From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Date: Wed, 15 Jan 2020 15:33:47 +0100 Subject: [Cluster-devel] RFC: hold i_rwsem until aio completes In-Reply-To: <20200115132428.GA25201@ziepe.ca> References: <20200114161225.309792-1-hch@lst.de> <20200114192700.GC22037@ziepe.ca> <20200115065614.GC21219@lst.de> <20200115132428.GA25201@ziepe.ca> Message-ID: <20200115143347.GL2827@hirez.programming.kicks-ass.net> List-Id: To: cluster-devel.redhat.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit On Wed, Jan 15, 2020 at 09:24:28AM -0400, Jason Gunthorpe wrote: > I was interested because you are talking about allowing the read/write side > of a rw sem to be held across a return to user space/etc, which is the > same basic problem. No it is not; allowing the lock to be held across userspace doesn't change the owner. This is a crucial difference, PI depends on there being a distinct owner. That said, allowing the lock to be held across userspace still breaks PI in that it completely wrecks the ability to analyze the critical section.