Calling Convention in x86 Assembly
- Memory and Addressing Modes
- Calling Convention
Modern (i.e 386 and beyond) x86 processor have eight 32-bit general purpose registers, as decipted in following Figure. The register names are mostly hsitorical. For example,
EAX used to be called accumulator since it was used to by a number of arithmetic operations, and
ECX was known as the counter since it was used to hold a loop index. Whereas most of the registers have lost their special purposes in the modern insturction set, by convention, two are reserved for special purposes —the stack pointer (
ESP) and the base pointer (
EDX registers, subsections may be used. For example, the least significant 2 bytes of
EAX can be treated as a 16-bit register called
AX. The least significant byte of
AX can be used as a signle 8-bit reigster called
AL, while the most siginificant byte of
AX can be used as a single 8-bit register called
AH. These names refer to the same physical register. When a two-byte quantity is placed into
DX, the update affects the value of
EDX. These sub-registers are mainly hold-overs from older, 16-bit version of the instruction set. However, they are sometimes convenient when dealing with data that are samller than 32-bits (e.g. 1-byte ASCII characters).
When referring to registers in assembly languages, the names are not case-sensitive. For example, the names
eax refer to the same register.
Memory and Addressing Modes
Declaring Static Data Regions
You can declare static data regions (analogous to global variables) in x86 assembly using special assembler directives for this purpose. Data declarations should be preceded by the
.DATA directive. Following this directive, the directive
DD can be used to declare one, two, and four byte data locations, respectively. Declared locations can be labeled with names for later reference —this is similar to declaring variables by name, but abides by some lower level rules. For example, location declared in sequence will be located in memory next to one another.
.DATA var DB 64 ; Declare a byte, referred to as location var, containing the value 64. var2 DB ? ; Declare an uninitialized byte, referred to as location var2. DB 10 ; Declare a byte with no label, containing the value 10. Its location is var2 + 1. X DW ? ; Declare a 2-byte uninitialized value, referred to as location X. Y DD 30000 ; Declare a 4-byte value, referred to as location Y, initialized to 30000.
Unlike in high level languages where arrays can have many dimensions and are accessed by indices, arrays in x86 assemlby language are simply a number of cells located contiguously in memory. An array can be declared by just listing the values, as in the first example below. Two other common methods used for declaring arrays of data are the
DUP directive and the use of string literals. The
DUP directive tells the assembler to duplicate an expression a given number of times. For example,
4 DUP(2) is equivalent to
2, 2, 2, 2.
Z DD 1, 2, 3 ; Declare three 4-byte values, initialized to 1, 2, and 3. The value of location Z + 8 will be 3. bytes DB 10 DUP(?) ; Declare 10 uninitialized bytes starting at location bytes. arr DD 100 DUP(0) ; Declare 100 4-byte words starting at location arr, all initialized to 0 str DB 'hello',0 ; Declare 6 bytes starting at the address str, initialized to the ASCII character values for hello and the null (0) byte.
Modern x86-compatible processors are capable of addressing up to 232 bytes of memory: memory addresses are 32-bits wide. In the examples above, where we used labels to refer to memory regions, these labels are actually replaced by the assembler with 32-bit quantities that specify addresses in memory. In addition to supporting referring to memory regions by labels (i.e. constant values), the x86 provides a flexible scheme for computing and referring to memory addresses: up to two of the 32-bit registers and a 32-bit signed constant can be added together to compute a memory address. One of the registers can be optionally pre-multiplied by 2, 4, or 8.
The addressing modes can be used with many x86 instructions. Here we illustrate some examples using the
mov instruction that moves (copies) data between registers and memory. This instruction has two operands: the first is the destination and the second specifies the source.
