[meta] Port Formats and Operands to the Rust crate;

cranelift/codegen/meta/src/cdsl/formats.rs

@@ -0,0 +1,246 @@
+use crate::cdsl::operands::{Operand, OperandKind};
+
+use std::collections::{HashMap, HashSet};
+use std::fmt;
+use std::slice;
+
+use cranelift_entity::{entity_impl, PrimaryMap};
+
+/// An immediate field in an instruction format.
+///
+/// This corresponds to a single member of a variant of the `InstructionData`
+/// data type.
+///
+/// :param iform: Parent `InstructionFormat`.
+/// :param kind: Immediate Operand kind.
+/// :param member: Member name in `InstructionData` variant.
+#[derive(Debug)]
+pub struct FormatField {
+    /// Immediate operand number in parent.
+    immnum: usize,
+
+    /// Immediate operand kind.
+    pub kind: OperandKind,
+
+    /// Member name in InstructionDate variant.
+    pub member: &'static str,
+}
+
+/// Every instruction opcode has a corresponding instruction format which
+/// determines the number of operands and their kinds. Instruction formats are
+/// identified structurally, i.e., the format of an instruction is derived from
+/// the kinds of operands used in its declaration.
+///
+/// The instruction format stores two separate lists of operands: Immediates
+/// and values. Immediate operands (including entity references) are
+/// represented as explicit members in the `InstructionData` variants. The
+/// value operands are stored differently, depending on how many there are.
+/// Beyond a certain point, instruction formats switch to an external value
+/// list for storing value arguments. Value lists can hold an arbitrary number
+/// of values.
+///
+/// All instruction formats must be predefined in the meta shared/formats module.
+///
+/// :param kinds: List of `OperandKind` objects describing the operands.
+/// :param name: Instruction format name in CamelCase. This is used as a Rust
+///     variant name in both the `InstructionData` and `InstructionFormat`
+///     enums.
+/// :param typevar_operand: Index of the value input operand that is used to
+///     infer the controlling type variable. By default, this is `0`, the first
+///     `value` operand. The index is relative to the values only, ignoring
+///     immediate operands.
+#[derive(Debug)]
+pub struct InstructionFormat {
+    pub name: &'static str,
+    pub num_value_operands: usize,
+    pub has_value_list: bool,
+    pub imm_fields: Vec<FormatField>,
+    pub typevar_operand: Option<usize>,
+}
+
+impl fmt::Display for InstructionFormat {
+    fn fmt(&self, fmt: &mut fmt::Formatter) -> Result<(), fmt::Error> {
+        let args = self
+            .imm_fields
+            .iter()
+            .map(|field| format!("{}: {}", field.member, field.kind.name))
+            .collect::<Vec<_>>()
+            .join(", ");
+        fmt.write_fmt(format_args!(
+            "{}(imms=({}), vals={})",
+            self.name, args, self.num_value_operands
+        ))?;
+        Ok(())
+    }
+}
+
+pub struct InstructionFormatBuilder {
+    name: &'static str,
+    num_value_operands: usize,
+    has_value_list: bool,
+    imm_fields: Vec<FormatField>,
+    typevar_operand: Option<usize>,
+}
+
+pub struct ImmParameter {
+    kind: OperandKind,
+    member: &'static str,
+}
+impl Into<ImmParameter> for (&'static str, &OperandKind) {
+    fn into(self) -> ImmParameter {
+        ImmParameter {
+            kind: self.1.clone(),
+            member: self.0,
+        }
+    }
+}
+impl Into<ImmParameter> for &OperandKind {
+    fn into(self) -> ImmParameter {
+        ImmParameter {
+            kind: self.clone(),
+            member: self.default_member.unwrap(),
+        }
+    }
+}
+
+impl InstructionFormatBuilder {
+    pub fn new(name: &'static str) -> Self {
+        Self {
+            name,
+            num_value_operands: 0,
+            has_value_list: false,
+            imm_fields: Vec::new(),
+            typevar_operand: None,
+        }
+    }
+
+    pub fn value(mut self) -> Self {
+        self.num_value_operands += 1;
+        self
+    }
+
+    pub fn varargs(mut self) -> Self {
