Change position of local variable to lowest address in stack - c

I have written simple code in order to study assembler.
Task im trying to do is in assembler part: mycode.s
Im curious, how would i be able to manually change position of my local variable int var ( it is located on stack, because its local) to lowest possible location(address) in stack...
Im using following code:
#include<stdio.h>
int c = 0;
int main()
{
int var = 0;
if( var == c)
{
printf("here");
}
return 0;
}
when i assemble my code with gcc -m32 -S mycode.c
i get:
.file "mycode.c"
.globl c
.bss
.align 4
.type c, #object
.size c, 4
c:
.zero 4
.section .rodata
.LC0:
.string "here"
.text
.globl main
.type main, #function
main:
leal 4(%esp), %ecx
andl $-16, %esp
pushl -4(%ecx)
pushl %ebp
movl %esp, %ebp
pushl %ecx
subl $20, %esp
movl $0, -12(%ebp)
movl c, %eax
cmpl %eax, -12(%ebp)
jne .L2
subl $12, %esp
pushl $.LC0
call printf
addl $16, %esp
.L2:
movl $0, %eax
movl -4(%ebp), %ecx
leave
leal -4(%ecx), %esp
ret
.size main, .-main
.ident "GCC: (Ubuntu 7.2.0-8ubuntu3.2) 7.2.0"
.section .note.GNU-stack,"",#progbits
What instructions can i use to check the current stack addresses and find the lowest possible address (for example: 0x003 < 0x004).. Thanks

Related

stackframe dosen't get eliminated from the stack?

I wrote a single c program that prints input to std output. Then I converted it to assembly language. By the way I am using AT&T Syntax.
This is the simple C code.
#include <stdio.h>
int main()
{
int c;
while ((c = getchar ()) != EOF)
{
putchar(c);
}
return 0;
}
int c is a local variable.
Then I converted it to assembly language.
.file "question_1.c"
.text
.globl main
.type main, #function
//prolog
main:
leal 4(%esp), %ecx
andl $-16, %esp
pushl -4(%ecx)
pushl %ebp
movl %esp, %ebp
pushl %ecx
subl $20, %esp // we add 20 bytes to the stack
jmp .L2
.L3:
subl $12, %esp
pushl -12(%ebp)
call putchar
addl $16, %esp
.L2:
call getchar
movl %eax, -12(%ebp)
cmpl $-1, -12(%ebp)
jne .L3
//assumption this is the epilog
movl $0, %eax
movl -4(%ebp), %ecx
leave
leal -4(%ecx), %esp
ret
.size main, .-main
.ident "GCC: (Ubuntu 4.9.4-2ubuntu1) 4.9.4"
.section .note.GNU-stack,"",#progbits
normally in the epilog we are supposed to addl 20 because in the prolog we subl 20.
So the is the stack frame still there?
Or am I missing out a crucial point?
I also have a question regarding the main function. Normally functions are normally "called" but where does it happen in the assembly code?
Thank you in advance.
Just after the main label, leal 4(%esp), %ecx saves four plus the stack pointer in %ecx. At the end of the routine, leal -4(%ecx), %esp writes four less than the saved value to the stack pointer. This directly restores the original value, instead of doing it by adding the amount that was subtracted.

