// Signal tapset
// Copyright (C) 2006 IBM Corp.
// Copyright (C) 2006 Intel Corporation.
// Copyright (C) 2008-2009 Red Hat, Inc.
//
// This file is part of systemtap, and is free software.  You can
// redistribute it and/or modify it under the terms of the GNU General
// Public License (GPL); either version 2, or (at your option) any
// later version.
//
//
// <tapsetdescription>
//  This family of probe points is used to probe signal activities.
//  Since there are so many signals sent to processes at any given
//  point, it is advisable to filter the information according to the
//  requirements.  For example, filter only for a particular signal
//  (if sig==2) or for a particular process (if pid_name==stap).
// </tapsetdescription>

/**
 * probe signal.send - Signal being sent to a process
 * Arguments:
 * @sig: The number of the signal
 * @sig_name: A string representation of the signal
 * @sig_pid: The PID of the process receiving the signal
 * @pid_name: The name of the signal recipient
 * @si_code: Indicates the signal type
 * @task: A task handle to the signal recipient
 * @sinfo: The address of siginfo struct
 * @shared: Indicates whether the signal is shared by the thread group
 * @send2queue: Indicates whether the signal is sent to an existing
 * sigqueue
 * @name: The name of the function used to send out the signal
 *
 * Context:
 *  The signal's sender.
 *
 */
probe signal.send = _signal.send.*
{
    sig_name = _signal_name(sig)
    sig_pid = task_pid(task)
    pid_name = task_execname(task)

    if (sinfo == 2)
        si_code ="SIGSTOP or SIGKILL"
    else if (sinfo > 0)
        si_code="SI_KERNEL (SIGFPE, SIGSEGV, SIGTRAP, SIGCHLD, SIGPOLL)"
    else if (sinfo <= 0)
        si_code="SI_USER or SI_TIMER or SI_ASYNCIO"
}

%( kernel_v <= "2.6.25" %?
probe _signal.send.part1 = kernel.function("__group_send_sig_info")
{
    name = "__group_send_sig_info"
    sig = $sig
    task = $p
    sinfo = $info
    shared = 1
    send2queue = 0
}

probe _signal.send.part2 = kernel.function("send_group_sigqueue")
{
    name = "send_group_sigqueue"
    sig = $sig
    task = $p
    sinfo = 0 # $q->info
    shared = 1
    send2queue = 1
}

probe _signal.send.part4 = kernel.function("specific_send_sig_info")
{
    name = "specific_send_sig_info"
    sig = $sig
    task = $t
    sinfo = $info
    shared = 0
    send2queue = 0
}
%)

%( kernel_v > "2.6.25" %?
probe _signal.send.part1 =
		kernel.function("__send_signal") !,
		kernel.function("send_signal")
{
    if ($group == 1) {
	name = "__group_send_sig_info"
	shared = 1
    }
    else if ($group == 0) {
	    name = "specific_send_sig_info"
	    shared = 0
    }
    sig = $sig
    task = $t
    sinfo = $info
    send2queue = 0
}
%)

probe _signal.send.part3 = kernel.function("send_sigqueue")
{
    name = "send_sigqueue"
%( kernel_v >=  "2.6.26" %?
    task = $t
    sig = $q->info->si_signo
%:
    task = $p
    sig = $sig
%)
    sinfo = 0 # $q->info
    shared = 0
    send2queue = 1
}

/**
 * probe signal.send.return - Signal being sent to a process completed
 * @retstr: The return value to either __group_send_sig_info,
 * specific_send_sig_info,
 * or send_sigqueue
 * @shared: Indicates whether the sent signal is shared by the thread group.
 * @send2queue: Indicates whether the sent signal was sent to an
 * existing sigqueue
 * @name: The name of the function used to send out the signal
 *
 * Context:
 * The signal's sender. (correct?)
 *
 * Possible __group_send_sig_info and
 * specific_send_sig_info return values are as follows;
 *
 * 0 -- The signal is sucessfully sent to a process,
 * which means that:
 * (1) the signal was ignored by the receiving process,
 * (2) this is a non-RT signal and the system already has one queued, and
 * (3) the signal was successfully added to the sigqueue of the receiving process.
 *
 * -EAGAIN -- The sigqueue of the receiving process is
 * overflowing, the signal was RT, and the signal was sent by a user using something other
 * than kill().
 *
 * Possible send_group_sigqueue and
 * send_sigqueue return values are as follows;
 *
 * 0 -- The signal was either sucessfully added into the
 * sigqueue of the receiving process, or a SI_TIMER entry is already
 * queued (in which case, the overrun count will be simply incremented).
 *
 * 1 -- The signal was ignored by the receiving process.
 *
 * -1 -- (send_sigqueue only) The task was marked
 * exiting, allowing * posix_timer_event to redirect it to the group
 * leader.
 *
 */
probe signal.send.return = _signal.send.*.return
{
    retstr = returnstr(1)
}

