Messages in this thread Patch in this message | | | Date | Thu, 9 Jul 2009 19:36:14 +0200 | From | Oleg Nesterov <> | Subject | [PATCH 1/1] proc_flush_task: flush /proc/tid/task/pid when a sub-thread exits |
| |
The exiting sub-thread flushes /proc/pid only, but this doesn't buy too much: ps and friends mostly use /proc/tid/task/pid.
Remove "if (thread_group_leader())" checks from proc_flush_task() path, this means we always remove /proc/tid/task/pid dentry on exit, and this actually matches the comment above proc_flush_task().
The test-case:
static void* tfunc(void *arg) { char name[256];
sprintf(name, "/proc/%d/task/%ld/status", getpid(), gettid()); close(open(name, O_RDONLY));
return NULL; }
int main(void) { pthread_t t;
for (;;) { if (!pthread_create(&t, NULL, &tfunc, NULL)) pthread_join(t, NULL); } }
slabtop shows that pid/proc_inode_cache/etc grow quickly and "indefinitely" until the task is killed or shrink_slab() is called, not good. And the main thread needs a lot of time to exit.
The same can happen if something like "ps -efL" runs continuously, while some application spawns short-living threads.
Reported-by: "James M. Leddy" <jleddy@redhat.com> Signed-off-by: Oleg Nesterov <oleg@redhat.com>
--- WAIT/fs/proc/base.c~PROC_FLUSH_THREAD 2009-06-23 20:16:44.000000000 +0200 +++ WAIT/fs/proc/base.c 2009-07-09 17:56:27.000000000 +0200 @@ -2601,9 +2601,6 @@ static void proc_flush_task_mnt(struct v dput(dentry); } - if (tgid == 0) - goto out; - name.name = buf; name.len = snprintf(buf, sizeof(buf), "%d", tgid); leader = d_hash_and_lookup(mnt->mnt_root, &name); @@ -2660,17 +2657,16 @@ out: void proc_flush_task(struct task_struct *task) { int i; - struct pid *pid, *tgid = NULL; + struct pid *pid, *tgid; struct upid *upid; pid = task_pid(task); - if (thread_group_leader(task)) - tgid = task_tgid(task); + tgid = task_tgid(task); for (i = 0; i <= pid->level; i++) { upid = &pid->numbers[i]; proc_flush_task_mnt(upid->ns->proc_mnt, upid->nr, - tgid ? tgid->numbers[i].nr : 0); + tgid->numbers[i].nr); } upid = &pid->numbers[pid->level];
| |