) all the + bound parameter definitions of E and (b) all the +global definitions introduced in the expansion E.
+The name of D is different from any name appearing as a qualifier in the +model. +Therefore the model uses only simple identifiers to refer to the +definitions in D.
+Use the rules stated in the previous section to resolve I +as if it appeared in D.
+Example 1:
+module M {
+ constant a = 0
+ module template T {
+ constant b = a
+ }
+}
+module N {
+ # Inside the expansion of M.T, names are resolved
+ # in a new scope at the point of the definition of M.T.
+ # So a refers to M.a, even though the expansion
+ # occurs inside N.
+ expand M.T
+}
+# The expansion introduces a definition b in module N,
+# so N.b is in scope here. Definitions introduced by
+# template expansion follow the standard rules for scoping.
+constant c = N.bExample 2:
+module template T(constant a: U32) {
+ constant b = a
+}
+
+# The expansion of T introduces a definition constant b = a.
+# Inside the expansion of T, the bound parameter definition
+# (constant a: U32, 0) is in scope, and the name a refers
+# to this definition.
+expand T(0)The standard way to implement the resolution semantics described in the +previous two sections is with stacks of scopes. +In this implementation, a scope is a map from identifiers to definitions. +At each point P in the model, there are two logical stacks for each name +group G:
+A definition stack describing how definitions at P in name group G are +nested in +enclosing definitions.
+A resolution stack describing how identifiers at P in name group G are +to be resolved.
+At each point at the top level of the model (outside of any definition), the +definition stack has one scope, the top-level scope. +At each point inside a definition, we construct the definition stack +by (1) starting with the definition stack at the point of the definition +and (2) pushing a new scope containing the global definitions in the body of +the definition.
+Outside of any expanded template, the definition stack and the resolution +stack are the same. +Inside an expanded template E that refers to a template definition T, +we construct the resolution stack by +(1) starting with the definition stack at the point of T and (2) +pushing a new scope containing the bound parameter definitions and +global definitions of E.
+In all cases, to resolve an identifier I in name group G at point P, we +search +upwards in the resolution stack for name group G at P until one of two +things happens:
+We find I as a key in some scope S. In this case I +refers to the definition to which I is mapped in S.
+We reach the top level scope S and find that I does +not exist as a key in this scope. +In this case we report an undefined symbol error.
+The following rules govern the resolution of qualified identifiers, i.e., -associating qualified identifiers with definitions:
+associating qualified identifiers Q with definitions:If a qualified identifier is an identifier, then resolve it as -stated in the +
If Q is an identifier I, then resolve I as stated in the previous section.
Otherwise, the qualified identifier has the form Q . I, where
-Q is a qualified identifier and I is an identifier. Do the
+
Otherwise Q has the form Q' . I, where
+Q' is a qualified identifier and I is an identifier. Do the
following:
Recursively resolve Q.
+Recursively resolve Q'.
If Q refers to a definition D +
If Q' refers to a definition D with a brace-delimited body, then do the following:
Determine the name group S of Q .
+
Determine the name group S of Q'
+.
I.
module M {
+constant a = 0
+module M {
constant a = 0
enum E {
b = 2
@@ -10932,16 +11766,14 @@ 15.6. Resolution o
A use is a qualified identifier -that refers to a -definition -according to the +that refers to a definition according to the scoping rules for names.
The set of definitions and non-qualifying uses @@ -11042,7 +11874,7 @@
So long as the use-def graph is acyclic, there is no @@ -11083,7 +11915,7 @@
At certain points in the analysis, the analyzer may treat an element of the model as if it contained uses of one or more @@ -11117,7 +11949,7 @@
An error also occurs if the rules for resolving dot expressions -causes an implied constant use to be resolved to a member +cause an implied constant use to be resolved to a member of a struct value instead of a constant definition. For example, suppose this definition exists in the model:
Expansion of templates performs partial analysis +and syntactic expansion. +It produces a new model in which each +template +expansion specifier S is transformed into an +expanded template \$E = S { X }\$, where X +is a copy of the syntax provided +by the template definition to which S refers. +The expanded template E is not expressible in the syntax +of the source language, but it is expressible in the internal +representation of the model. +The syntax X behaves as if it were literally included in +the model at the point of E, except that there are +special rules for resolving +resolving names that appear inside E.
+Let M be the model initially presented to the analyzer, +with include specifiers resolved. +Construct the output model M' as follows:
+Perform enough analysis on M to + check all the template +definitions and to match all template expansion + specifiers with the template definitions to which they refer.
+For each template expansion specifier S in M, +do the following:
+Let T be the template definition referred to in S.
+Let X be a copy of the syntax specified in the body of T.
+Replace S with \$S { X }\$.