// Return values for "__group_send_sig_info" and "specific_send_sig_info"
//
// - return 0 if  the signal is sucessfully sent to a process,
//   which means the following:
//     <1> the signal is ignored by receiving process
//     <2> this is a non-RT signal and we already have one queued
//     <3> the signal is successfully added into the sigqueue of
//        receiving process
//
// - return -EAGAIN if the sigqueue is overflow the signal was RT
//   and sent by user using something other than kill()

%( kernel_v > "2.6.25" %?
probe _signal.send.part1.return = kernel.function("send_signal").return
{
    if ($group == 1) {
	name = "__group_send_sig_info"
	shared = 1
    }
    else if ($group == 0) {
	    name = "specific_send_sig_info"
	    shared = 0
    }
    send2queue = 0
}
%)

%( kernel_v <= "2.6.25" %?
probe _signal.send.part1.return = kernel.function("__group_send_sig_info").return
{
    name = "__group_send_sig_info"
    shared = 1
    send2queue = 0
}

probe _signal.send.part4.return = kernel.function("specific_send_sig_info").return
{
    name = "specific_send_sig_info"
    shared = 0
    send2queue = 0
}

// - return 0 if the signal is either sucessfully added into the
//   sigqueue of receiving process or a SI_TIMER entry is already
//   queued so just increment the overrun count
//
// - return 1 if this signal is ignored by receiving process

probe _signal.send.part2.return = kernel.function("send_group_sigqueue").return
{
    name = "send_group_sigqueue"
    shared = 1
    send2queue = 1
}
%)

// - return 0 if the signal is either sucessfully added into the
//   sigqueue of receiving process or a SI_TIMER entry is already
//   queued so just increment the overrun count
//
// - return 1 if this signal is ignored by receiving process
//
// - return -1 if the task is marked exiting, so posix_timer_event
//   can redirect it to the group leader

probe _signal.send.part3.return = kernel.function("send_sigqueue").return
{
    name = "send_sigqueue"
    shared = 0
    send2queue = 1
}

/**
 * probe signal.checkperm - Check being performed on a sent signal
 * @sig: The number of the signal
 * @sig_name: A string representation of the signal
 * @sig_pid: The PID of the process receiving the signal
 * @pid_name: Name of the process receiving the signal
 * @si_code: Indicates the signal type
 * @task: A task handle to the signal recipient
 * @sinfo: The address of the siginfo structure
 * @name: Name of the probe point; default value is
 * signal.checkperm
 */
probe signal.checkperm = kernel.function("check_kill_permission")
{
    sig = $sig
    task = $t
    sinfo = $info
    name = "signal.checkperm"

    sig_name = _signal_name($sig)
    sig_pid = task_pid(task)
    pid_name = task_execname(task)

    if (sinfo == 2)
        si_code ="SIGSTOP or SIGKILL"
    else if (sinfo > 0)
        si_code="SI_KERNEL (SIGFPE, SIGSEGV, SIGTRAP, SIGCHLD, SIGPOLL)"
    else if (sinfo <= 0)
        si_code="SI_USER or SI_TIMER or SI_ASYNCIO"
}

/**
 * probe signal.checkperm.return - Check performed on a sent signal completed
 * @name: Name of the probe point; default value is
 * signal.checkperm
 * @retstr: Return value as a string
 */
probe signal.checkperm.return = kernel.function("check_kill_permission").return
{
    name = "signal.checkperm"
    retstr = returnstr(1)
}