+        self.has_value_list = true;
+        self
+    }
+
+    pub fn imm(mut self, param: impl Into<ImmParameter>) -> Self {
+        let imm_param = param.into();
+        let field = FormatField {
+            immnum: self.imm_fields.len(),
+            kind: imm_param.kind,
+            member: imm_param.member,
+        };
+        self.imm_fields.push(field);
+        self
+    }
+
+    pub fn typevar_operand(mut self, operand_index: usize) -> Self {
+        assert!(self.typevar_operand.is_none());
+        assert!(self.has_value_list || operand_index < self.num_value_operands);
+        self.typevar_operand = Some(operand_index);
+        self
+    }
+
+    pub fn finish(self) -> InstructionFormat {
+        let typevar_operand = if self.typevar_operand.is_some() {
+            self.typevar_operand
+        } else if self.has_value_list || self.num_value_operands > 0 {
+            // Default to the first value operand, if there's one.
+            Some(0)
+        } else {
+            None
+        };
+
+        InstructionFormat {
+            name: self.name,
+            num_value_operands: self.num_value_operands,
+            has_value_list: self.has_value_list,
+            imm_fields: self.imm_fields,
+            typevar_operand,
+        }
+    }
+}
+
+#[derive(Copy, Clone, PartialEq, Eq, Hash, PartialOrd, Ord)]
+pub struct InstructionFormatIndex(u32);
+entity_impl!(InstructionFormatIndex);
+
+pub struct FormatRegistry {
+    /// Map (immediate kinds names, number of values, has varargs) to an instruction format index
+    /// in the actual map.
+    sig_to_index: HashMap<(Vec<String>, usize, bool), InstructionFormatIndex>,
+    map: PrimaryMap<InstructionFormatIndex, InstructionFormat>,
+    name_set: HashSet<&'static str>,
+}
+
+impl FormatRegistry {
+    pub fn new() -> Self {
+        Self {
+            sig_to_index: HashMap::new(),
+            map: PrimaryMap::new(),
+            name_set: HashSet::new(),
+        }
+    }
+
+    /// Find an existing instruction format that matches the given lists of instruction inputs and
+    /// outputs.
+    pub fn lookup(&self, operands_in: &Vec<Operand>) -> InstructionFormatIndex {
+        let mut imm_keys = Vec::new();
+        let mut num_values = 0;
+        let mut has_varargs = false;
+
+        for operand in operands_in.iter() {
+            if operand.is_value() {
+                num_values += 1;
+            }
+            has_varargs = has_varargs || operand.is_varargs();
+            if let Some(imm_key) = operand.kind.imm_key() {
+                imm_keys.push(imm_key);
+            }
+        }
+
+        let sig = (imm_keys, num_values, has_varargs);
+        *self
+            .sig_to_index
+            .get(&sig)
+            .expect("unknown InstructionFormat; please define it in shared/formats.rs first")
+    }
+
+    pub fn get(&self, index: InstructionFormatIndex) -> &InstructionFormat {
+        self.map.get(index).unwrap()
+    }
+
+    pub fn insert(&mut self, inst_format: InstructionFormatBuilder) {
+        let name = &inst_format.name;
+        if !self.name_set.insert(name) {
+            panic!(
+                "Trying to add an InstructionFormat named {}, but it already exists!",
+                name
+            );
+        }
+
+        let format = inst_format.finish();
+
+        // Compute key.
+        let imm_keys = format
+            .imm_fields
+            .iter()
+            .map(|field| field.kind.imm_key().unwrap())
+            .collect();
+        let key = (imm_keys, format.num_value_operands, format.has_value_list);
+
+        let index = self.map.push(format);
+        if let Some(already_inserted) = self.sig_to_index.insert(key, index) {
+            panic!(
+                "duplicate InstructionFormat: trying to insert '{}' while '{}' already has the same structure.",
+                self.map.get(index).unwrap().name,
+                self.map.get(already_inserted).unwrap().name
+            );
+        }
+    }
+
+    pub fn iter(&self) -> slice::Iter<InstructionFormat> {
+        self.map.values()
+    }
+}

