From 6bdc893ae88a8505cfcd08c619e419e724468c59 Mon Sep 17 00:00:00 2001 From: Cliff Click Date: Tue, 6 Aug 2024 13:09:38 +0100 Subject: Chapter 14: Narrow types Co-authored-by: Thalia Archibald Co-authored-by: Dibyendu Majumdar Co-authored-by: XmiliaH <45106915+XmiliaH@users.noreply.github.com> Co-authored-by: Robert Obkircher --- docs/chapter14/14-grammar.md | 124 ++++++++ docs/chapter14/Makefile | 12 + docs/chapter14/README.md | 100 +++++++ docs/chapter14/lattice.gv | 202 +++++++++++++ docs/chapter14/lattice.svg | 691 +++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 1129 insertions(+) create mode 100644 docs/chapter14/14-grammar.md create mode 100644 docs/chapter14/Makefile create mode 100644 docs/chapter14/README.md create mode 100644 docs/chapter14/lattice.gv create mode 100644 docs/chapter14/lattice.svg (limited to 'docs/chapter14') diff --git a/docs/chapter14/14-grammar.md b/docs/chapter14/14-grammar.md new file mode 100644 index 00000000..a6d4df42 --- /dev/null +++ b/docs/chapter14/14-grammar.md @@ -0,0 +1,124 @@ +# Grammar for Chapter 14 + +```antlrv4 +grammar SimpleLanguage; + +program : block EOF ; + +block : statement+ ; + + +statement + : '{' block '}' + | returnStatement + | ifStatement + | whileStatement + | breakStatement + | continueStatement + | metaStatement + | structDeclaration + | expressionStatement + ; + + +returnStatement : 'return' expression ';' ; + +ifStatement : 'if' '(' expression ')' statement ('else' statement)? ; + +whileStatement : 'while' '(' expression ')' statement ; + +breakStatement : 'break' ';' ; + +continueStatement : 'continue' ';' ; + +metaStatement : '#showGraph' ';' ; + +structDeclaration : 'struct' IDENTIFIER '{' fields '}' ; + +fields : field+ ; + +field : type IDENTIFIER ';' ; + + +expressionStatement + : type IDENTIFIER ';' + | type IDENTIFIER '=' expression ';' + | IDENTIFIER '=' expression ';' + | expression + ; + +PRIMTYPE + : 'int' + | 'i8' + | 'i16' + | 'i32' + | 'i64' + | 'u8' + | 'u16' + | 'u32' + | 'flt' + | 'f32' + | 'f64' + | 'bool' + ; + +type + : PRIMTYPE + | typeName ('?')? + ; + +typeName : IDENTIFIER ; + + +expression : bitWiseExpression ; + +bitWiseExpression + : comparisonExpression ( '&' | '|' | '^' ) comparisonExpression)* + ; + +comparisonExpression + : shiftExpression (('==' | '!='| '>'| '<'| '>='| '<=') shiftExpression)* + ; + +shiftExpression + : additiveExpression ( '<<' | '>>>' | '>>' ) additiveExpression)* + ; + +additiveExpression + : multiplicativeExpression (('+' | '-') multiplicativeExpression)* + ; + +multiplicativeExpression + : unaryExpression (('*' | '/') unaryExpression)* + ; + +unaryExpression + : ('-') unaryExpression + | '!' unaryExpression + | primaryExpression postFix* + ; + +primaryExpression + : INTEGER_LITERAL + | '(' expression ')' + | 'true' + | 'false' + | 'null' + | newExpression + | IDENTIFIER + ; + +newExpression : 'new' IDENTIFIER ; + +postFix : '.' IDENTIFIER [ '=' expression ] ; + +INTEGER_LITERAL + : [1-9]DIGIT* + | [0] + ; + +IDENTIFIER : NON_DIGIT (NON_DIGIT | DIGIT)* ; + +NON_DIGIT: [a-zA-Z_]; +DIGIT: [0-9]; +``` diff --git a/docs/chapter14/Makefile b/docs/chapter14/Makefile new file mode 100644 index 