What is the interface for ARM system calls and where is it defined in the Linux kernel?

Asked 18/10, 2012 at 4:9 Answered 16/5, 2024 at 23:35

Solved linux linux-kernel arm system-calls

I have read about system calls in Linux, and everywhere description is given regarding x86 architecture (0x80 interrupt and SYSENTER). But I am not able to track down the files and process for a system call in ARM architecture. Can anyone please help.

Few relevant files which I got to know are:

arch/arm/kernel/calls.S

arch/arm/kernel/entry-common.S (explanation needed)

Erfurt answered 18/10, 2012 at 4:9 Comment(5)

Is it how user space makes a system call that you are wondering or how it is handled in kernel space? – Medea 18/10, 2012 at 6:47

@Medea how it is handles in kernel space. and as far as user space is concerned, i dont think it will be arm specific. thanks – Erfurt 18/10, 2012 at 7:20

thinking about the answer you accepted below, I wonder what you've seen there kernel specific? – Medea 18/10, 2012 at 7:32

i am reading and analysing how the kernel issues software interrupt in case of ARM to switch to system mode and how library calls are linked with system calls. It seems that swi is converted to svc. For further clarification i need to run a program using/calling a basic system call. Kindly correct me if i am wrong or if i am missing something. – Erfurt 18/10, 2012 at 7:50

It is the user space which issues a software interrupt (which looks like the part you are interested), then kernel takes over. The way you do it in arm is "supervisor call / svc" (previously called swi). see this boilerplate code from uclibc git.uclibc.org/uClibc/tree/libc/sysdeps/linux/arm/syscall.c – Medea 18/10, 2012 at 8:47

In ARM world, you do a software interrupt (mechanism to signal the kernel) by supervisor call / svc (previously called SWI).

ARM assembly (UAL) syntax looks like this:

SVC{<c>}{<q>} {#}<imm>

(In Linux you need to pass #0)

You should cheat from other projects like bionic or uClibc.

Medea answered 18/10, 2012 at 9:35 Comment(2)

thanks for the link, it is clear from that the arguments are loaded in the registers. Linux allows 5 arguments as mentioned, so there must be loading somewhere in General purpose registers in linux code(for arm). I am not able to find the corresponding file in linux code. – Erfurt 18/10, 2012 at 9:58

you can pass them as in uClibc example. – Medea 18/10, 2012 at 11:14

More generic answer than what you asked.

On Linux the man syscall (2) is a good start to find out how to make a system call in various architectures.

Copied from that manpage:

Architecture calling conventions
    Every architecture has its own way of invoking and passing arguments
    to the kernel.  The details for various architectures are listed in
    the two tables below.

    The first table lists the instruction used to transition to kernel
    mode (which might not be the fastest or best way to transition to the
    kernel, so you might have to refer to vdso(7)), the register used to
    indicate the system call number, the register used to return the sys‐
    tem call result, and the register used to signal an error.

    arch/ABI    instruction           syscall #  retval  error    Notes
    ────────────────────────────────────────────────────────────────────
    alpha       callsys               v0         a0      a3       [1]
    arc         trap0                 r8         r0      -
    arm/OABI    swi NR                -          a1      -        [2]
    arm/EABI    swi 0x0               r7         r0      -
    arm64       svc #0                x8         x0      -
    blackfin    excpt 0x0             P0         R0      -
    i386        int $0x80             eax        eax     -
    ia64        break 0x100000        r15        r8      r10      [1]
    m68k        trap #0               d0         d0      -
    microblaze  brki r14,8            r12        r3      -
    mips        syscall               v0         v0      a3       [1]
    nios2       trap                  r2         r2      r7
    parisc      ble 0x100(%sr2, %r0)  r20        r28     -
    powerpc     sc                    r0         r3      r0       [1]
    riscv       scall                 a7         a0      -
    s390        svc 0                 r1         r2      -        [3]
    s390x       svc 0                 r1         r2      -        [3]
    superh      trap #0x17            r3         r0      -        [4]
    sparc/32    t 0x10                g1         o0      psr/csr  [1]
    sparc/64    t 0x6d                g1         o0      psr/csr  [1]
    tile        swint1                R10        R00     R01      [1]
    x86-64      syscall               rax        rax     -        [5]
    x32         syscall               rax        rax     -        [5]
    xtensa      syscall               a2         a2      -

    Notes:

        [1] On a few architectures, a register is used as a boolean (0
            indicating no error, and -1 indicating an error) to signal
            that the system call failed.  The actual error value is still
            contained in the return register.  On sparc, the carry bit
            (csr) in the processor status register (psr) is used instead
            of a full register.

        [2] NR is the system call number.

        [3] For s390 and s390x, NR (the system call number) may be passed
            directly with svc NR if it is less than 256.

        [4] On SuperH, the trap number controls the maximum number of
            arguments passed.  A trap #0x10 can be used with only 0-argu‐
            ment system calls, a trap #0x11 can be used with 0- or
            1-argument system calls, and so on up to trap #0x17 for
            7-argument system calls.

        [5] The x32 ABI uses the same instruction as the x86-64 ABI and
            is used on the same processors.  To differentiate between
            them, the bit mask __X32_SYSCALL_BIT is bitwise-ORed into the
            system call number for system calls under the x32 ABI.  Both
            system call tables are available though, so setting the bit
            is not a hard requirement.

