README update

README.md

@@ -1,5 +1,3 @@
-# OUTDATED
-
 # APEX - Active code Path EXtractor
 
 APEX is a tool that can extract active code path from the C program source code
@@ -12,46 +10,57 @@ In order to run APEX, you need to have llvm installed:
 - https://llvm.org/docs/GettingStarted.html
 - https://releases.llvm.org/3.8.0/docs/CMake.html
 
+If you want to build LLVM from source, run:
+``` shell
+cd src
+./build_llvm.sh
+```
+
+After getting LLVM, APEX needs to be built. Run the following command
+to build APEX:
+``` shell
+make build
+```
 
 ### How to run APEX
 
-Input C source code is located in the [test.c](c-code/test.c). <br>
-Source and target functions defining start and end of the active code path can
-be specified in the [apex.h](apex/apex.h).
+First, you need to compile your C program into LLVM bitcode since APEX cannot
+work directly with the C source code.
 
-After providing input, run `make run`. Check `apex/build` for output.
+``` bash
+clang -c -g -emit-llvm main.c -o main.bc
+```
 
+After obtaining bitcode, we can now run APEX via `apex.py` launcher:
 
-### How it works
+```
+python apex.py main.bc main.c 16
+```
 
-Let's say we have C source file [test.c](c-code/test.c) and we want to
-extract active path between functions `main` and `y` (we specify these inputs
-according to the instructions above).
+`apex.py` requires three arguments. See usage for full description:
 
-After running `make run`, APEX:
-- uses llvm and translates `test.c` into IR (intermediate representation)
-- creates call graph from the IR
-- finds some active code path from source (`main`) to target (`y`)
-- removes functions that are not on the active code path
+``` bash
+usage: apex.py [-h] [--export EXPORT] code file line
 
-To run extracted active code path, type `lli apex/build/bytecode_from_input_basic_opts_and_apex.bc`
-(see [build_and_run.sh](build_and_run.sh) script for more info).
+positional arguments:
+  code             C source code compiled into LLVM bytecode.
+  file             Target file name (NOT FULL PATH).
+  line             Target line number.
 
-Image on the left represents call graph before we run APEX, while image on the
-right shows how the call graph looks after APEX transformed IR. <br>
+optional arguments:
+  -h, --help       show this help message and exit
+  --export EXPORT  true/false for exporting call graphs.
+```
 
-![](img/callgraph_default_opt.dot.svg) ![](img/callgraph_apex.dot.svg)
+APEX produces extracted executable called `extracted` along with the `build`
+directory (containing various files, logs, etc.).
 
 
 ### Current limitations:
 
 Since APEX is under development, there are currently some serious limitations:
 
-```
-- supports only one compilation module
-- handles only void returning functions on the active path
-- does not handle external libraries
-- (probably tons more that I don't know about)
-```
-
-[TODO](TODO.md) | [DOC](DOC.md)
+- It is possible to extract values only from integer variables.
+- APEX can handle small programs (see examples direcotry), but may have problems
+with bigger ones.
+- (Probably tons more that I don't know about.)

examples/domainname/domainname.c

@@ -0,0 +1,44 @@
+#include <err.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <limits.h>
+
+char *__progname;
+void usage(void);
+
+int
+main(int argc, char *argv[])
+{
+	int ch;
+	char domainname[HOST_NAME_MAX+1];
+
+	while ((ch = getopt(argc, argv, "")) != -1)
+		switch (ch) {
+		default:
+			usage();
+		}
+	argc -= optind;
+	argv += optind;
+
+	if (argc > 1)
+		usage();
+
+	if (*argv) {
+		if (setdomainname(*argv, strlen(*argv)))
+			err(1, "setdomainname");
+	} else {
+		if (getdomainname(domainname, sizeof(domainname)))
+			err(1, "getdomainname");
+		(void)printf("%s\n", domainname);
+	}
+	return(0);
+}
+
+
+void usage(void)
+{
+	(void)fprintf(stderr, "usage: %s [name-of-domain]\n", __progname);
+	exit(1);
+}

examples/example_mod1/example_mod1.c → examples/example_mod2/example_mod2.c

@@ -4,7 +4,7 @@ int foo(int n) {
 }
 
 int qux(void) {
-    return 42;
+	return 42;
 }
 
 int bar(void) {
@@ -13,9 +13,11 @@ int bar(void) {
 }
 
 int main(void) {
-    int some_int = 10;
-    int foo_result = foo(some_int);
-    int bar_result = bar();
-
+  int some_int = 10;
+  int foo_result = foo(some_int);
+  int n = 10;
+  if (n < 42) {
+		int bar_result = bar();
+	}
     return 0;
 }

examples/example_mod1/example_mod1.ll → examples/example_mod2/example_mod2.ll

@@ -1,5 +1,5 @@
-; ModuleID = '../examples/mod1/example_mod1.bc'
-source_filename = "example_mod1.c"
+; ModuleID = 'example_mod2.bc'
+source_filename = "example_mod2.c"
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
 