Some examples of
mov instructions using address computations are:
mov eax, [ebx] ; Move the 4 bytes in memory at the address contained in EBX into EAX mov [var], ebx ; Move the contents of EBX into the 4 bytes at memory address var. (Note, var is a 32-bit constant). mov eax, [esi-4] ; Move 4 bytes at memory address ESI + (-4) into EAX mov [esi+eax], cl ; Move the contents of CL into the byte at address ESI+EAX mov edx, [esi+4*ebx] ; Move the 4 bytes of data at address ESI+4*EBX into EDX
Some examples of invalid address calculations include:
mov eax, [ebx-ecx] ; Can only add register values mov [eax+esi+edi], ebx ; At most 2 registers in address computation
In general, the intended size of the data item at a given memory address can be inferred from the assembly code instruction in which it is referenced. For example, in all of the above instructions, the size of the memory regions could be inferred from the size of the register operand. When we were loading a 32-bit register, the assembler could infer that the region of memory we were referring to was 4 bytes wide. When we were storing the value of a one byte register to memory, the assembler could infer that we wanted the address to refer to a single byte in memory.
However, in some cases the size of a referred-to memory region is ambiguous. Consider the instruction
move [ebx], 2. Should this instruction move the value 2 into the single byte at address
EBX? Perhaps it should move the 32-bit integer representation of 2 into the 4-bytes starting at address
EBX. Since either is a valid possible interpretation, the assemler must be explicitly directed as to which is correct. The size directive
WORD PTR, and
DWORD PTR serve this purpose, indicating size of 1, 2, and 4 bytes respectively.
mov BYTE PTR [ebx], 2 ; Move 2 into the single byte at the address stored in EBX. mov WORD PTR [ebx], 2 ; Move the 16-bit integer representation of 2 into the 2 bytes starting at the address in EBX. mov DWORD PTR [ebx], 2 ; Move the 32-bit integer representation of 2 into the 4 bytes starting at the address in EBX.
Machine instructions generally fall into three categories: data movement, arithmetic/logic, and control-flow. In this section, we will look at important examples of x86 instructions from each category. This section should not be considered an exhausitive list of x86 instructions, but rather a useful subset. Fro a complete list, see Intel’s instruction set reference.
We use the follow notation:
<reg32> Any 32-bit register (EAX, EBX, ECX, EDX, ESI, EDI, ESP, or EBP) <reg16> Any 16-bit register (AX, BX, CX, or DX) <reg8> Any 8-bit register (AH, BH, CH, DH, AL, BL, CL, or DL) <reg> Any register <mem> A memory address (e.g., [eax], [var + 4], or dword ptr [eax+ebx]) <con32> Any 32-bit constant <con16> Any 16-bit constant <con8> Any 8-bit constant <con> Any 8-, 16-, or 32-bit constant
Data Movement Instructions
mov — Move (Opcodes: 88, 89, 8A, 8B, 8C, 8E, …)
movinstruction copies the data item referred to by its second operand (i.e. register contents, memory contents, or a constant value) into the location referred to by its first operand (i.e. a register or memory). While register-to-register moves are possible, direct memory-to-memory are not. In cases where memory transfers are desired, the source memory contents must first be loaded into a register, then can be stored to the destination memory address.
mov <reg>,<reg> mov <reg>,<mem> mov <mem>,<reg> mov <reg>,<const> mov <mem>,<const>
mov eax, ebx ;copy the value in ebx into eax mov byte ptr [var], 5 ;store the value 5 into the byte at location var
push — Push stack (Opcodes: FF, 89, 8A, 8B, 8C, 8E, …)
pushinstruction places its operand onto the top of the hardware supported stack in memory. Specifically,
pushfirst decrements ESP by 4, then places its operand into the contents of the 32-bit locations at address [ESP]. ESP (the stack pointer) is decremented by push since the x86 stack grows down - i.e. the stack grows from high addresses to lower addresses.
push <reg32> push <mem> push <con32>
push eax ;push eax on the stack push [var] ;push the 4 bytes at address var onto the stack
pop — Pop stack
popinstruction removes the 4-byte data element from the top of the hardware-supported into the specified operand (i.e. register or memory location). It first moves the 4 bytes located at memory location [ESP] into the specified register or memory location, and then increments ESP by 4.
pop <reg32> pop <mem>
pop edi ; pop the top element of the stack into EDI. pop [ebx] ; pop the top element of the stack into memory at the four bytes starting at location EBX.
lea — Load effective address
leainstruction places the address specified by its second operand into the register specified by it first operand. Note, the contents of the memory location are not loaded, only the effective address is computed and placed into the register. This is useful for obtaining a pointer into a memory region.
lea <reg32>, <mem>
lea edi, [ebx+4*esi] ;the quantity EBX+4*ESI is placed in EDI. lea eax, [var] ;the value in var is placed in EAX. lea eax, [val] ;the value val is placed in EAX.