IA32 assembly language stack trace

I have been struggling with an IA32 assembly language question I received during a university lab. I can't quite firgure out how to go about it, and was wondering if anyone here could give me a hand.
The question says: Draw the runtime stack at the following points during execution, starting from the parameter passed to fact_do and ending with the top of the stack. In each case label each 4 byte memory location and indicate the contents when known. Show where %ebp and %esp point to.
The point I can't quite draw the stack trace for is: "Just before the leave instruction in fact_do (leave is equivalent to movl %ebp, %esp followed by popl %ebp)."
I think what is expected is something along the lines of:
_____________
| | <- %ebp
| . |
| . |
| . |
|_____________|
|_5___________|
|_Rtn_Address_| <- %esp
The following is the code that is to be traced:
.file "fact.c"
.section .rodata
.LC0:
.string "%d! = %d\n" #string is initialized
.text
.globl main
.type main, #function
main:
pushl %ebp
movl %esp, %ebp
andl $-16, %esp
subl $16, %esp
movl $5, (%esp) #(%esp) <- 5 [(%esp) = n]
call fact_do
movl $.LC0, %edx #set up print statement
movl %eax, 8(%esp) #8(%esp) <- %eax [%eax = 120, return value of fact_do]
movl $5, 4(%esp) #4(%esp) <- 5
movl %edx, (%esp) #print statement location
call printf #print output
movl $0, %eax #%eax <- 0
leave
ret
.size main, .-main
.globl fact_do
.type fact_do, #function
fact_do:
pushl %ebp #push %ebp onto stack
movl %esp, %ebp #%ebp <- %esp
subl $16, %esp #%esp <- %esp - 16
movl $1, -4(%ebp) #-4(%ebp) <- 1 [-4(%ebp) = result]
.L4:
movl -4(%ebp), %eax #%eax <- -4(%ebp)
imull 8(%ebp), %eax #%eax <- %eax * 8(%ebp) [8(%ebp) = n]
movl %eax, -4(%ebp) #-4(%ebp) <- %eax [result <- result*n]
subl $1, 8(%ebp) #8(%ebp)-- [n--]
cmpl $1, 8(%ebp) #8(%ebp) > 1 ? [n > 1]
jg .L4 #Repeat loop if 8(%ebp) > 1
movl -4(%ebp), %eax #%eax <- -4(%ebp) [%eax <- 120, result]
leave
ret
.size fact_do, .-fact_do
.ident "GCC: (GNU) 4.4.7 20120313 (Red Hat 4.4.7-4)"
.section .note.GNU-stack,"",#progbits
Optimized:
.file "fact.c"
.text
.globl fact_do
.type fact_do, #function
fact_do:
pushl %ebp #push %ebp (for return)
movl %esp, %ebp #%ebp <- %esp
movl 8(%ebp), %edx #%edx <- 8(%ebp) [%edx = n]
movl $1, %eax #%eax <- 1 [%eax = result]
.L2:
imull %edx, %eax #%eax <- %edx * %eax [result <- n * result]
subl $1, %edx #%edx-- [n--]
cmpl $1, %edx #%edx > 1 ? [n > 1]
jg .L2 #repeat if %edx > 1 [if n > 1]
popl %ebp #%ebp <- top value on stack (for return)
ret
.size fact_do, .-fact_do
.section .rodata.str1.1,"aMS",#progbits,1
.LC0:
.string "%d! = %d\n" #prepare string
.text
.globl main
.type main, #function
main:
pushl %ebp
movl %esp, %ebp
andl $-16, %esp
subl $16, %esp
movl $5, (%esp) #%esp <- 5
call fact_do
movl %eax, 8(%esp) #8(%esp) <- %eax [%eax = result]
movl $5, 4(%esp) #4(%esp) <- 5
movl $.LC0, (%esp) #set up print statement
call printf #print
movl $0, %eax #%eax <- 0
leave
ret
.size main, .-main
.ident "GCC: (GNU) 4.4.7 20120313 (Red Hat 4.4.7-4)"
.section .note.GNU-stack,"",#progbits
Which translates to the following in C:
#include <stdio.h>
int fact_do(int n);
int main(){
printf("%d! = %d\n", 5, fact_do(5));
return 0;
}
int fact_do(int n){
int result = 1;
do {
result *= n;
n = n-1;
} while (n > 1);
return result;
}