@@ -44,6 +44,7 @@ entry:
   %retval = alloca i32, align 4
   %some_int = alloca i32, align 4
   %foo_result = alloca i32, align 4
+  %n = alloca i32, align 4
   %bar_result = alloca i32, align 4
   store i32 0, i32* %retval, align 4
   call void @llvm.dbg.declare(metadata i32* %some_int, metadata !31, metadata !12), !dbg !32
@@ -52,10 +53,20 @@ entry:
   %0 = load i32, i32* %some_int, align 4, !dbg !35
   %call = call i32 @foo(i32 %0), !dbg !36
   store i32 %call, i32* %foo_result, align 4, !dbg !34
-  call void @llvm.dbg.declare(metadata i32* %bar_result, metadata !37, metadata !12), !dbg !38
-  %call1 = call i32 @bar(), !dbg !39
-  store i32 %call1, i32* %bar_result, align 4, !dbg !38
-  ret i32 0, !dbg !40
+  call void @llvm.dbg.declare(metadata i32* %n, metadata !37, metadata !12), !dbg !38
+  store i32 10, i32* %n, align 4, !dbg !38
+  %1 = load i32, i32* %n, align 4, !dbg !39
+  %cmp = icmp slt i32 %1, 42, !dbg !41
+  br i1 %cmp, label %if.then, label %if.end, !dbg !42
+
+if.then:                                          ; preds = %entry
+  call void @llvm.dbg.declare(metadata i32* %bar_result, metadata !43, metadata !12), !dbg !45
+  %call1 = call i32 @bar(), !dbg !46
+  store i32 %call1, i32* %bar_result, align 4, !dbg !45
+  br label %if.end, !dbg !47
+
+if.end:                                           ; preds = %if.then, %entry
+  ret i32 0, !dbg !48
 }
 
 attributes #0 = { noinline nounwind optnone uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "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-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" }
@@ -66,7 +77,7 @@ attributes #1 = { nounwind readnone speculatable }
 !llvm.ident = !{!6}
 
 !0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 5.0.1 (tags/RELEASE_500/final)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
-!1 = !DIFile(filename: "example_mod1.c", directory: "/mnt/Documents/work/university/muni/msc/thesis/APEX/examples/mod1")
+!1 = !DIFile(filename: "example_mod2.c", directory: "/mnt/Documents/work/university/muni/msc/thesis/APEX/examples/experiments/example_mod2")
 !2 = !{}
 !3 = !{i32 2, !"Dwarf Version", i32 4}
 !4 = !{i32 2, !"Debug Info Version", i32 3}
@@ -97,12 +108,20 @@ attributes #1 = { nounwind readnone speculatable }
 !29 = !DILocation(line: 12, column: 5, scope: !24)
 !30 = distinct !DISubprogram(name: "main", scope: !1, file: !1, line: 15, type: !21, isLocal: false, isDefinition: true, scopeLine: 15, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
 !31 = !DILocalVariable(name: "some_int", scope: !30, file: !1, line: 16, type: !10)
-!32 = !DILocation(line: 16, column: 9, scope: !30)
+!32 = !DILocation(line: 16, column: 7, scope: !30)
 !33 = !DILocalVariable(name: "foo_result", scope: !30, file: !1, line: 17, type: !10)
-!34 = !DILocation(line: 17, column: 9, scope: !30)
-!35 = !DILocation(line: 17, column: 26, scope: !30)
-!36 = !DILocation(line: 17, column: 22, scope: !30)
-!37 = !DILocalVariable(name: "bar_result", scope: !30, file: !1, line: 18, type: !10)
-!38 = !DILocation(line: 18, column: 9, scope: !30)
-!39 = !DILocation(line: 18, column: 22, scope: !30)
-!40 = !DILocation(line: 20, column: 5, scope: !30)
+!34 = !DILocation(line: 17, column: 7, scope: !30)
+!35 = !DILocation(line: 17, column: 24, scope: !30)
+!36 = !DILocation(line: 17, column: 20, scope: !30)
+!37 = !DILocalVariable(name: "n", scope: !30, file: !1, line: 18, type: !10)
+!38 = !DILocation(line: 18, column: 7, scope: !30)
+!39 = !DILocation(line: 19, column: 7, scope: !40)
+!40 = distinct !DILexicalBlock(scope: !30, file: !1, line: 19, column: 7)
+!41 = !DILocation(line: 19, column: 9, scope: !40)
+!42 = !DILocation(line: 19, column: 7, scope: !30)
+!43 = !DILocalVariable(name: "bar_result", scope: !44, file: !1, line: 20, type: !10)
+!44 = distinct !DILexicalBlock(scope: !40, file: !1, line: 19, column: 15)
+!45 = !DILocation(line: 20, column: 7, scope: !44)
+!46 = !DILocation(line: 20, column: 20, scope: !44)
+!47 = !DILocation(line: 21, column: 2, scope: !44)
+!48 = !DILocation(line: 22, column: 5, scope: !30)