From mboxrd@z Thu Jan 1 00:00:00 1970 From: hch@infradead.org (Christoph Hellwig) Date: Wed, 10 May 2017 09:55:31 -0700 Subject: [PATCH] nvmet: release the sq ref on rdma read errors In-Reply-To: <014901d2c528$ec347430$c49d5c90$@attalasystems.com> References: <01cb01d2c446$20cd83a0$62688ae0$@attalasystems.com> <01d601d2c446$b087cb90$119762b0$@attalasystems.com> <00d6f8bf-07bd-03ca-3060-fd3def1ef1d7@grimberg.me> <20170504085631.GA5584@lst.de> <5e0ca611-97d8-8884-abc3-c0f3b6172a21@grimberg.me> <014901d2c528$ec347430$c49d5c90$@attalasystems.com> Message-ID: <20170510165531.GB7992@infradead.org> On Thu, May 04, 2017@03:51:09PM -0700, Vijay Immanuel wrote: > Thanks for the feedback. Here's an updated patch. > ---- > > On rdma read errors, release the sq ref that was taken > when the req was initialized. This avoids a hang in > nvmet_sq_destroy() when the queue is being freed. > > Signed-off-by: Vijay Immanuel Looks good: Reviewed-by: Christoph Hellwig