cranelift/codegen/meta/src/cdsl/mod.rs

@@ -3,7 +3,9 @@
 //! This module defines the classes that are used to define Cranelift
 //! instructions and other entities.
 
+pub mod formats;
 pub mod isa;
+pub mod operands;
 pub mod regs;
 pub mod settings;
 pub mod types;

cranelift/codegen/meta/src/cdsl/operands.rs

@@ -0,0 +1,285 @@
+use std::collections::HashMap;
+
+use crate::cdsl::camel_case;
+use crate::cdsl::typevar::TypeVar;
+
+/// An instruction operand can be an *immediate*, an *SSA value*, or an *entity reference*. The
+/// type of the operand is one of:
+///
+/// 1. A `ValueType` instance indicates an SSA value operand with a concrete type.
+///
+/// 2. A `TypeVar` instance indicates an SSA value operand, and the instruction is polymorphic over
+///    the possible concrete types that the type variable can assume.
+///
+/// 3. An `ImmediateKind` instance indicates an immediate operand whose value is encoded in the
+///    instruction itself rather than being passed as an SSA value.
+///
+/// 4. An `EntityRefKind` instance indicates an operand that references another entity in the
+///    function, typically something declared in the function preamble.
+#[derive(Clone, Debug)]
+pub struct Operand {
+    pub name: &'static str,
+    pub doc: Option<String>,
+    pub kind: OperandKind,
+}
+
+impl Operand {
+    pub fn is_value(&self) -> bool {
+        match self.kind.fields {
+            OperandKindFields::TypeVar(_) => true,
+            _ => false,
+        }
+    }
+
+    pub fn type_var(&self) -> Option<&TypeVar> {
+        match &self.kind.fields {
+            OperandKindFields::TypeVar(typevar) => Some(typevar),
+            _ => None,
+        }
+    }
+
+    pub fn is_varargs(&self) -> bool {
+        match self.kind.fields {
+            OperandKindFields::VariableArgs => true,
+            _ => false,
+        }
+    }
+
+    /// Returns true if the operand has an immediate kind or is an EntityRef.
+    // TODO inherited name from the python, rename to is_immediate_or_entityref later.
+    pub fn is_immediate(&self) -> bool {
+        match self.kind.fields {
+            OperandKindFields::ImmEnum(_)
+            | OperandKindFields::ImmValue
+            | OperandKindFields::EntityRef => true,
+            _ => false,
+        }
+    }
+
+    /// Returns true if the operand has an immediate kind.
+    pub fn is_pure_immediate(&self) -> bool {
+        match self.kind.fields {
+            OperandKindFields::ImmEnum(_) | OperandKindFields::ImmValue => true,
+            _ => false,
+        }
+    }
+
+    pub fn is_cpu_flags(&self) -> bool {
+        match &self.kind.fields {
+            OperandKindFields::TypeVar(type_var)
+                if type_var.name == "iflags" || type_var.name == "fflags" =>
+            {
+                true
+            }
+            _ => false,
+        }
+    }
+}
+
+pub struct OperandBuilder {
+    name: &'static str,
+    doc: Option<String>,
+    kind: OperandKind,
+}
+
+impl OperandBuilder {
+    pub fn new(name: &'static str, kind: OperandKind) -> Self {
+        Self {
+            name,
+            doc: None,
+            kind,
+        }
+    }
+    pub fn doc(mut self, doc: impl Into<String>) -> Self {
+        assert!(self.doc.is_none());
+        self.doc = Some(doc.into());
+        self
+    }
+    pub fn finish(self) -> Operand {
+        let doc = match self.doc {
+            Some(doc) => Some(doc),
+            None => match &self.kind.fields {
+                OperandKindFields::TypeVar(tvar) => Some(tvar.doc.clone()),
+                _ => self.kind.doc.clone(),
+            },
+        };
+
+        Operand {
+            name: self.name,
+            doc,
+            kind: self.kind,
+        }
+    }
+}
+
+type EnumValues = HashMap<&'static str, &'static str>;
+
+#[derive(Clone, Debug)]
+pub enum OperandKindFields {
+    EntityRef,
+    VariableArgs,
+    ImmValue,
+    ImmEnum(EnumValues),
+    TypeVar(TypeVar),
+}
+
+#[derive(Clone, Debug)]
+pub struct OperandKind {
+    pub name: &'static str,
+
+    doc: Option<String>,
+
+    pub default_member: Option<&'static str>,
+
+    /// The camel-cased name of an operand kind is also the Rust type used to represent it.
+    pub rust_type: String,
+
+    fields: OperandKindFields,
+}
+
+impl OperandKind {
+    pub fn imm_key(&self) -> Option<String> {
+        match self.fields {
+            OperandKindFields::ImmEnum(_)