Arithmetic and Logic Instructions
add — Integer Addition
addinstruction adds together its two operands, storing the result in its first operand. Note, whereas both operands may be registers, at most one operand may be a memory location.
add <reg>, <reg> add <reg>, <mem> add <mem>, <reg> add <reg>, <con> add <mem>, <con>
add eax, 10 ;EAX ← EAX + 10 add BYTE PTR [var], 10 ;add 10 to the single byte stored at memory address var
sub — Integer Subtraction
subinstruction stores in the value of its first operand the result of subtracting the value of its second operand from the value of its first operand. As with
sub <reg>,<reg> sub <reg>,<mem> sub <mem>,<reg> sub <reg>,<con> sub <mem>,<con>
sub al, ah ;AL ← AL - AH sub eax, 216 ;subtract 216 from the value stored in EAX
inc, dec — Increment, Decrement
incinstruction increments the contents of its operand by one. The
decinstruction decrements the contents of its operand by one.
inc <reg> inc <mem> dec <reg> dec <mem>
dec eax ;subtract one from the contents of EAX. inc DWORD PTR [var] ;add one to the 32-bit integer stored at location var
imul — Integer Multiplication
imulinstruction has two basic formats: two-operand (fist two syntax listings above) and three-operand (last two syntax list above).
The two-operand form multiplies its two operand together and stores the result in the first operand. The result (i.e. first) operand must be a register.
Thw three-operand form multiplies its second the third operands together and stores the result in its first operand. Again, the result operand must be a register. Furthermore, the third operand is restricted to being a constant value.
imul <reg32>,<reg32> imul <reg32>,<mem> imul <reg32>,<reg32>,<con> imul <reg32>,<mem>,<con>
imul eax, [var] ;multiply the contents of EAX by the 32-bit contents of the memory location var. Store the result in EAX. imul esi, edi, 25 ;ESI → EDI * 25
idiv — Integer Division
idivinstruction divides the contents of the 64-bit integer EDX:EAX (constructed by viewing EDX as the most significant four bytes and EAX as the least significant four bytes) by the specified operand value. The quotient result of the division is stored into EAX, while the reminder is placed in EDX.
idiv <reg32> idiv <mem>
idiv ebx ;divide the contents of EDX:EAX by the contents of EBX. Place the quotient in EAX and the remainder in EDX. idiv DWORD PTR [var] ;divide the contents of EDX:EAX by the 32-bit value stored at memory location var. Place the quotient in EAX and the remainder in EDX.
and, or, xor — Bitwise logical and, or and exclusive or
The instructions perform the specified logical opeartion (logical bitwise and, or, exclusive or, respectively) on their operands, placing the result in the first operand location.
and <reg>,<reg> and <reg>,<mem> and <mem>,<reg> and <reg>,<con> and <mem>,<con> or <reg>,<reg> or <reg>,<mem> or <mem>,<reg> or <reg>,<con> or <mem>,<con> xor <reg>,<reg> xor <reg>,<mem> xor <mem>,<reg> xor <reg>,<con> xor <mem>,<con>
and eax, 0fH ;clear all but the last 4 bits of EAX. xor edx, edx ;set the contents of EDX to zero.
not — Bitwise Logical Not
Logically negates the operan contents (that is, flips all bit values in the operand).
not <reg> not <mem>
not BYTE PTR [var] ;negate all bits in the byte at the memory location var.
neg — Negate
Performs the two’s complement negation of the operand contents.
neg <reg> neg <mem>
neg eax ;EAX → - EAX
shl, shr — Shift Left, Shift Right
These instructions shift the bits in their first operand’s content left and right, padding the resulting empty bit poistions with zeros. The shifited operand can be shifed up to 31 places. The number of bits to shift is specified by the second operand, which can be either an 8-bit constant or the register CL. In either case, shifts counts of greater then 31 are performed modulo 32.
shl <reg>,<con8> shl <mem>,<con8> shl <reg>,<cl> shl <mem>,<cl> shr <reg>,<con8> shr <mem>,<con8> shr <reg>,<cl> shr <mem>,<cl>
shl eax, 1 ;Multiply the value of EAX by 2 (if the most significant bit is 0) shr ebx, cl ;Store in EBX the floor of result of dividing the value of EBX by 2n wheren is the value in CL.
