[PATCH] mm: correctly synchronize rss-counters at exit/exec
From: Konstantin Khlebnikov <hidden>
Date: 2012-06-09 09:43:40
Also in:
linux-mm, lkml
Subsystem:
exec & binfmt api, elf, filesystems (vfs and infrastructure), the rest · Maintainers:
Kees Cook, Alexander Viro, Christian Brauner, Linus Torvalds
do_exit() and exec_mmap() call sync_mm_rss() before mm_release() does put_user(clear_child_tid) which can update task->rss_stat and thus make mm->rss_stat inconsistent. This triggers the "BUG:" printk in check_mm(). Let's fix this bug in the safest way, and optimize/cleanup this later. Reported-by: Markus Trippelsdorf <redacted> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Konstantin Khlebnikov <redacted> --- fs/exec.c | 2 +- kernel/exit.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/exec.c b/fs/exec.c
index a79786a..da27b91 100644
--- a/fs/exec.c
+++ b/fs/exec.c@@ -819,10 +819,10 @@ static int exec_mmap(struct mm_struct *mm) /* Notify parent that we're no longer interested in the old VM */ tsk = current; old_mm = current->mm; - sync_mm_rss(old_mm); mm_release(tsk, old_mm); if (old_mm) { + sync_mm_rss(old_mm); /* * Make sure that if there is a core dump in progress * for the old mm, we get out and die instead of going
diff --git a/kernel/exit.c b/kernel/exit.c
index 34867cc..c0277d3 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c@@ -643,6 +643,7 @@ static void exit_mm(struct task_struct * tsk) mm_release(tsk, mm); if (!mm) return; + sync_mm_rss(mm); /* * Serialize with any possible pending coredump. * We must hold mmap_sem around checking core_state