> On Jul 22, 2024, at 10:21 AM, Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote: > > On Sat, Jul 20, 2024 at 04:35:38PM +0800, Yang Erkun wrote: >> We use offset_readdir for tmpfs, and every we call rename, the offset >> for the parent dir will increase by 1. So for tmpfs we will always >> fail since the infinite readdir. > > This honestly sounds like a bug in tmpfs, so maybe we should discuss > the behavior first? Agreed, sounds like there should be some root cause analysis before the tests are altered. Is this problem addressed in recent kernels? -- Chuck Lever
