forkfd: update the API to add a reading and closing function

Right now the two functions being added are just wrappers, but this will
allow us in the future to support FreeBSD and Linux's system calls that
do the equivalent of forkfd, but have slightly different semantics in
the actual getting of the information from the file descriptor.

See-Also: https://lkml.org/lkml/2015/3/12/1044
See-Also: http://www.freebsd.org/cgi/man.cgi?query=pdfork
Change-Id: Ia0aac2f09e9245339951ffff13c94acb5f4ff204
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@theqtcompany.com>
Reviewed-by: Rafael Roquetto <rafael.roquetto@kdab.com>
Reviewed-by: Thiago Macieira <thiago.macieira@intel.com>

src/3rdparty/forkfd/forkfd.c

@@ -1,6 +1,6 @@
 /****************************************************************************
 **
-** Copyright (C) 2014 Intel Corporation
+** Copyright (C) 2015 Intel Corporation
 ** Copyright (C) 2015 Klarälvdalens Datakonsult AB, a KDAB Group company, info@kdab.com
 **
 ** Permission is hereby granted, free of charge, to any person obtaining a copy
@@ -31,6 +31,8 @@
 #include "forkfd.h"
 
 #include <sys/types.h>
+#include <sys/time.h>
+#include <sys/resource.h>
 #include <sys/wait.h>
 #include <assert.h>
 #include <errno.h>
@@ -38,6 +40,7 @@
 #include <signal.h>
 #include <stdlib.h>
 #include <string.h>
+#include <time.h>
 #include <unistd.h>
 
 #ifdef __linux__
@@ -79,6 +82,12 @@
         ret = call;           \
     } while (ret == -1 && errno == EINTR)
 
+struct pipe_payload
+{
+    struct forkfd_info info;
+    struct rusage rusage;
+};
+
 typedef struct process_info
 {
     ffd_atomic_int pid;
@@ -179,33 +188,43 @@ static int isChildReady(pid_t pid, siginfo_t *info)
 }
 #endif
 
-static int tryReaping(pid_t pid, siginfo_t *info)
+static int tryReaping(pid_t pid, struct pipe_payload *payload)
 {
     /* reap the child */
 #ifdef HAVE_WAITID
     if (waitid_works) {
         // we have waitid(2), which fills in siginfo_t for us
-        info->si_pid = 0;
-        return waitid(P_PID, pid, info, WEXITED | WNOHANG) == 0 && info->si_pid == pid;
+        siginfo_t info;
+        info.si_pid = 0;
+        int ret = waitid(P_PID, pid, &info, WEXITED | WNOHANG) == 0 && info.si_pid == pid;
+        if (!ret)
+            return ret;
+
+        payload->info.code = info.si_code;
+        payload->info.status = info.si_status;
+#  ifdef __linux__
+        payload->rusage.ru_utime.tv_sec = info.si_utime / CLOCKS_PER_SEC;
+        payload->rusage.ru_utime.tv_usec = info.si_utime % CLOCKS_PER_SEC;
+        payload->rusage.ru_stime.tv_sec = info.si_stime / CLOCKS_PER_SEC;
+        payload->rusage.ru_stime.tv_usec = info.si_stime % CLOCKS_PER_SEC;
+#  endif
+        return 1;
     }
 #endif
-
     int status;
     if (waitpid(pid, &status, WNOHANG) <= 0)
         return 0;     // child did not change state
 
-    info->si_signo = SIGCHLD;
-    info->si_pid = pid;
     if (WIFEXITED(status)) {
-        info->si_code = CLD_EXITED;
-        info->si_status = WEXITSTATUS(status);
+        payload->info.code = CLD_EXITED;
+        payload->info.status = WEXITSTATUS(status);
     } else if (WIFSIGNALED(status)) {
-        info->si_code = CLD_KILLED;
+        payload->info.code = CLD_KILLED;
 #  ifdef WCOREDUMP
         if (WCOREDUMP(status))
-            info->si_code = CLD_DUMPED;
+            payload->info.code = CLD_DUMPED;
 #  endif
-        info->si_status = WTERMSIG(status);
+        payload->info.status = WTERMSIG(status);
     }
 
     return 1;
@@ -220,10 +239,11 @@ static void freeInfo(Header *header, ProcessInfo *entry)
     assert(header->busyCount >= 0);
 }
 
-static void notifyAndFreeInfo(Header *header, ProcessInfo *entry, siginfo_t *info)
+static void notifyAndFreeInfo(Header *header, ProcessInfo *entry,
+                              const struct pipe_payload *payload)
 {
     ssize_t ret;
-    EINTR_LOOP(ret, write(entry->deathPipe, info, sizeof(*info)));
+    EINTR_LOOP(ret, write(entry->deathPipe, payload, sizeof(*payload)));
     EINTR_LOOP(ret, close(entry->deathPipe));
 
     freeInfo(header, entry);
@@ -243,9 +263,11 @@ static void sigchld_handler(int signum)
         /* is this one of our children? */
         BigArray *array;
         siginfo_t info;
+        struct pipe_payload payload;
         int i;
 
         memset(&info, 0, sizeof info);
+        memset(&payload, 0, sizeof payload);
 
