nuttx/sched/sched/sched_waitpid.c
hujun5 8a6aa5f5bf sched/sched: There is no need to use sched_[un]lock
purpose:
1 sched_lock is very time-consuming, and reducing its invocations can improve performance.
2 sched_lock is prone to misuse, and narrowing its scope of use is to prevent people from referencing incorrect code and using it

test:
We can use qemu for testing.
compiling
make distclean -j20; ./tools/configure.sh -l qemu-armv8a:nsh_smp ;make -j20
running
qemu-system-aarch64 -cpu cortex-a53 -smp 4 -nographic -machine virt,virtualization=on,gic-version=3 -net none -chardev stdio,id=con,mux=on -serial chardev:con -mon chardev=con,mode=readline -kernel ./nuttx

We have also tested this patch on other ARM hardware platforms.

Signed-off-by: hujun5 <hujun5@xiaomi.com>
2024-04-30 11:34:24 -03:00

628 lines
20 KiB
C

/****************************************************************************
* sched/sched/sched_waitpid.c
*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership. The
* ASF licenses this file to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance with the
* License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
* License for the specific language governing permissions and limitations
* under the License.
*
****************************************************************************/
/****************************************************************************
* Included Files
****************************************************************************/
#include <nuttx/config.h>
#include <sys/wait.h>
#include <signal.h>
#include <assert.h>
#include <errno.h>
#include <nuttx/sched.h>
#include <nuttx/signal.h>
#include <nuttx/cancelpt.h>
#include <nuttx/semaphore.h>
#include "sched/sched.h"
#include "group/group.h"
#ifdef CONFIG_SCHED_WAITPID
/****************************************************************************
* Public Functions
****************************************************************************/
/****************************************************************************
* Name: nxsched_waitpid
*
* Description:
* This functions will obtain status information pertaining to one
* of the caller's child processes. This function will suspend
* execution of the calling thread until status information for one of the
* terminated child processes of the calling process is available, or until
* delivery of a signal whose action is either to execute a signal-catching
* function or to terminate the process. If more than one thread is
* suspended in nxsched_waitpid() awaiting termination of the same process,
* exactly one thread will return the process status at the time of the
* target process termination. If status information is available prior to
* the call to nxsched_waitpid(), return will be immediate.
*
* Input Parameters:
* pid - The task ID of the thread to waid for
* stat_loc - The location to return the exit status
* options - Modifiable behavior, see sys/wait.h.
*
* Returned Value:
* If nxsched_waitpid() returns because the status of a child process is
* available, it will return a value equal to the process ID of the child
* process for which status is reported.
*
* If nxsched_waitpid() returns due to the delivery of a signal to the
* calling process, -1 will be returned and errno set to EINTR.
*
* If nxsched_waitpid() was invoked with WNOHANG set in options, it has
* at least one child process specified by pid for which status is not
* available, and status is not available for any process specified by
* pid, 0 is returned.
*
* Otherwise, (pid_t)-1 will be returned, and errno set to indicate the
* error:
*
* ECHILD - The process specified by pid does not exist or is not a child
* of the calling process, or the process group specified by pid
* does not exist does not have any member process that is a child
* of the calling process.
* EINTR - The function was interrupted by a signal. The value of the
* location pointed to by stat_loc is undefined.
* EINVAL - The options argument is not valid.
*
****************************************************************************/
#ifndef CONFIG_SCHED_HAVE_PARENT
pid_t nxsched_waitpid(pid_t pid, int *stat_loc, int options)
{
FAR struct tcb_s *ctcb;
FAR struct task_group_s *group;
bool mystat = false;
int ret;
/* NOTE: sched_lock() is not enough for SMP
* because the child task is running on another CPU
*/
irqstate_t flags = enter_critical_section();
/* Get the TCB corresponding to this PID */
ctcb = nxsched_get_tcb(pid);
if (ctcb == NULL)
{
ret = -ECHILD;
goto errout;
}
/* Then the task group corresponding to this PID */
group = ctcb->group;
if (group == NULL)
{
ret = -ECHILD;
goto errout;
}
/* Lock this group so that it cannot be deleted until the wait completes */
group_add_waiter(group);
/* "If more than one thread is suspended in waitpid() awaiting termination
* of the same process, exactly one thread will return the process status
* at the time of the target process termination." Hmmm.. what do we
* return to the others?
*/
if (group->tg_waitflags == 0)
{
/* Save the waitpid() options, setting the non-standard WCLAIMED bit to
* assure that tg_waitflags is non-zero.