Control Flow Instructions
The x86 processor maintains an instruction pointer (IP) register that is a 32-bit value indicating the location in memory where the current instruction starts. Normally, it increments to point the next instruction in memory begins after execution an instruction. The IP register cannot be mainipulated directly, but is updated implicitly by provided control flow instructions.
We use the notation
<label> to refer to labeled locations in the program text. Labels can be inserted anywhere in x86 code text by entering a label name followed by a colon. For example,
mov esi, [ebp+8] begin: xor ecx, ecx mov eax, [esi]
The second instruction in this code fragment is labeled
begin. Elsewhere in the code, we can refer to the memory location that this instruction is located at in memory using the more convenient symbolic name
begin. The label is just a convenient way of expressing the location instead of its 32-bit value.
jmp — Jump
Transfers program control flow to the instruction at the memory location indicated by the operand.
jmp begin — Jump to the instruction labeled begin.
jcondition — Conditional Jump
These instructions are coditional jumps that are based on the status of a set of condition codes that are stored in a special register called the machine status word. The contents of the machine status word include information about the last arithmetic operation performed. For example, one bit of this word indicates if the last result was zero. Another indicates if the last result was negative. Based on these condition codes, a number of conditional jumps can be performed. For example, the
jzinstruction performs a jump to the specified operand label if the result of the last arithmetic operation was zero. Otherwise, control proceeds to the next instruction in sequence.
A number of the conditional branches are given names that are intuitively based on the last operation performed being a special compare instruction,
cmp(see below). For example, conditional branches such as
jneare based on first performing a cmp operation on the desired operands.
je <label> (jump when equal) jne <label> (jump when not equal) jz <label> (jump when last result was zero) jg <label> (jump when greater than) jge <label> (jump when greater than or equal to) jl <label> (jump when less than) jle <label> (jump when less than or equal to)
cmp eax, ebx jle done ;If the contents of EAX are less than or equal to the contents of EBX, jump to the label done. Otherwise, continue to the next instruction.
cmp — Compare
Compare the values of the two specified operand, setting the condition codes in the machine status word appropriately. This instruction is equivalent to the
subinstruction, except the result of the substraction is discarded instead of replacing the first operand.
cmp <reg>,<reg> cmp <reg>,<mem> cmp <mem>,<reg> cmp <reg>,<con>
cmp DWORD PTR [var], 10 jeq loop ;If the 4 bytes stored at location var are equal to the 4-byte integer constant 10, jump to the location labeled loop.
call, ret — Subroutine call and return
These instructions implement a subroutine call and return. The
callinstruction first pushes the current code location onto the hardware supported stack in memory (see the
pushinstruction for details), and then performs an unconditional jump to the code location indicated by the label operand. Unlike the simple jump instructions, the
callinstruction saves the location to return to when the subroutine completes.
retinstruction implements a subroutine return mechanism. The instruction first pops a code location off the hardware supported in-memory stack (see the
popinstruction for details). It then performs an uncoditional jump to the retrieved code location.
call <label> ret
To allow separate programmers to share code and develop libraries for use by many programs, and to simplify the use of subroutines in general, programmers typically adopt a common calling convention. The calling convention is a protocl about how to call and return from routines. For example, given a set of calling convention rules, a programmer need not examine the definition of a subroutine to determine how paramters should be passed to that subroutine. Furthermore, given a set of calling convention rules, high-level language compilers can be made to follow the rules, thus allowing hand-coded assembly language routines and high-level language routines to call one another.
