Always do io_commit_cqring() after completing a request, even if it was accounted as overflowed on the CQ side. Failing to do that may lead to not to pushing deferred requests when needed, and so stalling the whole ring. Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx> --- fs/io_uring.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/io_uring.c b/fs/io_uring.c index 6e2322525da6..11c1abe8bd1a 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -7849,6 +7849,7 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx, } WRITE_ONCE(ctx->rings->cq_overflow, atomic_inc_return(&ctx->cached_cq_overflow)); + io_commit_cqring(ctx); spin_unlock_irq(&ctx->completion_lock); /* -- 2.24.0
