From 568ac888215c7fb2fabe8ea739b00ec3c1f5d440 Mon Sep 17 00:00:00 2001 From: Balbir Singh Date: Wed, 10 Aug 2016 15:43:06 -0400 Subject: cgroup: reduce read locked section of cgroup_threadgroup_rwsem during fork cgroup_threadgroup_rwsem is acquired in read mode during process exit and fork. It is also grabbed in write mode during __cgroups_proc_write(). I've recently run into a scenario with lots of memory pressure and OOM and I am beginning to see systemd __switch_to+0x1f8/0x350 __schedule+0x30c/0x990 schedule+0x48/0xc0 percpu_down_write+0x114/0x170 __cgroup_procs_write.isra.12+0xb8/0x3c0 cgroup_file_write+0x74/0x1a0 kernfs_fop_write+0x188/0x200 __vfs_write+0x6c/0xe0 vfs_write+0xc0/0x230 SyS_write+0x6c/0x110 system_call+0x38/0xb4 This thread is waiting on the reader of cgroup_threadgroup_rwsem to exit. The reader itself is under memory pressure and has gone into reclaim after fork. There are times the reader also ends up waiting on oom_lock as well. __switch_to+0x1f8/0x350 __schedule+0x30c/0x990 schedule+0x48/0xc0 jbd2_log_wait_commit+0xd4/0x180 ext4_evict_inode+0x88/0x5c0 evict+0xf8/0x2a0 dispose_list+0x50/0x80 prune_icache_sb+0x6c/0x90 super_cache_scan+0x190/0x210 shrink_slab.part.15+0x22c/0x4c0 shrink_zone+0x288/0x3c0 do_try_to_free_pages+0x1dc/0x590 try_to_free_pages+0xdc/0x260 __alloc_pages_nodemask+0x72c/0xc90 alloc_pages_current+0xb4/0x1a0 page_table_alloc+0xc0/0x170 __pte_alloc+0x58/0x1f0 copy_page_range+0x4ec/0x950 copy_process.isra.5+0x15a0/0x1870 _do_fork+0xa8/0x4b0 ppc_clone+0x8/0xc In the meanwhile, all processes exiting/forking are blocked almost stalling the system. This patch moves the threadgroup_change_begin from before cgroup_fork() to just before cgroup_canfork(). There is no nee to worry about threadgroup changes till the task is actually added to the threadgroup. This avoids having to call reclaim with cgroup_threadgroup_rwsem held. tj: Subject and description edits. Signed-off-by: Balbir Singh Acked-by: Zefan Li Cc: Oleg Nesterov Cc: Andrew Morton Cc: stable@vger.kernel.org # v4.2+ Signed-off-by: Tejun Heo --- kernel/fork.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/fork.c') diff --git a/kernel/fork.c b/kernel/fork.c index 52e725d4a866..aaf782327bf3 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1404,7 +1404,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->real_start_time = ktime_get_boot_ns(); p->io_context = NULL; p->audit_context = NULL; - threadgroup_change_begin(current); cgroup_fork(p); #ifdef CONFIG_NUMA p->mempolicy = mpol_dup(p->mempolicy); @@ -1556,6 +1555,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, INIT_LIST_HEAD(&p->thread_group); p->task_works = NULL; + threadgroup_change_begin(current); /* * Ensure that the cgroup subsystem policies allow the new process to be * forked. It should be noted the the new process's css_set can be changed @@ -1656,6 +1656,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, bad_fork_cancel_cgroup: cgroup_cancel_fork(p); bad_fork_free_pid: + threadgroup_change_end(current); if (pid != &init_struct_pid) free_pid(pid); bad_fork_cleanup_thread: @@ -1688,7 +1689,6 @@ bad_fork_cleanup_policy: mpol_put(p->mempolicy); bad_fork_cleanup_threadgroup_lock: #endif - threadgroup_change_end(current); delayacct_tsk_free(p); bad_fork_cleanup_count: atomic_dec(&p->cred->user->processes); -- cgit v1.2.3 From cd81a9170e69e018bbaba547c1fd85a585f5697a Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Tue, 23 Aug 2016 16:20:38 +0200 Subject: mm: introduce get_task_exe_file For more convenient access if one has a pointer to the task. As a minor nit take advantage of the fact that only task lock + rcu are needed to safely grab ->exe_file. This saves mm refcount dance. Use the helper in proc_exe_link. Signed-off-by: Mateusz Guzik Acked-by: Konstantin Khlebnikov Acked-by: Richard Guy Briggs Cc: # 4.3.x Signed-off-by: Paul Moore --- kernel/fork.