(again..) PATCH: CLONE_PPID and friends - comments>?

Tim Hockin (thockin@isunix.it.ilstu.edu)
Tue, 10 Aug 1999 01:53:42 -0500 (CDT)

Here is the (latest edition of the) final CLONE_PPID patch. I reworked
sys_wait4, and few other comments I received.

Anyone see anything wrong with this? Else I will make a 2.3.x diff and
pass it on to Linus before the 2.4 freeze..

Tim

diff -ru clean-linux-2.2.10/fs/exec.c clone-linux-2.2.10/fs/exec.c
--- clean-linux-2.2.10/fs/exec.c Sun Jun 13 10:50:04 1999
+++ clone-linux-2.2.10/fs/exec.c Thu Jul 22 03:17:00 1999
@@ -846,9 +846,12 @@

if (retval >= 0)
retval = search_binary_handler(&bprm,regs);
- if (retval >= 0)
+ if (retval >= 0) {
+ /* clear any thread-special bits - JIC */
+ current->flags &= ~(PF_PPIDOK | PF_CLWAIT);
/* execve success */
return retval;
+ }

/* Something went wrong, return the inode and free the argument pages*/
if (bprm.dentry)
diff -ru clean-linux-2.2.10/include/linux/sched.h clone-linux-2.2.10/include/linux/sched.h
--- clean-linux-2.2.10/include/linux/sched.h Tue May 11 10:35:45 1999
+++ clone-linux-2.2.10/include/linux/sched.h Thu Jul 22 02:51:17 1999
@@ -34,6 +34,10 @@
#define CLONE_PID 0x00001000 /* set if pid shared */
#define CLONE_PTRACE 0x00002000 /* set if we want to let tracing continue on the child too */
#define CLONE_VFORK 0x00004000 /* set if the parent wants the child to wake it up on mm_release */
+#define CLONE_PPIDOK 0x00008000 /* set if the child should be able to create siblings */
+#define CLONE_PPID 0x00010000 /* set if we want to create a sibling, not a child */
+#define CLONE_WAIT 0x00020000 /* set if child is allowed to wait() for siblings */
+#define CLONE_SUSPENDED 0x00040000 /* create a process in the stopped state */

/*
* These are the constant used to fake the fixed-point load-average
@@ -332,6 +336,8 @@

#define PF_USEDFPU 0x00100000 /* task used FPU this quantum (SMP) */
#define PF_DTRACE 0x00200000 /* delayed trace (used on m68k, i386) */
+#define PF_PPIDOK 0x00400000 /* task is allowed to use CLONE_PPID */
+#define PF_CLWAIT 0x00800000 /* allowed to wait for siblings */

/*
* Limit the stack by to some sane default: root can always
diff -ru clean-linux-2.2.10/kernel/exit.c clone-linux-2.2.10/kernel/exit.c
--- clean-linux-2.2.10/kernel/exit.c Fri Apr 30 08:13:37 1999
+++ clone-linux-2.2.10/kernel/exit.c Sun Aug 8 16:28:33 1999
@@ -145,6 +145,7 @@
read_lock(&tasklist_lock);
for_each_task(p) {
if (p->p_opptr == father) {
+ p->flags &= ~(PF_PPIDOK | PF_CLWAIT);
p->exit_signal = SIGCHLD;
p->p_opptr = child_reaper; /* init */
if (p->pdeath_signal) send_sig(p->pdeath_signal, p, 0);
@@ -407,16 +408,22 @@
{
int flag, retval;
struct wait_queue wait = { current, NULL };
+ struct wait_queue wait_sib = { current, NULL };
struct task_struct *p;
+ struct task_struct *start;

if (options & ~(WNOHANG|WUNTRACED|__WCLONE))
return -EINVAL;

add_wait_queue(&current->wait_chldexit,&wait);
+ if (current->flags & PF_CLWAIT)
+ add_wait_queue(&current->p_pptr->wait_chldexit,&wait_sib);
+
+ start = current->p_cptr;
repeat:
flag = 0;
read_lock(&tasklist_lock);
- for (p = current->p_cptr ; p ; p = p->p_osptr) {
+ for (p = start ; p ; p = p->p_osptr) {
if (pid>0) {
if (p->pid != pid)
continue;
@@ -484,10 +491,19 @@
current->state=TASK_INTERRUPTIBLE;
schedule();
goto repeat;
+ } else {
+ /* try the siblings ? */
+ if ((current->flags & PF_CLWAIT) && (start != current->p_pptr->p_cptr)) {
+ start = current->p_pptr->p_cptr;
+ goto repeat;
+ }
}
retval = -ECHILD;
end_wait4:
remove_wait_queue(&current->wait_chldexit,&wait);
+ if (current->flags & PF_CLWAIT)
+ remove_wait_queue(&current->p_pptr->wait_chldexit,&wait_sib);
+
return retval;
}

diff -ru clean-linux-2.2.10/kernel/fork.c clone-linux-2.2.10/kernel/fork.c
--- clean-linux-2.2.10/kernel/fork.c Mon Apr 12 12:44:26 1999
+++ clone-linux-2.2.10/kernel/fork.c Thu Jul 29 10:08:13 1999
@@ -517,12 +517,18 @@
{
unsigned long new_flags = p->flags;

- new_flags &= ~(PF_SUPERPRIV | PF_USEDFPU | PF_VFORK);
+ new_flags &= ~(PF_SUPERPRIV | PF_USEDFPU | PF_VFORK
+ | PF_PPIDOK | PF_CLWAIT);
new_flags |= PF_FORKNOEXEC;
if (!(clone_flags & CLONE_PTRACE))
new_flags &= ~(PF_PTRACED|PF_TRACESYS);
if (clone_flags & CLONE_VFORK)
new_flags |= PF_VFORK;
+ if (clone_flags & CLONE_PPIDOK)
+ new_flags |= PF_PPIDOK;
+ if (clone_flags & CLONE_WAIT)
+ new_flags |= PF_CLWAIT;
+
p->flags = new_flags;
}

@@ -587,7 +593,14 @@
p->next_run = p;
p->prev_run = p;

- p->p_pptr = p->p_opptr = current;
+ if (!(clone_flags & CLONE_PPID)) {
+ p->p_pptr = p->p_opptr = current;
+ } else if (current->flags & PF_PPIDOK) {
+ p->p_pptr = p->p_opptr = current->p_opptr;
+ } else {
+ retval = -EPERM;
+ goto bad_fork_cleanup;
+ }
p->p_cptr = NULL;
init_waitqueue(&p->wait_chldexit);
p->vfork_sem = NULL;
@@ -668,7 +681,10 @@

p->next_run = NULL;
p->prev_run = NULL;
- wake_up_process(p); /* do this last */
+ if (!(clone_flags & CLONE_SUSPENDED))
+ wake_up_process(p); /* do this last */
+ else
+ p->state = TASK_STOPPED;
}
++total_forks;
bad_fork:

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/