eventfd: track eventfd_signal() recursion depth

eventfd use cases from aio and io_uring can deadlock due to circular
or resursive calling, when eventfd_signal() tries to grab the waitqueue
lock. On top of that, it's also possible to construct notification
chains that are deep enough that we could blow the stack.

Add a percpu counter that tracks the percpu recursion depth, warn if we
exceed it. The counter is also exposed so that users of eventfd_signal()
can do the right thing if it's non-zero in the context where it is
called.

Cc: stable@vger.kernel.org # 4.19+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Jens Axboe
2020-02-02 08:23:03 -07:00
parent d7f62e825f
commit b5e683d5ca
2 changed files with 29 additions and 0 deletions

View File

@@ -12,6 +12,8 @@
#include <linux/fcntl.h>
#include <linux/wait.h>
#include <linux/err.h>
#include <linux/percpu-defs.h>
#include <linux/percpu.h>
/*
* CAREFUL: Check include/uapi/asm-generic/fcntl.h when defining
@@ -40,6 +42,13 @@ __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n);
int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx, wait_queue_entry_t *wait,
__u64 *cnt);
DECLARE_PER_CPU(int, eventfd_wake_count);
static inline bool eventfd_signal_count(void)
{
return this_cpu_read(eventfd_wake_count);
}
#else /* CONFIG_EVENTFD */
/*
@@ -68,6 +77,11 @@ static inline int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx,
return -ENOSYS;
}
static inline bool eventfd_signal_count(void)
{
return false;
}
#endif
#endif /* _LINUX_EVENTFD_H */