+Model M:
+module M {
+
+ constant a = 0
+
+ module template T(constant b: U32) {
+
+ constant c = a + b
+
+ }
+
+}
+
+module N {
+
+ expand M.T(constant 1)
+
+}Model M':
+module M {
+
+ constant a = 0
+
+ module template T(constant b: U32) {
+
+ constant c = a + b
+
+ }
+
+}
+
+module N {
+
+ # The expanded template is expand M.T(constant 1) { constant c = a + b }
+ expand M.T(constant 1)
+
+}A type is a set of values.
The primitive integer types correspond to the primitive integer type names @@ -11155,7 +12096,7 @@
The floating-point types correspond to the floating-point type names @@ -11163,7 +12104,7 @@
The primitive integer types together with the floating-point types are @@ -11171,7 +12112,7 @@
The Boolean type corresponds to the Boolean type name. @@ -11179,7 +12120,7 @@
Together, the primitive numeric types and the @@ -11188,7 +12129,7 @@
The string types correspond to the string type names. @@ -11221,7 +12162,7 @@
An array type is a type associated with an array definition. @@ -11234,7 +12175,7 @@
An enum type is a type associated with an enum definition. @@ -11245,7 +12186,7 @@
A struct type is a type associated with a struct definition. @@ -11256,7 +12197,7 @@
An abstract type is a type associated with an abstract type definition. @@ -11267,7 +12208,7 @@
An alias type is a type associated with an alias type definition. @@ -11278,20 +12219,20 @@
Internal types do not have syntactic names in FPP source models. The compiler assigns these types to expressions during type checking.
The type Integer represents all integer values, without regard to bit width.
Integer together with the primitive integer types are called @@ -11299,7 +12240,7 @@
Integer together with the primitive numeric types are called @@ -11307,7 +12248,7 @@
The type [ n ] T, where n is an integer and T is a type, represents an array of n elements, @@ -11318,7 +12259,7 @@
The type { \$m_1\$ : \$T_1, ...,\$ \$m_n\$ : \$T_n\$ }, where each \$m_i\$ is an identifier and each \$T_i\$ is a type, @@ -11332,13 +12273,13 @@
A type is a canonical type if it is not an alias type.
The underlying type of a type T is the canonical type associated with T.
@@ -11357,7 +12298,7 @@A displayable type is a type that the F Prime ground data system can display. It is one of the following:
@@ -11390,7 +12331,7 @@A type has numeric members if its underlying type is one of the following:
@@ -11414,7 +12355,7 @@The alias list of a type \$T\$ is a list \$L\$ defined as follows:
Every type T with a syntactic name in FPP has an associated default value. @@ -11501,7 +12442,7 @@
Every displayable type has a serialized size. This is the maximum number of bytes required to @@ -11579,7 +12520,7 @@
In this section, we explain the rules used to assign types to @@ -11607,7 +12548,7 @@
The type of a
boolean
@@ -11653,7 +12594,7 @@ 18.3. Boolean Literals
The type of a
string
@@ -11661,7 +12602,7 @@ 18.4. String Literals
To type an identifier expression \$e\$, the semantic analyzer @@ -11679,7 +12620,7 @@
To type an array expression @@ -11769,7 +12710,7 @@
To type an array subscript expression @@ -11799,13 +12740,13 @@
To type a struct expression @@ -11837,7 +12778,7 @@
To type a binary expression \$e_1\$ op \$e_2\$ the semantic analyzer does the following:
@@ -11894,7 +12835,7 @@To type a binary expression \$e_1\$ op
\$e_2\$ where op is +, -, *, or /, the
@@ -11939,7 +12880,7 @@
To type a binary expression \$e_1\$ op
\$e_2\$, where op is << or >>,
@@ -11987,7 +12928,7 @@
We say that types \$T_1\$ and \$T_2\$ are identical if one of the following holds:
@@ -12053,7 +12994,7 @@We say that a type \$T_1\$ may be converted to another type \$T_2\$ if every value represented by type \$T_1\$ can be @@ -12156,9 +13097,9 @@
Here are the rules for resolving two types \$T_1\$ and \$T_2\$ (e.g., the @@ -12284,7 +13225,7 @@
To compute a common type for a list of types \$T_1, ... , T_n\$, do the following:
@@ -12311,7 +13252,7 @@A type option is one of the following:
@@ -12332,7 +13273,7 @@We say that a type option \$O_1\$ may be converted to another type option \$O_2\$ in the following cases.
@@ -12382,9 +13323,9 @@Here are the rules for resolving two type options \$O_1\$ and \$O_2\$ to a common type option \$O\$:
@@ -12442,7 +13383,7 @@To compute a common type for a list of type options \$O_1, ... , O_n\$, do the following:
@@ -12470,7 +13411,7 @@A value is one of the following:
@@ -12531,7 +13472,7 @@A primitive integer value is an ordinary (mathematical) integer value together with its canonical type, which is a @@ -12562,7 +13503,7 @@
An integer value is an ordinary (mathematical) integer value. Its canonical type is Integer. @@ -12571,7 +13512,7 @@
A floating-point value is an IEEE floating-point value of 4- or 8-byte width. Formally, the set of floating-point values is the disjoint union @@ -12594,14 +13535,14 @@
A Boolean value is one of the values true and false.