*/
group->tg_waitflags = (uint8_t)options | WCLAIMED;
/* Save the return status location (which may be NULL) */
group->tg_statloc = stat_loc;
/* We are the waipid() instance that gets the return status */
mystat = true;
}
/* Then wait for the task to exit */
if ((options & WNOHANG) != 0)
{
/* Don't wait if status is not available */
ret = nxsem_trywait(&group->tg_exitsem);
}
else
{
/* Wait if necessary for status to become available */
ret = nxsem_wait(&group->tg_exitsem);
}
group_del_waiter(group);
if (ret < 0)
{
/* Handle the awkward case of whether or not we
* need to nullify the stat_loc value.
*/
if (mystat)
{
group->tg_statloc = NULL;
group->tg_waitflags = 0;
}
if ((options & WNOHANG) != 0)
{
pid = 0;
}
else
{
goto errout;
}
}
/* On success, return the PID */
ret = pid;
errout:
leave_critical_section(flags);
return ret;
}
/****************************************************************************
*
* If CONFIG_SCHED_HAVE_PARENT is defined, then waitpid will use the SIGCHLD
* signal. It can also handle the pid == INVALID_PROCESS_ID argument. This
* is slightly more spec-compliant.
*
* But then I have to be concerned about the fact that NuttX does not queue
* signals. This means that a flurry of signals can cause signals to be
* lost (or to have the data in the struct siginfo to be overwritten by
* the next signal).
*
****************************************************************************/
#else
pid_t nxsched_waitpid(pid_t pid, int *stat_loc, int options)
{
FAR struct tcb_s *rtcb = this_task();
FAR struct tcb_s *ctcb;
#ifdef CONFIG_SCHED_CHILD_STATUS
FAR struct child_status_s *child = NULL;
bool retains;
#endif
FAR struct siginfo info;
irqstate_t flags;
sigset_t set;
int ret;
/* Create a signal set that contains only SIGCHLD */
sigemptyset(&set);
nxsig_addset(&set, SIGCHLD);
flags = enter_critical_section();
/* Verify that this task actually has children and that the requested PID
* is actually a child of this task.
*/
#ifdef CONFIG_SCHED_CHILD_STATUS
/* Does this task retain child status? */
retains = ((rtcb->group->tg_flags & GROUP_FLAG_NOCLDWAIT) == 0);
if (rtcb->group->tg_children == NULL && retains)
{
ret = -ECHILD;
goto errout;
}
else if (pid != INVALID_PROCESS_ID)
{
/* Get the TCB corresponding to this PID. NOTE: If the child has
* already exited, then the PID will not map to a valid TCB.
*/
ctcb = nxsched_get_tcb(pid);
if (ctcb && ctcb->group)
{
/* Make sure that the thread it is our child. */
if (ctcb->group->tg_ppid != rtcb->group->tg_pid)
{
ret = -ECHILD;
goto errout;
}
}
/* The child task is ours or it is no longer active. Does the parent
* task retain child status?
*/
if (retains)
{
/* Yes.. Check if this specific pid has allocated child status? */
if (group_find_child(rtcb->group, pid) == NULL)
{
ret = -ECHILD;
goto errout;
}
}
}
#else /* CONFIG_SCHED_CHILD_STATUS */
if (rtcb->group->tg_nchildren == 0)
{
/* There are no children */
ret = -ECHILD;
goto errout;
}
else if (pid != INVALID_PROCESS_ID)
{
/* Get the TCB corresponding to this PID and make sure that the
* thread it is our child.
*/
ctcb = nxsched_get_tcb(pid);
if (!ctcb || !ctcb->group || ctcb->group->tg_ppid != rtcb->pid)
{
ret = -ECHILD;
goto errout;
}
}
#endif /* CONFIG_SCHED_CHILD_STATUS */
/* Loop until the child that we are waiting for dies */
for (; ; )
{
#ifdef CONFIG_SCHED_CHILD_STATUS
/* Check if the task has already died. Signals are not queued in
* NuttX. So a possibility is that the child has died and we
* missed the death of child signal (we got some other signal
* instead).
*/
if (pid == INVALID_PROCESS_ID)
{
/* We are waiting for any child, check if there are still
* children.
*/
DEBUGASSERT(!retains || rtcb->group->tg_children);
if (retains && (child = group_exit_child(rtcb->group)) != NULL)
{
/* A child has exited. Apparently we missed the signal.
* Return the saved exit status.