evaluation of operand in sizeof operator

As sizeof operator evaluates operand if it's a VLA so I tried to test it as :
#include<stdio.h>
int main(void)
{
int sz=20,i=0,j=0;
int arr[sz];
printf("%d\n",sizeof((++i,sz)));
printf("%d\n",sizeof((++j,arr)));
printf("%d\n%d\n",i,j);
}
I thought that i won't increment as sz is not VLA but j would increment as arr is VLA.
But in the output, none of the i and j incremented.
Quoting my answer to another question:
The "conversion" is due to the subtraction operator. You can see a similar, and perhaps more surprising, result with the comma operator:
printf("%zu\n", sizeof(1, a));
will also print sizeof(int *), because of the comma operator resulting in a getting used in a value context.
Basically, due to the comma operator, the type of arr is a pointer, and the size of the VLA doesn't come into the picture. See my linked answer for details.
Not much of an explanation but I suspect it's some compiler optimization for the comma operator. The value of a comma operation is the value of the last expression. With the compiler knowing sizeof is a unary operator and presented with a comma operation, it doesn't bother to evaluate any but the last expression (regardless whether the last was a reference to a VLA or not).
I wrote some test programs (gcc 4.3.3 on Ubuntu 9.04):
$ cat test.c # with sizeof
#include <stdio.h>
int main(void)
{
int x = 0;
printf("%d\n",
sizeof( printf("%s%d\n", "comma!", ++x), x));
}
$ gcc -S test.c
$ cat test.s
.file "test.c"
.section .rodata
.LC0:
.string "%d\n"
.text
.globl main
.type main, #function
main:
leal 4(%esp), %ecx
andl $-16, %esp
pushl -4(%ecx)
pushl %ebp
movl %esp, %ebp
pushl %ecx
subl $36, %esp
movl $0, -8(%ebp)
movl $4, 4(%esp)
movl $.LC0, (%esp)
call printf
addl $36, %esp
popl %ecx
popl %ebp
leal -4(%ecx), %esp
ret
.size main, .-main
.ident "GCC: (Ubuntu 4.3.3-5ubuntu4) 4.3.3"
.section .note.GNU-stack,"",#progbits
Note the absences of the string literals and the second printf() call.
$ cat test-alt.c # without sizeof
#include <stdio.h>
int main(void)
{
int x = 0;
printf("%d\n",
( printf("%s%d\n", "comma!", ++x), x));
}
$ gcc -S test-alt.c
$ cat test-alt.s
.file "test-alt.c"
.section .rodata
.LC0:
.string "comma!"
.LC1:
.string "%s%d\n"
.LC2:
.string "%d\n"
.text
.globl main
.type main, #function
main:
leal 4(%esp), %ecx
andl $-16, %esp
pushl -4(%ecx)
pushl %ebp
movl %esp, %ebp
pushl %ecx
subl $36, %esp
movl $0, -8(%ebp)
addl $1, -8(%ebp)
movl -8(%ebp), %eax
movl %eax, 8(%esp)
movl $.LC0, 4(%esp)
movl $.LC1, (%esp)
call printf
movl -8(%ebp), %eax
movl %eax, 4(%esp)
movl $.LC2, (%esp)
call printf
addl $36, %esp
popl %ecx
popl %ebp
leal -4(%ecx), %esp
ret
.size main, .-main
.ident "GCC: (Ubuntu 4.3.3-5ubuntu4) 4.3.3"
.section .note.GNU-stack,"",#progbits
It may be documented somewhere but I wouldn't know where to look.
sizeof evaluates at compile time. In C99 for variable length arrays it will wait until runtime. Check this answer to a similar question.
Compiler knows the array size: it is obviously 20. I don't think that sz is VLA. Try to use use array size as function parameter, for eaxmple:
void Function(int size)
{
int arr[size];
...
}
BTW, to understand what happens, it is recommended to read Assembly code produced by compiler. Check whether sizeof is replaced by a constant already at compile time.

C - what is the return value of a semicolon?