00000000..12659b8f --- /dev/null +++ b/docs/chapter14/Makefile @@ -0,0 +1,12 @@ +GV := $(wildcard *.gv) +SVG := $(patsubst %.gv,%.svg,$(GV)) + +.PHONY: all +all: $(SVG) + +%.svg: %.gv + dot -Tsvg $< > $@ + +.PHONY: clean +clean: + @rm -f $(SVG) diff --git a/docs/chapter14/README.md b/docs/chapter14/README.md new file mode 100644 index 00000000..21f8fbd6 --- /dev/null +++ b/docs/chapter14/README.md @@ -0,0 +1,100 @@ +# Chapter 14: Narrow types + +In this chapter, we add sub-word integer types. + +You can also read [this chapter](https://github.com/SeaOfNodes/Simple/tree/linear-chapter14) in a linear Git revision history on the [linear](https://github.com/SeaOfNodes/Simple/tree/linear) branch and [compare](https://github.com/SeaOfNodes/Simple/compare/linear-chapter13...linear-chapter14) it to the previous chapter. + +## Narrow or sub-word types + +As a prelude to arrays, especially arrays of bytes common in all networking +codes, Simple needs some way to manipulate sub-word integer types. +Computations are always done in 64 bits for both integer and float ops, but +will be truncated (rounded) on a variable or field assignment, and (sign) +extended on a load. + +```java +i8 x = 123456789; +if( x != 21 ) + return false; +return true; +``` + +The variable `x` is limited to 8 bits signed. Assigning the vary large value +`123456789` (hex `0x75BCD15`) simply truncates to 8 bits yielding `21 (0x15)`. +The value is sign-extended on a load. + +```java +i16 x = 123456789; +if( x != -13035 ) + return false; +return true; +``` + +The variable `x` is limited to 16 bits signed. Assigning the vary large value +`0x75BCD15` simply truncates to 16 bits yielding `0xCD15`. The value is +sign-extended on a load yielding `-13035 (0xFFFF FFFF FFFF CD15)`. + +```java +u16 x = 123456789; +if( x != 52501 ) + return false; +return true; +``` + +The variable `x` is limited to 16 bits unsigned. Assigning the vary large value +`0x75BCD15` simply truncates to 16 bits yielding `0xCD15`. The value is +zero-extended on a load yielding `52501 (0x0000 0000 0000 CD15)`. + +Similarly, floats can be limited to IEEE754 32-bit standard format and will be +rounded when assigned, and inflated to 64-bits when loaded. + +The complete list of primitive types is now: +```java +int, i64 // 64-bits, matching modern hardware +// u64 // NOT implemented, as it hints at some kind of >64 bit behavior +i32, u32 // 32 bits, sign or zero extended on load +i16, u16 // 16 bits, sign or zero extended on load + i8, u8 // 8 bits, sign or zero extended on load + i1, u1 // 1 bit , sign or zero extended on load +bool // Same as u1 +flt, f64 // IEEE 754 64-bit float, sometimes called a "double" +f32 // IEEE 754 32-bit float +``` + + + +## Type Implementation + +The TypeInteger type class is +reworked to support a full range of min/max values. At this time, only some +power-of-2 sized ranges are exposed to the programmer but the optimizer +internally supports all ranges. The MEET operation takes the min-of-mins and +max-of-maxes. The DUAL operation inverts the min and max. + +Example: the constant `0` has `[0...0]`. + +Example: the `bool` type has the range `[0...1]`. + +Example: the `u8` type has the range `[0...255]`. + +Example: the `i8` type has the range `[-128...127]`. + +Example: the `dual` of `bool` is `[1...0]` (just swap min and max). + +The TypeFloat class is also reworked to support 32-bit and 64-bit sizes. + +![Graph1](./lattice.svg) + + + + +## Nodes + +Added this Chapter are some bit-manipulation Nodes representing common hardware +ops to mask, shift, sign-extend and round bits. The truncation and extension +logic uses these. They are otherwise very similar to the previous +`Add/Sub/Mul/Div` Nodes but with slightly different constant math and +slightly different `idealize()` calls. + + + diff --git a/docs/chapter14/lattice.gv b/docs/chapter14/lattice.gv new file mode 100644 index 00000000..4a8b1d4c --- /dev/null +++ b/docs/chapter14/lattice.gv @@ -0,0 +1,202 @@ +digraph Lattice { + compound = true; + // Define nodes + splines=false; + node [shape=circle, style=filled]; + nodesep=0.6; + + TOP [label=<⊤>, fillcolor=black, fontcolor=white]; + BOT [label=<⊥>, fillcolor=black, fontcolor=white]; + + // ---------------------- + // Control + XCTRL [label=<⊤:ctrl>, fillcolor=yellow, shape=oval]; + CTRL [label=<⊥:ctrl>, fillcolor=yellow, shape=oval]; + + TOP -> XCTRL; + XCTRL -> CTRL; + CTRL -> BOT; + + // ---------------------- + // Integers + INT_TOP [label=<⊤:int>, fillcolor=lightblue, shape=oval]; + INT_BOT [label=<⊥:int>, fillcolor=lightblue, shape=oval]; + + INT_HI [label="max < min" , fillcolor=lightblue, shape=oval]; + INT_XBOOL [label="1...0", fillcolor=lightblue, shape=oval]; + INT_CON [label="constants", fillcolor=lightblue, shape=oval]; + INT_CON0 [label="0", fillcolor=lightblue]; + INT_CON1 [label="1", fillcolor=lightblue]; + INT_BOOL [label="0...1", fillcolor=lightblue,shape=oval]; + INT_LO [label="max > min" , fillcolor=lightblue, shape=oval]; + + TOP -> INT_TOP; + INT_TOP -> INT_HI; + INT_HI -> INT_XBOOL; + INT_XBOOL -> INT_CON0; + INT_XBOOL -> INT_CON1; + INT_HI -> INT_CON; + INT_CON0 -> INT_BOOL; + INT_CON1 -> INT_BOOL; + INT_BOOL -> INT_LO; + INT_CON -> INT_LO; + INT_LO -> INT_BOT; + INT_BOT -> BOT; + + INT_CON0 -> INT_CON1 -> INT_CON [style=invis]; + + // ---------------------- + // Integers + FLT_TOP [label=<⊤:flt>, fillcolor=aqua, shape=oval]; + FLT_BOT [label=<⊥:flt>, fillcolor=aqua, shape=oval]; + + FLT_CON64 [label="64b consts\n[excludes\n32b consts]" , fillcolor=aqua, shape=oval]; + FLT_CON32 [label="32b consts", fillcolor=aqua, shape=oval]; + FLT_CONPI [label="3.14159...", fillcolor=aqua, shape=oval]; + FLT_X32 [label="~f32", fillcolor=aqua]; + FLT_CON0 [label="0." , fillcolor=aqua]; + FLT_32 [label="f32" , fillcolor=aqua]; + + TOP -> FLT_TOP; + FLT_TOP -> FLT_CON64 + FLT_TOP -> FLT_CONPI + FLT_CON64 -> FLT_BOT + FLT_CONPI -> FLT_BOT + + FLT_TOP -> FLT_X32; + FLT_X32 -> FLT_CON0; + FLT_X32 -> FLT_CON32; + FLT_CON0 -> FLT_32; + FLT_CON32 -> FLT_32; + FLT_32 -> FLT_BOT + + FLT_BOT -> BOT; + + FLT_CON0 -> FLT_CON32 -> FLT_CONPI -> FLT_CON64 [style=invis]; + + + // ---------------------- + // Pointers, just a boolean nil/not and a struct + PTR [shape=none, fillcolor=none, label=< + + + +
*ptr
struct
+ >]; + PTR0 [shape=none, fillcolor=none, label=< + + + +
*ptr?