*/
/* The child has exited. Return the saved exit status */
if (stat_loc != NULL)
{
*stat_loc = child->ch_status << 8;
}
pid = child->ch_pid;
/* Discard the child entry and break out of the loop */
group_remove_child(rtcb->group, child->ch_pid);
group_free_child(child);
break;
}
}
/* We are waiting for a specific PID. Does this task retain child
* status?
*/
else if (retains)
{
/* Get the current status of the child task. */
child = group_find_child(rtcb->group, pid);
DEBUGASSERT(child);
/* Did the child exit? */
if ((child->ch_flags & CHILD_FLAG_EXITED) != 0)
{
/* The child has exited. Return the saved exit status */
if (stat_loc != NULL)
{
*stat_loc = child->ch_status << 8;
}
/* Discard the child entry and break out of the loop */
if ((options & WNOWAIT) == 0)
{
group_remove_child(rtcb->group, pid);
group_free_child(child);
}
break;
}
}
else
{
/* We can use nxsig_kill() with signal number 0 to determine if
* that task is still alive.
*/
ret = nxsig_kill(pid, 0);
if (ret < 0)
{
/* It is no longer running. We know that the child task
* was running okay when we started, so we must have lost
* the signal. In this case, we know that the task exit'ed,
* but we do not know its exit status. It would be better
* to reported ECHILD than bogus status.
*/
ret = -ECHILD;
goto errout;
}
}
#else /* CONFIG_SCHED_CHILD_STATUS */
/* Check if the task has already died. Signals are not queued in
* NuttX. So a possibility is that the child has died and we
* missed the death of child signal (we got some other signal
* instead).
*/
if (rtcb->group->tg_nchildren == 0 ||
(pid != INVALID_PROCESS_ID && nxsig_kill(pid, 0) < 0))
{
/* We know that the child task was running okay when we started,
* so we must have lost the signal. What can we do?
* Let's return ECHILD.. that is at least informative.
*/
ret = -ECHILD;
goto errout;
}
#endif /* CONFIG_SCHED_CHILD_STATUS */
if ((options & WNOHANG) != 0)
{
pid = 0;
break;
}
/* Wait for any death-of-child signal */
ret = nxsig_waitinfo(&set, &info);
if (ret < 0)
{
goto errout;
}
/* Was this the death of the thread we were waiting for? In the of
* pid == INVALID_PROCESS_ID, we are waiting for any child thread.
*/
if (info.si_signo == SIGCHLD &&
(pid == INVALID_PROCESS_ID || info.si_pid == pid))
{
/* Yes... return the status and PID (in the event it was -1) */
if (stat_loc != NULL)
{
*stat_loc = info.si_status << 8;
}
pid = info.si_pid;
#ifdef CONFIG_SCHED_CHILD_STATUS
if (retains)
{
/* Recover the exiting child */
child = group_find_child(rtcb->group, info.si_pid);
/* Discard the child entry, if we have one */
if (child != NULL && (options & WNOWAIT) == 0)
{
group_remove_child(rtcb->group, child->ch_pid);
group_free_child(child);
}
}
#endif /* CONFIG_SCHED_CHILD_STATUS */
break;
}
}
/* On success, return the PID */
ret = pid;
errout:
leave_critical_section(flags);
return ret;
}
#endif /* CONFIG_SCHED_HAVE_PARENT */
/****************************************************************************
* Name: waitpid
*
* Description:
* The waitpid() functions will obtain status information pertaining to one
* of the caller's child processes. The waitpid() function will suspend
* execution of the calling thread until status information for one of the
* terminated child processes of the calling process is available, or until
* delivery of a signal whose action is either to execute a signal-catching
* function or to terminate the process. If more than one thread is
* suspended in waitpid() awaiting termination of the same process, exactly
* one thread will return the process status at the time of the target
* process termination. If status information is available prior to the
* call to waitpid(), return will be immediate.
*
* The pid argument specifies a set of child processes for which status is
* requested. The waitpid() function will only return the status of a child
* process from this set:
*
* - If pid is equal to (pid_t)-1, status is requested for any child
* process. In this respect, waitpid() is then equivalent to wait().
* - If pid is greater than 0, it specifies the process ID of a single
* child process for which status is requested.
* - If pid is 0, status is requested for any child process whose process
* group ID is equal to that of the calling process.
* - If pid is less than (pid_t)-1, status is requested for any child
* process whose process group ID is equal to the absolute value of pid.