/**
 * probe signal.wakeup - Sleeping process being wakened for signal
 * @sig_pid: The PID of the process to wake
 * @pid_name: Name of the process to wake
 * @resume: Indicates whether to wake up a task in a
 * STOPPED or TRACED state
 * @state_mask: A string representation indicating the mask
 * of task states to wake. Possible values are
 * TASK_INTERRUPTIBLE, TASK_STOPPED,
 * TASK_TRACED, and TASK_INTERRUPTIBLE.
 */
probe signal.wakeup = kernel.function("signal_wake_up")
{
    sig_pid = $t->pid
    pid_name = kernel_string($t->comm)
    resume   = $resume
    if (resume == 0) {
        state_mask = "TASK_INTERRUPTIBLE"
    } else {
        state_mask = "TASK_INTERRUPTIBLE | TASK_STOPPED | TASK_TRACED"
    }
}


/**
 * probe signal.check_ignored - Checking to see signal is ignored
 * @sig_pid: The PID of the process receiving the signal
 * @pid_name: Name of the process receiving the signal
 * @sig: The number of the signal
 * @sig_name: A string representation of the signal
 */
probe signal.check_ignored = kernel.function("sig_ignored")
{
    sig_pid = $t->pid
    pid_name = kernel_string($t->comm)
    sig = $sig
    sig_name = _signal_name($sig)
}

/**
 * probe signal.check_ignored.return - Check to see signal is ignored completed
 * @name: Name of the probe point; default value is
 * signal.checkperm
 * @retstr: Return value as a string
 */
probe signal.check_ignored.return = kernel.function("sig_ignored").return ?
{
    name = "sig_ignored"
    retstr = returnstr(1)
}


// probe signal.handle_stop
//  For now, just comment it out since at the time handle_stop_signal()
//  is called, it doesn't know whether current signal is STOP/COUNT.
//  So the calling of handle_stop_signal() doesn't mean that the Kernel
//  is now processing the STOP/COUNT signal
//
/*
probe signal.handle_stop = kernel.function("handle_stop_signal")
{
    sig_pid = $p->pid
    pid_name = kernel_string($p->comm)
    sig_info = $sig
    sig_name = _signal_name($sig)
}
*/


/**
 * probe signal.force_segv - Forcing send of SIGSEGV
 * @sig_pid: The PID of the process receiving the signal
 * @pid_name: Name of the process receiving the signal
 * @sig: The number of the signal
 * @sig_name: A string representation of the signal
 */
probe signal.force_segv = _signal.force_segv.*
{
}

probe _signal.force_segv.part1 = kernel.function("force_sigsegv")
{
    sig_pid = $p->pid
    pid_name = kernel_string($p->comm)
    sig = $sig
    sig_name = _signal_name($sig)
}

probe _signal.force_segv.part2 = kernel.function("force_sigsegv_info") ?
{
    sig_pid = pid()
    pid_name = execname()
    sig = $sig
    sig_name = _signal_name($sig)
}

/**
 * probe signal.force_segv.return - Forcing send of SIGSEGV complete
 * @name: Name of the probe point; default value is
 * force_sigsegv
 * @retstr: Return value as a string
 */
probe signal.force_segv.return =
		kernel.function("force_sigsegv").return,
		kernel.function("force_sigsegv_info").return ?
{
    name = "force_sigsegv"
    retstr = returnstr(1)
}


/**
 * probe signal.syskill - Sending kill signal to a process
 * @pid: The PID of the process receiving the signal
 * @sig: The specific signal sent to the process
 */
probe signal.syskill = syscall.kill
{
    sig_name = _signal_name($sig)
}

/**
 * probe signal.syskill.return - Sending kill signal completed
 */
probe signal.syskill.return = syscall.kill.return
{
}

/**
 * probe signal.sys_tkill - Sending a kill signal to a thread
 * @pid: The PID of the process receiving the kill signal
 * @sig: The specific signal sent to the process
 * @sig_name: The specific signal sent to the process
 *
 * The tkill call is analogous to kill(2),
 * except that it also allows a process within a specific thread group to
 * be targeted. Such processes are targeted through their unique
 * thread IDs (TID).
 */
probe signal.systkill = syscall.tkill
{
    sig_name = _signal_name($sig)
}