In practice, many calling conventions are possible. We will use the widely used C language calling convention. Following this convention will allow you to write assembly language subroutine that are safely callable from C (and C++) code, and will also enable you to call C library funcitons from your assembly language code.
The C calling convention is based heavily on the use of the hardware-supported stack. It is based on the
ret instructions. Subroutine parameters are passed on the stack. Registers are saved on the stack, and local variables used by the subroutines are placed in memory on the stack. The vast majority of high-level procedural languages implemented on most processors have used similar calling conventions.
The calling convention is broken into two sets of rules. The fist set of rules is employed by the caller of the subroutine, and the second set of rule is observed by the writer of the subroutine (the callee. It should be emphasized that mistakes in the observance of these rules quickly result in fatal program errors since the stack will be left in an incosistent state; thus metriculous care should be used when implementing the call convention in your own subroutines.
A good way to visualize the operation of the calling convention is to draw the contents of the nearby region of the stack during subroutine execution. The image above depicts the contents of the stack during the execution of a subroutine with three paramters and three local variables. The cells depicted in the stack are 32-bit wide memory locations, thus the memory addresses of the cells are 4 bytes apart. The first paramter resides at an offest of 8 bytes from the base pointer. Above the paramters on the stack (and below the base pointer), the
call instruction place the return address, thus leading to an extra 4 bytes of offset from the base pointer to the first parameter. When the
ret instruction is used to return from the subroutine, it will jump to the return address stored on the stack.
To make a subrouting call, the caller should:
- Before calling a subroutine, the caller should save the contents of certain registers that are designated caller-saved. The caller-saved register are EAX, ECX, EDX. Since the called subroutine is allowed to modify these registers, it the called relies on their values after the subroutine returns, the called must push the values in these registers onto the stack (so they can be restore after the subroutine returns).
- To pass paramters to the subroutine, push them onto the stack before the call. The paramters should be pushed in inverted order (i.e. last paramter first). Since the stack grows down, the first paramter will be stored at the lowest address (this inversion of paramter was historically used to allow function to be passed a variable number of paramters).
- To call the subroutine, use the
callinstruction. This instruction places the return address on top of the paramters on the stack, and branches to the subroutine code. This invokes the subroutine, which should follow the callee rules below.
After the subroutine returns (immediately following the
call instruction), the caller can expect to find the return value of the subroutine in the register EAX. To restore the machine state, the called should:
Remove the parameters from stack. This restores the stack to its state before the call was performed.
Restore the contents of caller-saved register (EAX, ECX, EDX) by poping them off of the stack. The caller assume that no other registers were modified by the subroutine.
The code below shows a function call that follows the caller rules. The caller is calling a function _myFunc that takes three paramters. First paramter is in EAX, the second paramter is the constant 216; the third paramter is in memory location var.
push [var] ; Push last parameter first push 216 ; Push the second parameter push eax ; Push first parameter last call _myFunc ; Call the function (assume C naming) add esp, 12
Note that after the call returns, the caller cleans up the stack using the
add instruction. We have 12 bytes (3 paramters * 4 bytes each) on the stack, and the stack grows down. Thus, to get rid of the paramters, we can simply add 12 to the stack pointer.
The result produced by _myFunc is now available for use in the register EAX. The values of the caller-saved register (ECX and EDX), may been changed. If the caller uses them after the call, it would have needed to save them on the stack before the call and restore them after it.
The definition of the subroutine should adhere to the following rules at the begining of the subroutine:
- Push the value of EBP onto the stack, the the copy the value of ESP into EBP using the folowing instruction:
push ebp mov ebp, esp
This initial action maintains the base pointer, EBP. The base pointer is used by convention as a point of reference for finding paramters and local variables on the stack. When a subroutine is executing, the base pointer holds a copy of the stack pointer vlaue from when subroutine started executing. Paramters and local variables will always be located at known, constant offsets aways from the base pointer value. We put the old base pointer value at the begining of the subroutine so that we can later restore the appropriate base pointer value for the caller when the subroutine returns. Remember, the caller is not expecting the subroutine to change the value of the base pointer. We then move the stack pointer into EBP to obtain our pointer of reference for accessing parameters and local variables.