*
* The options argument is constructed from the bitwise-inclusive OR of
* zero or more of the following flags, defined in the <sys/wait.h> header:
*
* WCONTINUED - The waitpid() function will report the status of any
* continued child process specified by pid whose status has not been
* reported since it continued from a job control stop.
* WNOHANG - The waitpid() function will not suspend execution of the
* calling thread if status is not immediately available for one of the
* child processes specified by pid.
* WUNTRACED - The status of any child processes specified by pid that are
* stopped, and whose status has not yet been reported since they stopped,
* will also be reported to the requesting process.
*
* If the calling process has SA_NOCLDWAIT set or has SIGCHLD set to
* SIG_IGN, and the process has no unwaited-for children that were
* transformed into zombie processes, the calling thread will block until
* all of the children of the process containing the calling thread
* terminate, and waitpid() will fail and set errno to ECHILD.
*
* If waitpid() returns because the status of a child process is available,
* these functions will return a value equal to the process ID of the child
* process. In this case, if the value of the argument stat_loc is not a
* null pointer, information will be stored in the location pointed to by
* stat_loc. The value stored at the location pointed to by stat_loc will
* be 0 if and only if the status returned is from a terminated child
* process that terminated by one of the following means:
*
* 1. The process returned 0 from main().
* 2. The process called _exit() or exit() with a status argument of 0.
* 3. The process was terminated because the last thread in the process
* terminated.
*
* Regardless of its value, this information may be interpreted using the
* following macros, which are defined in <sys/wait.h> and evaluate to
* integral expressions; the stat_val argument is the integer value pointed
* to by stat_loc.
*
* WIFEXITED(stat_val) - Evaluates to a non-zero value if status was
* returned for a child process that terminated normally.
* WEXITSTATUS(stat_val) - If the value of WIFEXITED(stat_val) is non-zero,
* this macro evaluates to the low-order 8 bits of the status argument
* that the child process passed to _exit() or exit(), or the value the
* child process returned from main().
* WIFSIGNALED(stat_val) - Evaluates to a non-zero value if status was
* returned for a child process that terminated due to the receipt of a
* signal that was not caught (see <signal.h>).
* WTERMSIG(stat_val) - If the value of WIFSIGNALED(stat_val) is non-zero,
* this macro evaluates to the number of the signal that caused the
* termination of the child process.
* WIFSTOPPED(stat_val) - Evaluates to a non-zero value if status was
* returned for a child process that is currently stopped.
* WSTOPSIG(stat_val) - If the value of WIFSTOPPED(stat_val) is non-zero,
* this macro evaluates to the number of the signal that caused the child
* process to stop.
* WIFCONTINUED(stat_val) - Evaluates to a non-zero value if status was
* returned for a child process that has continued from a job control
* stop.
*
* Input Parameters:
* pid - The task ID of the thread to waid for
* stat_loc - The location to return the exit status
* options - Modifiable behavior, see sys/wait.h.
*
* Returned Value:
* If waitpid() returns because the status of a child process is available,
* it will return a value equal to the process ID of the child process for
* which status is reported.
*
* If waitpid() returns due to the delivery of a signal to the calling
* process, -1 will be returned and errno set to EINTR.
*
* If waitpid() was invoked with WNOHANG set in options, it has at least
* one child process specified by pid for which status is not available,
* and status is not available for any process specified by pid, 0 is
* returned.
*
* Otherwise, (pid_t)-1 will be returned, and errno set to indicate the
* error:
*
* ECHILD - The process specified by pid does not exist or is not a child
* of the calling process, or the process group specified by pid
* does not exist does not have any member process that is a child
* of the calling process.
* EINTR - The function was interrupted by a signal. The value of the
* location pointed to by stat_loc is undefined.
* EINVAL - The options argument is not valid.
*
* Assumptions:
*
* Compatibility
* If there is no SIGCHLD signal supported (CONFIG_SCHED_HAVE_PARENT not
* defined), then waitpid() is still available, but does not obey the
* restriction that the pid be a child of the caller.
*
****************************************************************************/
pid_t waitpid(pid_t pid, int *stat_loc, int options)
{
pid_t ret;
/* waitpid() is a cancellation point */
enter_cancellation_point();
/* Let nxsched_waitpid() do the work. */
ret = nxsched_waitpid(pid, stat_loc, options);
if (ret < 0)
{
set_errno(-ret);
ret = ERROR;
}
leave_cancellation_point();
return ret;
}
#endif /* CONFIG_SCHED_WAITPID */