On 2/3/26 2:59 PM, Al Viro wrote:
On Tue, Feb 03, 2026 at 02:44:33PM -0500, Waiman Long wrote:
+ /*
+ * Defer the path_put() call of the old pwd to workqueue as
+ * we may be in an atomic context that cannot call path_put()
+ * directly because of might_sleep().
+ */
+ pp = kmalloc(sizeof(*pp), GFP_NOFS);
If that really gets called in atomic context, this kmalloc will
in itself be enough to fuck you over, won't it?
Looking at the bug report again, I realized that the problem is because
of the get_cond_fs_pwd() helper in my v1 patch call path_put() in a
spinlock critical section. So we don't need to use workqueue here, just
doing a path_put() should be good enough.
Thanks, Longman