im just curious about the following example
#include<stdio.h>
int test();
int test(){
// int a = 5;
// int b = a+1;
return ;
}
int main(){
printf("%u\n",test());
return 0;
}
i compiled it with 'gcc -Wall -o semicolon semicolon.c' to create an executable
and 'gcc -Wall -S semicolon.c' to get the assembler code which is:
.file "semicolon.c"
.text
.globl test
.type test, #function
test:
pushl %ebp
movl %esp, %ebp
subl $4, %esp
leave
ret
.size test, .-test
.section .rodata
.LC0:
.string "%u\n"
.text
.globl main
.type main, #function
main:
leal 4(%esp), %ecx
andl $-16, %esp
pushl -4(%ecx)
pushl %ebp
movl %esp, %ebp
pushl %ecx
subl $20, %esp
call test
movl %eax, 4(%esp)
movl $.LC0, (%esp)
call printf
movl $0, %eax
addl $20, %esp
popl %ecx
popl %ebp
leal -4(%ecx), %esp
ret
.size main, .-main
.ident "GCC: (Ubuntu 4.3.3-5ubuntu4) 4.3.3"
.section .note.GNU-stack,"",#progbits
since im not such an assembler pro, i only know that printf prints what is in eax
but i dont fully understand what 'movl %eax, 4(%esp)' means which i assume fills eax before calling test
but what is the value then? what means 4(%esp) and what does the value of esp mean?
if i uncomment the lines in test() printf prints 6 - which is written in eax ^^
Your assembly language annotated:
test:
pushl %ebp # Save the frame pointer
movl %esp, %ebp # Get the new frame pointer.
subl $4, %esp # Allocate some local space on the stack.
leave # Restore the old frame pointer/stack
ret
Note that nothing in test touches eax.
.size test, .-test
.section .rodata
.LC0:
.string "%u\n"
.text
.globl main
.type main, #function
main:
leal 4(%esp), %ecx # Point past the return address.
andl $-16, %esp # Align the stack.
pushl -4(%ecx) # Push the return address.
pushl %ebp # Save the frame pointer
movl %esp, %ebp # Get the new frame pointer.
pushl %ecx # save the old top of stack.
subl $20, %esp # Allocate some local space (for printf parameters and ?).
call test # Call test.
Note that at this point, nothing has modified eax. Whatever came into main is still here.
movl %eax, 4(%esp) # Save eax as a printf argument.
movl $.LC0, (%esp) # Send the format string.
call printf # Duh.
movl $0, %eax # Return zero from main.
addl $20, %esp # Deallocate local space.
popl %ecx # Restore the old top of stack.
popl %ebp # And the old frame pointer.
leal -4(%ecx), %esp # Fix the stack pointer,
ret
So, what gets printed out is whatever came in to main. As others have pointed out it is undefined: It depends on what the startup code (or the OS) has done to eax previously.
The semicolon has no return value, what you have there is an "empty return", like the one used to return from void functions - so the function doesn't return anything.
This actually causes a warning when compiling:
warning: `return' with no value, in function returning non-void
And I don't see anything placed in eax before calling test.
About 4(%esp), this means taking the value from the stack pointer (esp) + 4. I.e. the one-before-last word on the stack.
The return value of an int function is passed in the EAX register. The test function does not set the EAX register because no return value is given. The result is therefore undefined.
A semicolon indeed has no value.
I think the correct answer is that a return <nothing> for an int function is an error, or at least has undefined behavor. That's why compiling this with -Wall yields
semi.c: In function ‘test’:
semi.c:6: warning: ‘return’ with no value, in function returning non-void
As for what the %4,esp holds... it's a location on the stack where nothing was (intentionally) stored, so it will likely return whatever junk is found at that location. This could be the last expression evaluated to variables in the function (as in your example) or something completely different. This is what "undefined" is all about. :)

How do C compilers implement functions that return large structures?