 #ifdef HAVE_WAITID
         if (!waitid_works)
@@ -275,8 +297,8 @@ search_next_child:
                                             FFD_ATOMIC_ACQUIRE, FFD_ATOMIC_RELAXED)) {
                 /* this is our child, send notification and free up this entry */
                 /* ### FIXME: what if tryReaping returns false? */
-                if (tryReaping(pid, &info))
-                    notifyAndFreeInfo(&children.header, &children.entries[i], &info);
+                if (tryReaping(pid, &payload))
+                    notifyAndFreeInfo(&children.header, &children.entries[i], &payload);
                 goto search_next_child;
             }
         }
@@ -290,8 +312,8 @@ search_next_child:
                                                 FFD_ATOMIC_ACQUIRE, FFD_ATOMIC_RELAXED)) {
                     /* this is our child, send notification and free up this entry */
                     /* ### FIXME: what if tryReaping returns false? */
-                    if (tryReaping(pid, &info))
-                        notifyAndFreeInfo(&array->header, &array->entries[i], &info);
+                    if (tryReaping(pid, &payload))
+                        notifyAndFreeInfo(&array->header, &array->entries[i], &payload);
                     goto search_next_child;
                 }
             }
@@ -321,9 +343,9 @@ search_arrays:
                     continue;
             }
 #endif
-            if (tryReaping(pid, &info)) {
+            if (tryReaping(pid, &payload)) {
                 /* this is our child, send notification and free up this entry */
-                notifyAndFreeInfo(&children.header, &children.entries[i], &info);
+                notifyAndFreeInfo(&children.header, &children.entries[i], &payload);
             }
         }
 
@@ -344,9 +366,9 @@ search_arrays:
                         continue;
                 }
 #endif
-                if (tryReaping(pid, &info)) {
+                if (tryReaping(pid, &payload)) {
                     /* this is our child, send notification and free up this entry */
-                    notifyAndFreeInfo(&array->header, &array->entries[i], &info);
+                    notifyAndFreeInfo(&array->header, &array->entries[i], &payload);
                 }
             }
 
@@ -626,7 +648,7 @@ int spawnfd(int flags, pid_t *ppid, const char *path, const posix_spawn_file_act
 {
     Header *header;
     ProcessInfo *info;
-    siginfo_t si;
+    struct pipe_payload payload;
     pid_t pid;
     int death_pipe[2];
     int ret = -1;
@@ -664,8 +686,8 @@ int spawnfd(int flags, pid_t *ppid, const char *path, const posix_spawn_file_act
     ffd_atomic_store(&info->pid, pid, FFD_ATOMIC_RELEASE);
 
     /* check if the child has already exited */
-    if (tryReaping(pid, &si))
-        notifyAndFreeInfo(header, info, &si);
+    if (tryReaping(pid, &payload))
+        notifyAndFreeInfo(header, info, &payload);
 
     ret = death_pipe[0];
     return ret;
@@ -682,3 +704,28 @@ out:
     return -1;
 }
 #endif // _POSIX_SPAWN && !FORKFD_NO_SPAWNFD
+
+
+int forkfd_wait(int ffd, forkfd_info *info, struct rusage *rusage)
+{
+    struct pipe_payload payload;
+    int ret;
+
+    ret = read(ffd, &payload, sizeof(payload));
+    if (ret == -1)
+        return ret;     /* pass errno, probably EINTR, EBADF or EWOULDBLOCK */
+
+    assert(ret == sizeof(payload));
+    if (info)
+        *info = payload.info;
+    if (rusage)
+        *rusage = payload.rusage;
+
+    return 0;           /* success */
+}
+
+
+int forkfd_close(int ffd)
+{
+    return close(ffd);
+}

src/3rdparty/forkfd/forkfd.h

@@ -1,6 +1,6 @@
 /****************************************************************************
 **
-** Copyright (C) 2014 Intel Corporation
+** Copyright (C) 2015 Intel Corporation
 **
 ** Permission is hereby granted, free of charge, to any person obtaining a copy
 ** of this software and associated documentation files (the "Software"), to deal
@@ -26,6 +26,7 @@
 #define FORKFD_H
 
 #include <fcntl.h>
+#include <stdint.h>
 #include <unistd.h> // to get the POSIX flags
 
 #ifdef _POSIX_SPAWN
@@ -41,7 +42,14 @@ extern "C" {
 
 #define FFD_CHILD_PROCESS (-2)
 
+struct forkfd_info {
+    int32_t code;
+    int32_t status;
+};
+
 int forkfd(int flags, pid_t *ppid);
+int forkfd_wait(int ffd, forkfd_info *info, struct rusage *rusage);
+int forkfd_close(int ffd);
 
 #ifdef _POSIX_SPAWN
 /* only for spawnfd: */

src/corelib/io/qprocess_unix.cpp

@@ -1091,20 +1091,17 @@ bool QProcessPrivate::waitForDeadChild()
         return true; // child has already exited
 
     // read the process information from our fd
-    siginfo_t info;
-    qint64 ret = qt_safe_read(forkfd, &info, sizeof info);
-    Q_ASSERT(ret == sizeof info);
-    Q_UNUSED(ret);
+    forkfd_info info;
+    int ret;
+    EINTR_LOOP(ret, forkfd_wait(forkfd, &info, Q_NULLPTR));
 
-    Q_ASSERT(info.si_pid == pid_t(pid));
-
-    exitCode = info.si_status;
-    crashed = info.si_code != CLD_EXITED;
+    exitCode = info.status;
+    crashed = info.code != CLD_EXITED;
 
     delete deathNotifier;
     deathNotifier = 0;
 
-    qt_safe_close(forkfd);
+    EINTR_LOOP(ret, forkfd_close(forkfd));
     forkfd = -1; // Child is dead, don't try to kill it anymore
 
 #if defined QPROCESS_DEBUG