/**
 * probe signal.systkill.return - Sending kill signal to a thread completed
 */
probe signal.systkill.return = syscall.tkill.return
{
}

/**
 * probe signal.sys_tgkill - Sending kill signal to a thread group
 * @pid: The PID of the thread receiving the kill signal
 * @tgid: The thread group ID of the thread receiving the kill signal
 * @sig: The specific kill signal sent to the process
 * @sig_name: A string representation of the signal
 *
 * The tgkill call is similar to tkill,
 * except that it also allows the caller to specify the thread group ID of
 * the thread to be signalled. This protects against TID reuse.
 */
probe signal.systgkill = syscall.tgkill
{
    sig_name = _signal_name($sig)
}

/**
 * probe signal.sys_tgkill.return - Sending kill signal to a thread group completed
 */
probe signal.systgkill.return = syscall.tgkill.return
{
}

/**
 * probe signal.send_sig_queue - Queuing a signal to a process
 * @sig: The queued signal
 * @sig_name: A string representation of the signal
 * @sig_pid: The PID of the process to which the signal is queued
 * @pid_name: Name of the process to which the signal is queued
 * @sigqueue_addr: The address of the signal queue
 */
probe signal.send_sig_queue =
        kernel.function("send_sigqueue"),
        kernel.function("send_group_sigqueue") ?
{
    sig = $sig
    sig_name = _signal_name($sig)
    sig_pid = $p->pid
    pid_name = kernel_string($p->comm)
    sigqueue_addr = $q
}

/**
 * probe signal.send_sig_queue.return  - Queuing a signal to a process completed
 * @retstr: Return value as a string
 */
probe signal.send_sig_queue.return =
        kernel.function("send_sigqueue").return,
        kernel.function("send_group_sigqueue").return ?
{
    retstr = returnstr(1)
}


/**
 * probe signal.pending - Examining pending signal
 * @sigset_add: The address of the user-space signal set
 * (sigset_t)
 * @sigset_size: The size of the user-space signal set
 *
 * This probe is used to examine a set of signals pending for delivery
 * to a specific thread. This normally occurs when the
 * do_sigpending kernel function is executed.
 */
probe signal.pending = kernel.function("do_sigpending")
{
    sigset_add=$set
    sigset_size=$sigsetsize
}

/**
 * probe signal.pending.return - Examination of pending signal completed
 * @retstr: Return value as a string
 */
probe signal.pending.return = kernel.function("do_sigpending").return
{
    retstr = returnstr(1)
}


/**
 * probe signal.handle - Signal handler being invoked
 * @sig: The signal number that invoked the signal handler
 * @sinfo: The address of the siginfo table
 * @sig_code: The si_code value of the
 * siginfo signal
 * @ka_addr: The address of the k_sigaction table
 * associated with the signal
 * @oldset_addr: The address of the bitmask array of blocked signals
 * @regs: The address of the kernel-mode stack area
 * @sig_mode: Indicates whether the signal was a user-mode or kernel-mode signal
 */
probe signal.handle = kernel.function("handle_signal")
{
    sig = $sig
    sig_name = _signal_name($sig)
    sinfo = $info
    sig_code = $info->si_code
    ka_addr = $ka
    oldset_addr =$oldset
    regs = $regs

    // Check whether the signal is a User Mode or Kernel mode Signal.

    if (sinfo == 0 && sig_code <= 0)
        sig_mode = "User Mode Signal"
    else if (sinfo >=  1)
        sig_mode = "Kernel Mode Signal"
}

/**
 * probe signal.handle.return - Signal handler invocation completed
 * @retstr: Return value as a string
 */
probe signal.handle.return = kernel.function("handle_signal").return ?
{
    retstr = returnstr(1)
}


/**
 * probe signal.do_action - Examining or changing a signal action
 * @sig: The signal to be examined/changed
 * @sigact_addr: The address of the new sigaction
 * struct associated with the signal
 * @oldsigact_addr: The address of the old sigaction
 * struct associated with the signal
 * @sa_handler: The new handler of the signal
 * @sa_mask: The new mask of the signal
 */
probe signal.do_action = kernel.function("do_sigaction")
{
    sig = $sig
    sigact_addr = $act
    oldsigact_addr = $oact
    if(sigact_addr != 0)
    {
        sa_handler = $act->sa->sa_handler
        sa_mask = __get_action_mask($act)
    }
}

