drummyfish/less_retarded_wiki
5
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Update

Browse source
This commit is contained in:
Miloslav Ciz 2023-12-22 00:43:13 +01:00
parent b931aa36c3
commit b6b5090c4c
5 changed files with 183 additions and 26 deletions
Download patch file Download diff file Expand all files Collapse all files

155
bytecode.md
View file

@ -4,7 +4,7 @@ TODO
## Example
Let's consider a simple algorithm that tests the [Collatz conjecture](collatz_conjecture.md) (which says that applying a simple operation from any starting number over and over will always lead to number 1). The algorithm in [C](c.md) would look as follows:
Let's consider a simple algorithm that tests the [Collatz conjecture](collatz_conjecture.md) (which says that applying a simple operation from any starting number over and over will always lead to number 1). The program reads a number (one digit for simplicity) and then prints the sequence until reaching the final number 1. The algorithm in [C](c.md) would look as follows:
```
// Collatz conjecture
@ -35,7 +35,88 @@ int main(void)
}
```
The program reads a number (one digit for simplicity) and then prints the sequence until reaching the final number 1. Now let's rewrite the same algorithm in [comun](comun.md), a language which will allow us to produce bytecode:
C will be normally compiled to [machine code](machine_code.md), however we can take a look at some immediate representation bytecode that compilers internally use to generate the machine code. The following is [LLVM](llvm.md), a widely used bytecode that can be produced from the above C code with [clang](clang.md) compiler (e.g. as `clang -cc1 tmp.c -S -emit-llvm -o -`):
```
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-pc-linux-gnu"
@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1
; Function Attrs: noinline nounwind optnone
define i32 @next(i32 %n) #0 {
entry:
%n.addr = alloca i32, align 4
store i32 %n, i32* %n.addr, align 4
%0 = load i32, i32* %n.addr, align 4
%rem = srem i32 %0, 2
%tobool = icmp ne i32 %rem, 0
br i1 %tobool, label %cond.true, label %cond.false
cond.true: ; preds = %entry
%1 = load i32, i32* %n.addr, align 4
%mul = mul nsw i32 3, %1
%add = add nsw i32 %mul, 1
br label %cond.end
cond.false: ; preds = %entry
%2 = load i32, i32* %n.addr, align 4
%div = sdiv i32 %2, 2
br label %cond.end
cond.end: ; preds = %cond.false, %cond.true
%cond = phi i32 [ %add, %cond.true ], [ %div, %cond.false ]
ret i32 %cond
}
; Function Attrs: noinline nounwind optnone
define i32 @main() #0 {
entry:
%retval = alloca i32, align 4
%n = alloca i32, align 4
store i32 0, i32* %retval, align 4
%call = call i32 (...) @getchar()
%sub = sub nsw i32 %call, 48
store i32 %sub, i32* %n, align 4
br label %while.body
while.body: ; preds = %entry, %if.end
%0 = load i32, i32* %n, align 4
%call1 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i32 0, i32 0), i32 %0)
%1 = load i32, i32* %n, align 4
%cmp = icmp eq i32 %1, 1
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %while.body
br label %while.end
if.end: ; preds = %while.body
%2 = load i32, i32* %n, align 4
%call2 = call i32 @next(i32 %2)
store i32 %call2, i32* %n, align 4
br label %while.body
while.end: ; preds = %if.then
ret i32 0
}
declare i32 @getchar(...) #1
declare i32 @printf(i8*, ...) #1
attributes #0 = { noinline nounwind optnone "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-features"="+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-features"="+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
!llvm.module.flags = !{!0}
!llvm.ident = !{!1}
!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{!"clang version 7.0.1-8+deb10u2 (tags/RELEASE_701/final)"}
```
TODO: analyze the above
Now let's rewrite the same algorithm in [comun](comun.md), a different language which will allow us to produce another kind of bytecode (obtained with `comun -T program.cmn`):
```
# Collatz conjecture
@ -65,7 +146,7 @@ next:
.
```
Here is annotated bytecode this compiles to:
Here is annotated comun bytecode this compiles to:
```
000000: DES 00 0111 # func \ next:
@ -127,4 +208,70 @@ Here is annotated bytecode this compiles to:
000038: END 00 0000
```
TODO: analyze the above, show other bytecodes (python, java, ...)
TODO: analyze the above, show other bytecodes (python, java, ...)
Let's try the same in [Python](python.md). The code we'll examine will look like this:
```
# Collatz conjecture
def next(n):
return 3 * n + 1 if n % 2 != 0 else n / 2
n = ord(raw_input()[0]) - ord('0')
while True:
print(n)
if n == 1:
break
n = next(n)
```
And the bytecode we get (e.g. with `python -m dis program.py`):
```
3 0 LOAD_CONST 0 (<code object next at ...)
3 MAKE_FUNCTION 0
6 STORE_NAME 0 (next)
6 9 LOAD_NAME 1 (ord)
12 LOAD_NAME 2 (raw_input)
15 CALL_FUNCTION 0
18 LOAD_CONST 1 (0)
21 BINARY_SUBSCR
22 CALL_FUNCTION 1
25 LOAD_NAME 1 (ord)
28 LOAD_CONST 2 ('0')
31 CALL_FUNCTION 1
34 BINARY_SUBTRACT
35 STORE_NAME 3 (n)
8 38 SETUP_LOOP 43 (to 84)
>> 41 LOAD_NAME 4 (True)
44 POP_JUMP_IF_FALSE 83
9 47 LOAD_NAME 3 (n)
50 PRINT_ITEM
51 PRINT_NEWLINE
11 52 LOAD_NAME 3 (n)
55 LOAD_CONST 3 (1)
58 COMPARE_OP 2 (==)
61 POP_JUMP_IF_FALSE 68
12 64 BREAK_LOOP
65 JUMP_FORWARD 0 (to 68)
14 >> 68 LOAD_NAME 0 (next)
71 LOAD_NAME 3 (n)
74 CALL_FUNCTION 1
77 STORE_NAME 3 (n)
80 JUMP_ABSOLUTE 41
>> 83 POP_BLOCK
>> 84 LOAD_CONST 4 (None)
87 RETURN_VALUE
```
TODO: make sense of it and analyze it