    The second table shows the registers used to pass the system call
    arguments.

    arch/ABI      arg1  arg2  arg3  arg4  arg5  arg6  arg7  Notes
    ──────────────────────────────────────────────────────────────
    alpha         a0    a1    a2    a3    a4    a5    -
    arc           r0    r1    r2    r3    r4    r5    -
    arm/OABI      a1    a2    a3    a4    v1    v2    v3
    arm/EABI      r0    r1    r2    r3    r4    r5    r6
    arm64         x0    x1    x2    x3    x4    x5    -
    blackfin      R0    R1    R2    R3    R4    R5    -
    i386          ebx   ecx   edx   esi   edi   ebp   -
    ia64          out0  out1  out2  out3  out4  out5  -
    m68k          d1    d2    d3    d4    d5    a0    -
    microblaze    r5    r6    r7    r8    r9    r10   -
    mips/o32      a0    a1    a2    a3    -     -     -     [1]
    mips/n32,64   a0    a1    a2    a3    a4    a5    -
    nios2         r4    r5    r6    r7    r8    r9    -
    parisc        r26   r25   r24   r23   r22   r21   -
    powerpc       r3    r4    r5    r6    r7    r8    r9
    riscv         a0    a1    a2    a3    a4    a5    -
    s390          r2    r3    r4    r5    r6    r7    -
    s390x         r2    r3    r4    r5    r6    r7    -
    superh        r4    r5    r6    r7    r0    r1    r2
    sparc/32      o0    o1    o2    o3    o4    o5    -
    sparc/64      o0    o1    o2    o3    o4    o5    -
    tile          R00   R01   R02   R03   R04   R05   -
    x86-64        rdi   rsi   rdx   r10   r8    r9    -
    x32           rdi   rsi   rdx   r10   r8    r9    -
    xtensa        a6    a3    a4    a5    a8    a9    -

    Notes:

        [1] The mips/o32 system call convention passes arguments 5
            through 8 on the user stack.

    Note that these tables don't cover the entire calling convention—some
    architectures may indiscriminately clobber other registers not listed
    here.

So it depends whether the system uses OABI or EABI.

So in EABI you use r7 to pass the system call number, use r0-r6 to pass the arguments, use SWI 0 to make the system call, expect the result in r0.

In OABI everything is the same except you use SWI <number> to make a system call.

Chondrite answered 11/8, 2016 at 20:56 Comment(4)

Any idea what the # after SWI represents? Is is the address of the service table? – Valonia 10/7, 2017 at 12:16

I did find "immed_8 is ignored by the processor. However, it is present in bits[7:0] of the instruction opcode. It can be retrieved by the exception handler to determine what service is being requested." here, a while back but still never understood what was really happenening. For example, why would it want to know what service was being requested? – Valonia 10/7, 2017 at 12:26

This table misses the sysenter mechanism used on i686 in addition to int 0x80. – Bentham 15/8, 2018 at 21:32

@Valonia On arm the number after SWI doesn't do anything from a hardware point of view. Traditionally the kernel would use the user-space instruction pointer to read the system call number from the instruction and dispatch the call. When EABI was designed this was deemed to be inefficient. On EABI the number after the SVC is always 0. On pure EABI systems it will be ignored, on EABI with OABI compatibility enabled the number after SVC will be used to distinguish between OABI and EABI syscalls. – Insincere 13/12, 2018 at 19:25

In ARM world, you do a software interrupt (mechanism to signal the kernel) by supervisor call / svc (previously called SWI).

ARM assembly (UAL) syntax looks like this:

SVC{<c>}{<q>} {#}<imm>

(In Linux you need to pass #0)

You should cheat from other projects like bionic or uClibc.

Medea answered 18/10, 2012 at 9:35 Comment(2)

you can pass them as in uClibc example. – Medea 18/10, 2012 at 11:14

If you're looking for syscall number in Linux system, take a look at w3challs.

That site helps finding out which syscall number and which registers used for passing arguments. It supports many architectures listed below:

Edit: Filed https://w3challs.com/forum/general/dev/bugs#04 to track the incorrect bug.

Ultra answered 14/8, 2018 at 9:6 Comment(2)

Many of these are just plain wrong, sadly --- look at pread64, which shows both buf and count in r1. – Nealey 20/1, 2020 at 22:56

It is. Maybe some bugs exist in the program used ot generating those tables. – Ultra 21/1, 2020 at 22:2

The disassembly of hello world in n900 shows svc #0

http://brnz.org/hbr/?m=201102

Humoresque answered 18/10, 2012 at 4:21 Comment(0)

arm system call numbers

v4.19 defines them at arch/arm/tools/syscall.tbl: https://github.com/torvalds/linux/blob/v4.19/arch/arm/tools/syscall.tbl

Excerpt:

0   common  restart_syscall     sys_restart_syscall
1   common  exit            sys_exit
2   common  fork            sys_fork
3   common  read            sys_read
4   common  write           sys_write
5   common  open            sys_open
6   common  close           sys_close

I have tested that exit and write have those exact numbers on this Linux assembly hello world on QEMU user mode.

arm64 system call numbers

aarch64 uses a new mechanism, see: https://reverseengineering.stackexchange.com/questions/16917/arm64-syscalls-table/18834#18834

Flavory answered 20/11, 2018 at 16:59 Comment(0)

The Linux system call numbers are defined in the C include file: /usr/include/asm-generic/unistd.h

In this file, there are define statements such as the following:

/* fs/read_write.c */
#define __NR_write 64

/* kernel/exit.c */
#define __NR_exit 93

This defines the symbol __NR_write to represent the magic number 64 for the write Linux system call and 93 for exit.

Excerpt from Programming with 64-Bit ARM Assembly Language: Single Board Computer Development for Raspberry Pi and Mobile Devices | Chapter 7: Linux Operating System Services - Calling Convention - Linux System Call Numbers

Valora answered 16/5, 2024 at 23:35 Comment(0)

Hot tags

Godot Unity Godot Help Programming Godot 4.X GUI GDScript 3D 2D Physics CSharp Godot 3.X VR XR Projects C++

Recommended topics

Hot tags