/**
 * probe signal.do_action.return - Examining or changing a signal action completed
 * @retstr: Return value as a string
 */
probe signal.do_action.return = kernel.function("do_sigaction").return
{
    retstr = returnstr(1)
}

function __get_action_mask:long(act:long) %{ /* pure */
    int i;
    struct k_sigaction *act = (struct k_sigaction *)((long)THIS->act);
    sigset_t *sigset = &act->sa.sa_mask;
    THIS->__retvalue = kread(&(sigset->sig[0]));
    for (i=1; i<_NSIG_WORDS; ++i) {
        uint64_t part = kread(&(sigset->sig[i]));
        THIS->__retvalue |= part << (_NSIG_BPW*i);
    }
    CATCH_DEREF_FAULT();
%}


/**
 * probe signal.procmask - Examining or changing blocked signals
 * @how: Indicates how to change the blocked signals; possible values are
 * SIG_BLOCK=0 (for blocking signals),
 * SIG_UNBLOCK=1 (for unblocking signals), and
 * SIG_SETMASK=2 for setting the signal mask.
 * @sigset_addr: The address of the signal set (sigset_t)
 * to be implemented
 * @oldsigset_addr: The old address of the signal set
 * (sigset_t)
 * @sigset: The actual value to be set for sigset_t
 * (correct?)
 */
probe signal.procmask = kernel.function("sigprocmask")
{
    how=$how
    sigset_addr = $set
    oldsigset_addr = $oldset
    sigset = get_sigset($set)
}

function get_sigset:long(sigset:long) %{ /* pure */
    int i;
    sigset_t *sigset = (sigset_t *)((long)THIS->sigset);
    THIS->__retvalue = kread(&(sigset->sig[0]));
    for (i=1; i<_NSIG_WORDS; ++i) {
        uint64_t part = kread(&(sigset->sig[i]));
        THIS->__retvalue |= part << (_NSIG_BPW*i);
    }
    CATCH_DEREF_FAULT();
%}

probe signal.procmask.return = kernel.function("sigprocmask").return
{
    retstr = returnstr(1)
}


/**
 * probe signal.flush - Flushing all pending signals for a task
 * @task: The task handler of the process performing the flush
 * @sig_pid: The PID of the process associated with the task
 * performing the flush
 * @pid_name: The name of the process associated with the task
 * performing the flush
 */
probe signal.flush = kernel.function("flush_signals")
{
    task = $t
    sig_pid = $t->pid
    pid_name = kernel_string($t->comm)
}

function get_sa_flags:long (act:long)
{
	return @cast(act, "k_sigaction", "kernel")->sa->sa_flags
}

function get_sa_handler:long (act:long)
{
	return @cast(act, "k_sigaction", "kernel")->sa->sa_handler
}

// sa_mask contains the set of signals to be blocked when executing the
// signal handler. This function returns a string, delimited by ",".
//
// struct task_struct {
// [...]
//         struct signal_struct//signal;
//         struct sighand_struct//sighand;
// [...]
// struct sighand_struct {
//         atomic_t                count;
//         struct k_sigaction      action[_NSIG];
// [...]
// struct k_sigaction {
//         struct sigaction sa;
// };
//
// struct sigaction {
// [...]
//         sigset_t sa_mask;
// };

function sigset_mask_str:string (mask:long) %{ /* pure */
	int i, len;
	char *str = THIS->__retvalue, tmp[5];
	for (i = 1; i < _NSIG; ++i, THIS->mask >>=1)
		if (THIS->mask & 1) {
			sprintf(tmp, "%u,", i);
			strlcat(str, tmp, MAXSTRINGLEN);
		}
	len = strlen(str);
	if (len) str[len - 1] = '\0';
%}