Its canonical type is bool.
A string value is a sequence of characters that can be represented as a string literal expression. @@ -12624,7 +13565,7 @@
An abstract type value is a value associated with an abstract type. @@ -12637,7 +13578,7 @@
An anonymous array value is a value associated with an anonymous array type. @@ -12669,7 +13610,7 @@
An array value is a value associated with an array type.
We write an array value like an anonymous array
@@ -12707,7 +13648,7 @@ 20.8. Array Values
An enumeration value is a value associated with an enumerated constant definition. @@ -12719,7 +13660,7 @@
An anonymous struct value is a value associated with an
anonymous struct
@@ -12739,7 +13680,7 @@ 20.10. Anonymous Struct Values
A struct value is a value associated with a struct type. @@ -12776,7 +13717,7 @@
Every value v of a displayable type has a serialized size. This is the number of bytes required to represent v @@ -12819,7 +13760,7 @@
Evaluation is the process of transforming an expression into @@ -12829,7 +13770,7 @@
Evaluation of expressions occurs as stated in the expression descriptions. Evaluation of integer @@ -12840,14 +13781,14 @@
The following rules govern the conversion of a value \$v_1\$ of canonical type \$T_1\$ to a value \$v_2\$ of canonical type \$T_2\$.
If \$T_1\$ and \$T_2\$ are primitive integer types with one signed and one unsigned, @@ -12914,7 +13855,7 @@
If \$T_1\$ is Integer and \$T_2\$ is a primitive integer type, then proceed as if \$T_1\$ were a signed primitive integer @@ -12929,7 +13870,7 @@
We use the standard rules for IEEE floating-point values to convert among integer values to and from floating-point values and @@ -12937,7 +13878,7 @@
If \$T_1\$ is a string type and \$T_2\$ is a string type, then let \$v_2 = v_1\$.
@@ -12951,7 +13892,7 @@If \$T_2\$ is an array type and \$T_1 = T_2\$, then let \$v_2 = v_1\$.
@@ -13000,7 +13941,7 @@If \$T_2\$ is a struct type and \$T_1 = T_2\$, then let \$v_2 = v_1\$.
@@ -13034,13 +13975,13 @@This section specifies the semantics of FPP ports.
A port instance I is an instantiated port instance specifier. @@ -13071,7 +14012,7 @@
A port interface is a set I of port instances. @@ -13091,10 +14032,13 @@
For every port interface definition D, there is an associated port interface I. @@ -13115,7 +14059,7 @@
For every component definition D, there is an associated port interface I. @@ -13136,7 +14080,7 @@
For every component instance definition D, there is an associated port interface I. @@ -13146,7 +14090,7 @@
For every topology definition D, there is an associated port interface I. This port interface is computed as follows. @@ -13169,24 +14113,75 @@
For every port interface instance template +parameter P, there +is an associated port interface I. +It is given by the port interface definition referred to in P.
+A port interface instance is either a -component instance definition -or a topology definition. -Each of these elements provides concrete port instances -that can serve as the endpoints of topology connections. -A port interface instance has an associated port interface I; -this is the port interface associated with the -component instance -definition or -topology definition.
+A port interface instance is one of the following:
+A port interface +instance parameter definition \$(p, A)\$, where +p is a port interface instance template +parameter, and A refers to +a port interface instance.
+Each of these elements has an associated port interface I:
+For a component instance definition D, the port interface is given by the +component definition referred to in D.
+For a topology definition, the port interface is given by the definition.
+For a port interface instance template definition \$(p, A)\$, the port +interface is specified in the parameter p.
+Each port interface instance represents a component instance definition +or topology definition D:
+A component instance definition D or topology definition D represents +D.
+A bound parameter definition B represents a definition D as +described in the section on definitions.
+A port interface I is a sub-interface of a port interface I' if the set of port instances represented by I is a subset of the set of @@ -13201,10 +14196,10 @@
Analysis is the process of checking a source model. It usually involves the following steps:
@@ -13215,13 +14210,17 @@Lexing and parsing to generate an abstract syntax tree (AST).
If step (1) succeeded, semantic analysis of the AST.
+If step (1), succeeded, analysis and expansion of +templates to generate a new AST.
+If step (2) succeeded, semantic analysis of the AST generated in step (2).
An analysis tool is a tool that reads in and analyzes FPP source files, but does not generate any code. @@ -13245,7 +14244,7 @@
Translation is the process of performing analysis and generating code.
@@ -13270,7 +14269,7 @@A translation tool is a tool that translates FPP source files. A translation tool typically accepts the following two kinds of @@ -13309,7 +14308,7 @@