The patch titled
Subject: exit: reparent: avoid find_new_reaper() if no children
has been removed from the -mm tree. Its filename was
exit-reparent-avoid-find_new_reaper-if-no-children.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Oleg Nesterov <oleg@xxxxxxxxxx>
Subject: exit: reparent: avoid find_new_reaper() if no children
Now that pid_ns logic was isolated we can change forget_original_parent()
to return right after find_child_reaper() when father->children is empty,
there is nothing to reparent in this case.
In particular this avoids find_alive_thread() and this can help if the
whole process exits and it has a lot of PF_EXITING threads at the start of
the thread list, this can easily lead to O(nr_threads ** 2) iterations.
Trivial test case (tested under KVM, 2 CPUs):
static void *tfunc(void *arg)
{
pause();
return NULL;
}
static int child(unsigned int nt)
{
pthread_t pt;
while (nt--)
assert(pthread_create(&pt, NULL, tfunc, NULL) == 0);
pthread_kill(pt, SIGTRAP);
pause();
return 0;
}
int main(int argc, const char *argv[])
{
int stat;
unsigned int nf = atoi(argv[1]);
unsigned int nt = atoi(argv[2]);
while (nf--) {
if (!fork())
return child(nt);
wait(&stat);
assert(stat == SIGTRAP);
}
return 0;
}
$ time ./test 16 16536 shows:
real user sys
- 5m37.628s 0m4.437s 8m5.560s
+ 0m50.032s 0m7.130s 1m4.927s
Signed-off-by: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: Aaron Tomlin <atomlin@xxxxxxxxxx>
Cc: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx>
Cc: Sterling Alexander <stalexan@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
kernel/exit.c | 3 +++
1 file changed, 3 insertions(+)
diff -puN kernel/exit.c~exit-reparent-avoid-find_new_reaper-if-no-children kernel/exit.c
--- a/kernel/exit.c~exit-reparent-avoid-find_new_reaper-if-no-children
+++ a/kernel/exit.c
@@ -571,6 +571,8 @@ static void forget_original_parent(struc
/* Can drop and reacquire tasklist_lock */
reaper = find_child_reaper(father);
+ if (list_empty(&father->children))
+ goto unlock;
reaper = find_new_reaper(father, reaper);
list_for_each_entry(p, &father->children, sibling) {
@@ -591,6 +593,7 @@ static void forget_original_parent(struc
reparent_leader(father, p, &dead_children);
}
list_splice_tail_init(&father->children, &reaper->children);
+ unlock:
write_unlock_irq(&tasklist_lock);
list_for_each_entry_safe(p, n, &dead_children, ptrace_entry) {
_
Patches currently in -mm which might be from oleg@xxxxxxxxxx are
origin.patch
mmfs-introduce-helpers-around-the-i_mmap_mutex.patch
mm-use-new-helper-functions-around-the-i_mmap_mutex.patch
mm-convert-i_mmap_mutex-to-rwsem.patch
mm-rmap-share-the-i_mmap_rwsem.patch
uprobes-share-the-i_mmap_rwsem.patch
mm-xip-share-the-i_mmap_rwsem.patch
mm-memory-failure-share-the-i_mmap_rwsem.patch
mm-nommu-share-the-i_mmap_rwsem.patch
mm-memoryc-share-the-i_mmap_rwsem.patch
remove-unnecessary-is_valid_nodemask.patch
oom-dont-assume-that-a-coredumping-thread-will-exit-soon.patch
oom-dont-assume-that-a-coredumping-thread-will-exit-soon-v2.patch
oom-dont-assume-that-a-coredumping-thread-will-exit-soon-v2-fix.patch
oom-kill-the-insufficient-and-no-longer-needed-pt_trace_exit-check.patch
linux-next.patch
all-arches-signal-move-restart_block-to-struct-task_struct.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