// task_struct->blocked signal mask contains the set of signals that are
// currently blocked.
//
// struct task_struct {
// [...]
//         sigset_t blocked, real_blocked;

function is_sig_blocked:long (task:long, sig:long) %{ /* pure */
	int i;
	sigset_t blocked;
	struct task_struct *p = (struct task_struct *)((long)THIS->task);
	for (i = 0; i < _NSIG_WORDS; ++i)
		blocked.sig[i] = kread(&p->blocked.sig[i]);
	THIS->__retvalue = sigismember(&blocked, THIS->sig);
	CATCH_DEREF_FAULT();
%}

function sa_flags_str:string (sa_flags:long) %{ /* pure */
	int len;
	char *str = THIS->__retvalue;
	if (THIS->sa_flags & 0x00000001u) strlcat(str, "NOCLDSTOP|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x00000002u) strlcat(str, "NOCLDWAIT|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x00000004u) strlcat(str, "SIGINFO|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x08000000u) strlcat(str, "ONSTACK|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x10000000u) strlcat(str, "RESTART|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x40000000u) strlcat(str, "NODEFER|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x80000000u) strlcat(str, "RESETHAND|", MAXSTRINGLEN);
	if (THIS->sa_flags & 0x04000000) strlcat(str, "RESTORER|", MAXSTRINGLEN);
	len = strlen(str);
	if (len) str[len - 1] = '\0';
%}

function sa_handler_str(handler) {
	if (handler == 0) return "default"; /* SIG_DFL */
	if (handler == 1) return "ignored"; /* SIG_IGN */
	if (handler == -1) return "error"; /* SIG_ERR */
	return sprintf("%p", handler); /* userspace address */
}

// Signals start from 1 not 0.

global __sig[64]

function signal_str(num) {
    if (!(64 in __sig)) {
	__sig[1] = "HUP"
	__sig[2] = "INT"
	__sig[3] = "QUIT"
	__sig[4] = "ILL"
	__sig[5] = "TRAP"
	__sig[6] = "ABRT" /* or IOT */
	__sig[7] = "BUS"
	__sig[8] = "FPE"
	__sig[9] = "KILL"
	__sig[10] = "USR1"
	__sig[11] = "SEGV"
	__sig[12] = "USR2"
	__sig[13] = "PIPE"
	__sig[14] = "ALRM"
	__sig[15] = "TERM"
	__sig[16] = "STKFLT"
	__sig[17] = "CHLD" /* or CLD */
	__sig[18] = "CONT"
	__sig[19] = "STOP"
	__sig[20] = "TSTP"
	__sig[21] = "TTIN"
	__sig[22] = "TTOU"
	__sig[23] = "URG"
	__sig[24] = "XCPU"
	__sig[25] = "XFSZ"
	__sig[26] = "VTALRM"
	__sig[27] = "PROF"
	__sig[28] = "WINCH"
	__sig[29] = "IO/POLL"
	__sig[30] = "PWR"
	__sig[31] = "SYS" /* or UNUSED */
	__sig[32] = "RTMIN"
	__sig[33] = "RTMIN+1"
	__sig[34] = "RTMIN+2"
	__sig[35] = "RTMIN+3"
	__sig[36] = "RTMIN+4"
	__sig[37] = "RTMIN+5"
	__sig[38] = "RTMIN+6"
	__sig[39] = "RTMIN+7"
	__sig[40] = "RTMIN+8"
	__sig[41] = "RTMIN+9"
	__sig[42] = "RTMIN+10"
	__sig[43] = "RTMIN+11"
	__sig[44] = "RTMIN+12"
	__sig[45] = "RTMIN+13"
	__sig[46] = "RTMIN+14"
	__sig[47] = "RTMIN+15"
	__sig[48] = "RTMIN+16"
	__sig[49] = "RTMIN+17"
	__sig[50] = "RTMIN+18"
	__sig[51] = "RTMIN+19"
	__sig[52] = "RTMIN+20"
	__sig[53] = "RTMIN+21"
	__sig[54] = "RTMIN+22"
	__sig[55] = "RTMIN+23"
	__sig[56] = "RTMIN+24"
	__sig[57] = "RTMIN+25"
	__sig[58] = "RTMIN+26"
	__sig[59] = "RTMIN+27"
	__sig[60] = "RTMIN+28"
	__sig[61] = "RTMIN+29"
	__sig[62] = "RTMIN+30"
	__sig[63] = "RTMIN+31"
	__sig[64] = "RTMIN+32"
    }
    return __sig[num]
}