+            | OperandKindFields::ImmValue
+            | OperandKindFields::EntityRef => Some(self.name.to_string()),
+            _ => None,
+        }
+    }
+
+    pub fn type_var(&self) -> TypeVar {
+        match &self.fields {
+            OperandKindFields::TypeVar(tvar) => tvar.clone(),
+            _ => panic!("not a typevar"),
+        }
+    }
+}
+
+pub struct OperandKindBuilder {
+    name: &'static str,
+
+    doc: Option<String>,
+
+    default_member: Option<&'static str>,
+
+    /// The camel-cased name of an operand kind is also the Rust type used to represent it.
+    rust_type: Option<String>,
+
+    fields: OperandKindFields,
+}
+
+impl OperandKindBuilder {
+    pub fn new(name: &'static str, fields: OperandKindFields) -> Self {
+        Self {
+            name,
+            doc: None,
+            default_member: None,
+            rust_type: None,
+            fields,
+        }
+    }
+
+    pub fn new_imm(name: &'static str) -> Self {
+        Self {
+            name,
+            doc: None,
+            default_member: None,
+            rust_type: None,
+            fields: OperandKindFields::ImmValue,
+        }
+    }
+
+    pub fn new_enum(name: &'static str, values: EnumValues) -> Self {
+        Self {
+            name,
+            doc: None,
+            default_member: None,
+            rust_type: None,
+            fields: OperandKindFields::ImmEnum(values),
+        }
+    }
+
+    pub fn doc(mut self, doc: &'static str) -> Self {
+        assert!(self.doc.is_none());
+        self.doc = Some(doc.to_string());
+        self
+    }
+    pub fn default_member(mut self, default_member: &'static str) -> Self {
+        assert!(self.default_member.is_none());
+        self.default_member = Some(default_member);
+        self
+    }
+    pub fn rust_type(mut self, rust_type: &'static str) -> Self {
+        assert!(self.rust_type.is_none());
+        self.rust_type = Some(rust_type.to_string());
+        self
+    }
+
+    pub fn finish(self) -> OperandKind {
+        let default_member = match self.default_member {
+            Some(default_member) => Some(default_member),
+            None => match &self.fields {
+                OperandKindFields::ImmEnum(_) | OperandKindFields::ImmValue => Some("imm"),
+                OperandKindFields::TypeVar(_) | OperandKindFields::EntityRef => Some(self.name),
+                OperandKindFields::VariableArgs => None,
+            },
+        };
+
+        let rust_type = match self.rust_type {
+            Some(rust_type) => rust_type.to_string(),
+            None => match &self.fields {
+                OperandKindFields::ImmEnum(_) | OperandKindFields::ImmValue => {
+                    format!("ir::immediates::{}", camel_case(self.name))
+                }
+                OperandKindFields::VariableArgs => "&[Value]".to_string(),
+                OperandKindFields::TypeVar(_) | OperandKindFields::EntityRef => {
+                    format!("ir::{}", camel_case(self.name))
+                }
+            },
+        };
+
+        let doc = match self.doc {
+            Some(doc) => Some(doc),
+            None => match &self.fields {
+                OperandKindFields::TypeVar(type_var) => Some(type_var.doc.clone()),
+                OperandKindFields::ImmEnum(_)
+                | OperandKindFields::ImmValue
+                | OperandKindFields::EntityRef
+                | OperandKindFields::VariableArgs => None,
+            },
+        };
+
+        OperandKind {
+            name: self.name,
+            doc,
+            default_member,
+            rust_type,
+            fields: self.fields,
+        }
+    }
+}
+
+impl Into<OperandKind> for &TypeVar {
+    fn into(self) -> OperandKind {
+        OperandKindBuilder::new("value", OperandKindFields::TypeVar(self.into())).finish()
+    }
+}
+impl Into<OperandKind> for &OperandKind {
+    fn into(self) -> OperandKind {
+        self.clone()
+    }
+}
+
+/// Helper to create an operand in definitions files.
+pub fn create_operand(name: &'static str, kind: impl Into<OperandKind>) -> Operand {
+    OperandBuilder::new(name, kind.into()).finish()
+}
+
+/// Helper to create an operand with a documentation in definitions files.
+pub fn create_operand_doc(
+    name: &'static str,
+    kind: impl Into<OperandKind>,
+    doc: &'static str,
+) -> Operand {
+    OperandBuilder::new(name, kind.into()).doc(doc).finish()
+}