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'kernel/fork.c') diff --git a/kernel/fork.c b/kernel/fork.c index d277e83ed3e0..42451aeb245f 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -773,6 +773,29 @@ struct file *get_mm_exe_file(struct mm_struct *mm) } EXPORT_SYMBOL(get_mm_exe_file); +/** + * get_task_exe_file - acquire a reference to the task's executable file + * + * Returns %NULL if task's mm (if any) has no associated executable file or + * this is a kernel thread with borrowed mm (see the comment above get_task_mm). + * User must release file via fput(). + */ +struct file *get_task_exe_file(struct task_struct *task) +{ + struct file *exe_file = NULL; + struct mm_struct *mm; + + task_lock(task); + mm = task->mm; + if (mm) { + if (!(task->flags & PF_KTHREAD)) + exe_file = get_mm_exe_file(mm); + } + task_unlock(task); + return exe_file; +} +EXPORT_SYMBOL(get_task_exe_file); + /** * get_task_mm - acquire a reference to the task's mm * -- cgit v1.2.3 From 735f2770a770156100f534646158cb58cb8b2939 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Thu, 1 Sep 2016 16:15:13 -0700 Subject: kernel/fork: fix CLONE_CHILD_CLEARTID regression in nscd Commit fec1d0115240 ("[PATCH] Disable CLONE_CHILD_CLEARTID for abnormal exit") has caused a subtle regression in nscd which uses CLONE_CHILD_CLEARTID to clear the nscd_certainly_running flag in the shared databases, so that the clients are notified when nscd is restarted. Now, when nscd uses a non-persistent database, clients that have it mapped keep thinking the database is being updated by nscd, when in fact nscd has created a new (anonymous) one (for non-persistent databases it uses an unlinked file as backend). The original proposal for the CLONE_CHILD_CLEARTID change claimed (https://lkml.org/lkml/2006/10/25/233): : The NPTL library uses the CLONE_CHILD_CLEARTID flag on clone() syscalls : on behalf of pthread_create() library calls. This feature is used to : request that the kernel clear the thread-id in user space (at an address : provided in the syscall) when the thread disassociates itself from the : address space, which is done in mm_release(). : : Unfortunately, when a multi-threaded process incurs a core dump (such as : from a SIGSEGV), the core-dumping thread sends SIGKILL signals to all of : the other threads, which then proceed to clear their user-space tids : before synchronizing in exit_mm() with the start of core dumping. This : misrepresents the state of process's address space at the time of the : SIGSEGV and makes it more difficult for someone to debug NPTL and glibc : problems (misleading him/her to conclude that the threads had gone away : before the fault). : : The fix below is to simply avoid the CLONE_CHILD_CLEARTID action if a : core dump has been initiated. The resulting patch from Roland (https://lkml.org/lkml/2006/10/26/269) seems to have a larger scope than the original patch asked for. It seems that limitting the scope of the check to core dumping should work for SIGSEGV issue describe above. [Changelog partly based on Andreas' description] Fixes: fec1d0115240 ("[PATCH] Disable CLONE_CHILD_CLEARTID for abnormal exit") Link: http://lkml.kernel.org/r/1471968749-26173-1-git-send-email-mhocko@kernel.org Signed-off-by: Michal Hocko Tested-by: William Preston Acked-by: Oleg Nesterov Cc: Roland McGrath Cc: Andreas Schwab Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'kernel/fork.c') diff --git a/kernel/fork.c b/kernel/fork.c index aaf782327bf3..93bdba13d7d9 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -913,14 +913,12 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm) deactivate_mm(tsk, mm); /* - * If we're exiting normally, clear a user-space tid field if - * requested. We leave this alone when dying by signal, to leave - * the value intact in a core dump, and to save the unnecessary - * trouble, say, a killed vfork parent shouldn't touch this mm. - * Userland only wants this done for a sys_exit. + * Signal userspace if we're not exiting with a core dump + * because we want to leave the value intact for debugging + * purposes. */ if (tsk->clear_child_tid) { - if (!(tsk->flags & PF_SIGNALED) && + if (!(tsk->signal->flags & SIGNAL_GROUP_COREDUMP) && atomic_read(&mm->mm_users) > 1) { /* * We don't check the error code - if userspace has -- cgit v1.2.3