struct
+ >]; + + TOP -> PTR; + PTR -> PTR0; + PTR0 -> BOT; + + // ---------------------- + // Structs + STRUCT_TOP [label=<⊤:struct>, fillcolor=lightgreen, shape=oval]; + STRUCT_BOT [label=<⊥:struct>, fillcolor=lightgreen, shape=oval]; + + STRUCT_S1 [shape=none, fillcolor=none, label=< + + +
fieldint/ptr
+ >]; + STRUCT_S2 [shape=none, fillcolor=none, label=< + + + +
fieldint/ptr
fieldint/ptr
+ >]; + STRUCT_SN [shape=none, fillcolor=none, label=< + + + + +
fieldint/ptr
... int/ptr
fieldint/ptr
+ >]; + + + TOP -> STRUCT_TOP; + STRUCT_TOP -> STRUCT_S1; + STRUCT_TOP -> STRUCT_S2; + STRUCT_TOP -> STRUCT_SN; + STRUCT_S1 -> STRUCT_BOT; + STRUCT_S2 -> STRUCT_BOT; + STRUCT_SN -> STRUCT_BOT; + STRUCT_BOT -> BOT; + + STRUCT_S1 -> STRUCT_S2 -> STRUCT_SN [style=invis]; + + // ---------------------- + // Memory + MEM_TOP [label=<⊤:mem>, fillcolor=blue, fontcolor=white, shape=oval]; + MEM_BOT [label=<⊥:mem>, fillcolor=blue, fontcolor=white, shape=oval]; + MEM_ALIAS1 [label="#1" , fillcolor=blue, fontcolor=white]; + MEM_ALIASDOTS [label="..." , fillcolor=blue, fontcolor=white, shape=oval]; + MEM_ALIASN [label="#n" , fillcolor=blue, fontcolor=white]; + + TOP -> MEM_TOP; + MEM_TOP -> MEM_ALIAS1; + MEM_TOP -> MEM_ALIASDOTS; + MEM_TOP -> MEM_ALIASN; + MEM_ALIAS1 -> MEM_BOT; + MEM_ALIASDOTS -> MEM_BOT; + MEM_ALIASN -> MEM_BOT; + MEM_BOT -> BOT; + + MEM_ALIAS1 -> MEM_ALIASDOTS -> MEM_ALIASN [style=invis]; + + // ---------------------- + // Tuples + + TUPLE_TOP [label=<⊤:tuple>, fillcolor=pink, shape=oval]; + TUPLE_BOT [label=<⊥:tuple>, fillcolor=pink, shape=oval]; + + TUPLE1 [shape=none, fillcolor=none, label=< + + + +
Tuple
type
+ >]; + TUPLE2 [shape=none, fillcolor=none, label=< + + + + +
Tuple
type
type
+ >]; + TUPLEN [shape=none, fillcolor=none, label=< + + + + + +
Tuple
type
...
type
+ >]; + + + TOP -> TUPLE_TOP; + TUPLE_TOP -> TUPLE1; + TUPLE_TOP -> TUPLE2; + TUPLE_TOP -> TUPLEN; + TUPLE1 -> TUPLE_BOT; + TUPLE2 -> TUPLE_BOT; + TUPLEN -> TUPLE_BOT; + TUPLE_BOT -> BOT; + + // ---------------------- + // Additional customization + {rank=same; TOP;} + {rank=same; INT_TOP; FLT_TOP; XCTRL; PTR; STRUCT_TOP; MEM_TOP; TUPLE_TOP} + {rank=same; INT_CON; INT_CON0; INT_CON1; + FLT_CON32; FLT_CON64; FLT_CON0; FLT_CONPI; + STRUCT_S1; STRUCT_S2; STRUCT_SN; + MEM_ALIAS1; MEM_ALIASDOTS; MEM_ALIASN; + TUPLE1; TUPLE2; TUPLEN; + } + {rank=same; INT_BOT; FLT_BOT; CTRL; PTR0; STRUCT_BOT; MEM_BOT; TUPLE_BOT} + {rank=same; BOT;} +} diff --git a/docs/chapter14/lattice.svg b/docs/chapter14/lattice.svg new file mode 100644 index 00000000..99f74ed0 --- /dev/null +++ b/docs/chapter14/lattice.svg @@ -0,0 +1,691 @@ + + + + + + +Lattice + + + +TOP + +⊤ + + + +XCTRL + +⊤ +:ctrl + + + +TOP->XCTRL + + + + + +INT_TOP + +⊤ +:int + + + +TOP->INT_TOP + + + + + +FLT_TOP + +⊤ +:flt + + + +TOP->FLT_TOP + + + + + +PTR + + + +*ptr + + +struct + + + +TOP->PTR + + + + + +STRUCT_TOP + +⊤ +:struct + + + +TOP->STRUCT_TOP + + + + + +MEM_TOP + +⊤ +:mem + + + +TOP->MEM_TOP + + + + + +TUPLE_TOP + +⊤ +:tuple + + + +TOP->TUPLE_TOP + + + + + +BOT + +⊥ + + + +CTRL + +⊥ +:ctrl + + + +XCTRL->CTRL + + + + + +CTRL->BOT + + + + + +INT_HI + +max < min + + + +INT_TOP->INT_HI + + + + + +INT_BOT + +⊥ +:int + + + +INT_BOT->BOT + + + + + +INT_XBOOL + +1...0 + + + +INT_HI->INT_XBOOL + + + + + +INT_CON + +constants + + + +INT_HI->INT_CON + + + + + +INT_CON0 + +0 + + + +INT_XBOOL->INT_CON0 + + + + + +INT_CON1 + +1 + + + +INT_XBOOL->INT_CON1 + + + + + +INT_LO + +max > min + + + +INT_CON->INT_LO + + + + + + +INT_BOOL + +0...1 + + + +INT_CON0->INT_BOOL + + + + + + +INT_CON1->INT_BOOL + + + + + +INT_BOOL->INT_LO + + + + + +INT_LO->INT_BOT + + + + + +FLT_CON64 + +64b consts +[excludes +32b consts] + + + +FLT_TOP->FLT_CON64 + + + + + +FLT_CONPI + +3.14159... + + + +FLT_TOP->FLT_CONPI + + + + + +FLT_X32 + +~f32 + + + +FLT_TOP->FLT_X32 + + + + + +FLT_BOT + +⊥ +:flt + + + +FLT_BOT->BOT + + + + + +FLT_CON64->FLT_BOT + + + + + +FLT_CON32 + +32b consts + + + + +FLT_32 + +f32 + + + +FLT_CON32->FLT_32 + + + + + +FLT_CONPI->FLT_BOT + + + + + + +FLT_X32->FLT_CON32 + + + + + +FLT_CON0 + +0. + + + +FLT_X32->FLT_CON0 + + + + + + +FLT_CON0->FLT_32 + + + + + +FLT_32->FLT_BOT + + + + + +PTR0 + + + +*ptr? + + +struct + + + +PTR->PTR0 + + + + + +PTR0->BOT + + + + + +STRUCT_S1 + + + +field + + + + + + + + +int/ptr + + + +STRUCT_TOP->STRUCT_S1 + + + + + +STRUCT_S2 + + + +field + + + + + + + + +int/ptr + + +field + + + + + + + + +int/ptr + + + +STRUCT_TOP->STRUCT_S2 + + + + + +STRUCT_SN + + + +field + + + + + + + + +int/ptr + + +...   + + + + + + + + +int/ptr + + +field + + + + + + + + +int/ptr + + + +STRUCT_TOP->STRUCT_SN + + + + + +STRUCT_BOT + +⊥ +:struct + + + +STRUCT_BOT->BOT + + + + + +STRUCT_S1->STRUCT_BOT + + + + + + +STRUCT_S2->STRUCT_BOT + + + + + + +STRUCT_SN->STRUCT_BOT + + + + + +MEM_ALIAS1 + +#1 + + + +MEM_TOP->MEM_ALIAS1 + + + + + +MEM_ALIASDOTS + +... + + + +MEM_TOP->MEM_ALIASDOTS + + + + + +MEM_ALIASN + +#n + + + +MEM_TOP->MEM_ALIASN + + + + + +MEM_BOT + +⊥ +:mem + + + +MEM_BOT->BOT + + + + + +MEM_ALIAS1->MEM_BOT + + + + + + +MEM_ALIASDOTS->MEM_BOT + + + + + + +MEM_ALIASN->MEM_BOT + + + + + +TUPLE1 + + + +Tuple + + +type + + + +TUPLE_TOP->TUPLE1 + + + + + +TUPLE2 + + + +Tuple + + +type + + +type + + + +TUPLE_TOP->TUPLE2 + + + + + +TUPLEN + + + +Tuple + + +type + + +... + + +type + + + +TUPLE_TOP->TUPLEN + + + + + +TUPLE_BOT + +⊥ +:tuple + + + +TUPLE_BOT->BOT + + + + + +TUPLE1->TUPLE_BOT + + + + + +TUPLE2->TUPLE_BOT + + + + + +TUPLEN->TUPLE_BOT + + + + + -- cgit v1.2.3