Next, allocate local variables by making space on the stack. Recall, the stack grow down, so to make space on the top of the stack, the stack pointer should be decremented. The amount by which the stack pointer is decremented depends on the number and size of local variables needed. For example, if 3 local integers (4 bytes each) were required, the stack pointer would need to be decremented by 12 to make space for these local variables (i.e.
sub esp, 12). As with paramters, local variables will be located at known offests from the base pointer.
Next, save the value of the callee-saved registers that will be used by the function. To save registers, push them onto the stack. The callee-saved registers are EBX, EDI, and ESI (ESP and EBP will also be preserved by the calling convention, but need not be pused on the stack during this stemp).
After these three action are performed, the body of the subroutine may proceed. When the subroutine is returns, it must follow these steps:
Leave the return value in EAX.
Restore the old values of any callee-saved registers (EDI and ESI) that were modified. The register contents are restored by popping them from the stack. The registers should be popped in the inverse order that they were pushed.
Deallocate local variables. The obvious way to do this might be to add the appropriate value to the stack pointer (since the space was allocated by subtracting the needed amount from the stack poitner). In praticice, a less error-prone way to deallocate the variables is to move the value in the base pointer into the stack pointer:
mov esp, ebp. This works because the base pointer always contains the value that the stack pointer contained immediately prior to the allocation of the local variables.
Immediately before returing, restore the caller’s base ponter value by poping EBP off the stack. Recall that the first thing we did on entry to the subroutine was to push the base pointer to save its old value.
Finally, return to the caller by executing a
retinstruction. This instruction will find and remove the approxiate return address from the stack.
Note that the callee’s rules fall cleanly into two halves that are basically mirror images of one another. The first half of the rules apply to the begining of the function, and are commonly said to define the prologue to the function. The later half of the rules apply to the end of the function, and are thus commonly said to defined the epilogue of the function.
Here is an example function definition that follow the callee rules:
.486 .MODEL FLAT .CODE PUBLIC _myFunc _myFunc PROC ; Subroutine Prologue push ebp ; Save the old base pointer value. mov ebp, esp ; Set the new base pointer value. sub esp, 4 ; Make room for one 4-byte local variable. push edi ; Save the values of registers that the function push esi ; will modify. This function uses EDI and ESI. ; (no need to save EBX, EBP, or ESP) ; Subroutine Body mov eax, [ebp+8] ; Move value of parameter 1 into EAX mov esi, [ebp+12] ; Move value of parameter 2 into ESI mov edi, [ebp+16] ; Move value of parameter 3 into EDI mov [ebp-4], edi ; Move EDI into the local variable add [ebp-4], esi ; Add ESI into the local variable add eax, [ebp-4] ; Add the contents of the local variable ; into EAX (final result) ; Subroutine Epilogue pop esi ; Recover register values pop edi mov esp, ebp ; Deallocate local variables pop ebp ; Restore the caller's base pointer value ret _myFunc ENDP END
The subroutine prologue performs the standard actions of saving a snapshot of the stack pointer in EBP (the base pointer), allocating local variables by decremnting the stack pointer, and saving register values on the stack.
In the body of the subroutine we can see the use of the base pointer. Both paramters and local variables are located at constant offests from the base pointer for the duration of the subroutines execution. In particular, we notice that since paramters were placed onto the stack before the subroutine was called, they are always located below the base pointer (i.e. at higher addresses) on the stack. The first paramter to the subroutine can always be found at memory location [EBP+8], the second at [EBP+12], the third at [EBP+16]. Simillarly, since local variables are allocated after the base pointers is set, they always reside above the base pointer (i.e. at lower addresses) ont the stack. In particular, the first local variable is always located at [EBP-4], the second at [EBP-8], and so on. The conventional use of the base pointer allows us to quickly indentify the use of local variables and paramters within a function body.
The function epilogue is basically a mirror image of the function prologue. The caller’s register values are recovered from the stack, the local variables are deallocated by resetting the stack pointer, the caller’s bse pointer value is recovered, the the
ret instruction is used to return the appropriate code location in the caller.