The return value of a function is usually stored on the stack or in a register. But for a large structure, it has to be on the stack. How much copying has to happen in a real compiler for this code? Or is it optimized away?
For example:
struct Data {
unsigned values[256];
};
Data createData()
{
Data data;
// initialize data values...
return data;
}
(Assuming the function cannot be inlined..)
None; no copies are done.
The address of the caller's Data return value is actually passed as a hidden argument to the function, and the createData function simply writes into the caller's stack frame.
This is known as the named return value optimisation. Also see the c++ faq on this topic.
commercial-grade C++ compilers implement return-by-value in a way that lets them eliminate the overhead, at least in simple cases
...
When yourCode() calls rbv(), the compiler secretly passes a pointer to the location where rbv() is supposed to construct the "returned" object.
You can demonstrate that this has been done by adding a destructor with a printf to your struct. The destructor should only be called once if this return-by-value optimisation is in operation, otherwise twice.
Also you can check the assembly to see that this happens:
Data createData()
{
Data data;
// initialize data values...
data.values[5] = 6;
return data;
}
here's the assembly:
__Z10createDatav:
LFB2:
pushl %ebp
LCFI0:
movl %esp, %ebp
LCFI1:
subl $1032, %esp
LCFI2:
movl 8(%ebp), %eax
movl $6, 20(%eax)
leave
ret $4
LFE2:
Curiously, it allocated enough space on the stack for the data item subl $1032, %esp, but note that it takes the first argument on the stack 8(%ebp) as the base address of the object, and then initialises element 6 of that item. Since we didn't specify any arguments to createData, this is curious until you realise this is the secret hidden pointer to the parent's version of Data.
But for a large structure, it has to be on the heap stack.
Indeed so! A large structure declared as a local variable is allocated on the stack. Glad to have that cleared up.
As for avoiding copying, as others have noted:
Most calling conventions deal with "function returning struct" by passing an additional parameter that points the location in the caller's stack frame in which the struct should be placed. This is definitely a matter for the calling convention and not the language.
With this calling convention, it becomes possible for even a relatively simple compiler to notice when a code path is definitely going to return a struct, and for it to fix assignments to that struct's members so that they go directly into the caller's frame and don't have to be copied. The key is for the compiler to notice that all terminating code paths through the function return the same struct variable. If that's the case, the compiler can safely use the space in the caller's frame, eliminating the need for a copy at the point of return.
There are many examples given, but basically
This question does not have any definite answer. it will depend on the compiler.
C does not specify how large structs are returned from a function.
Here's some tests for one particular compiler, gcc 4.1.2 on x86 RHEL 5.4
gcc trivial case, no copying
[00:05:21 1 ~] $ gcc -O2 -S -c t.c
[00:05:23 1 ~] $ cat t.s
.file "t.c"
.text
.p2align 4,,15
.globl createData
.type createData, #function
createData:
pushl %ebp
movl %esp, %ebp
movl 8(%ebp), %eax
movl $1, 24(%eax)
popl %ebp
ret $4
.size createData, .-createData
.ident "GCC: (GNU) 4.1.2 20080704 (Red Hat 4.1.2-46)"
.section .note.GNU-stack,"",#progbits
gcc more realistic case , allocate on stack, memcpy to caller
#include <stdlib.h>
struct Data {
unsigned values[256];
};
struct Data createData()
{
struct Data data;
int i;
for(i = 0; i < 256 ; i++)
data.values[i] = rand();
return data;
}
[00:06:08 1 ~] $ gcc -O2 -S -c t.c
[00:06:10 1 ~] $ cat t.s
.file "t.c"
.text
.p2align 4,,15
.globl createData
.type createData, #function
createData:
pushl %ebp
movl %esp, %ebp
pushl %edi
pushl %esi
pushl %ebx
movl $1, %ebx
subl $1036, %esp
movl 8(%ebp), %edi
leal -1036(%ebp), %esi
.p2align 4,,7
.L2:
call rand
movl %eax, -4(%esi,%ebx,4)
addl $1, %ebx
cmpl $257, %ebx
jne .L2
movl %esi, 4(%esp)
movl %edi, (%esp)
movl $1024, 8(%esp)
call memcpy
addl $1036, %esp
movl %edi, %eax
popl %ebx
popl %esi
popl %edi
popl %ebp
ret $4
.size createData, .-createData
.ident "GCC: (GNU) 4.1.2 20080704 (Red Hat 4.1.2-46)"
.section .note.GNU-stack,"",#progbits
gcc 4.4.2### has grown a lot, and does not copy for the above non-trivial case.
.file "t.c"
.text
.p2align 4,,15
.globl createData
.type createData, #function
createData:
pushl %ebp
movl %esp, %ebp
pushl %edi
pushl %esi
pushl %ebx
movl $1, %ebx
subl $1036, %esp
movl 8(%ebp), %edi
leal -1036(%ebp), %esi
.p2align 4,,7
.L2:
call rand
movl %eax, -4(%esi,%ebx,4)
addl $1, %ebx
cmpl $257, %ebx
jne .L2
movl %esi, 4(%esp)
movl %edi, (%esp)
movl $1024, 8(%esp)
call memcpy
addl $1036, %esp
movl %edi, %eax
popl %ebx
popl %esi
popl %edi
popl %ebp
ret $4
.size createData, .-createData
.ident "GCC: (GNU) 4.1.2 20080704 (Red Hat 4.1.2-46)"
.section .note.GNU-stack,"",#progbits
In addition, VS2008 (compiled the above as C) will reserve struct Data on the stack of createData() and do a rep movsd loop to copy it back to the caller in Debug mode, in Release mode it will move the return value of rand() (%eax) directly back to the caller
typedef struct {
unsigned value[256];
} Data;
Data createData(void) {
Data r;
calcualte(&r);
return r;
}
Data d = createData();
msvc(6,8,9) and gcc mingw(3.4.5,4.4.0) will generate code like the following pseudocode
void createData(Data* r) {
calculate(&r)
}
Data d;
createData(&d);
gcc on linux will issue a memcpy() to copy the struct back on the stack of the caller. If the function has internal linkage, more optimizations become available though.

Resources