Return-Path: Date: Mon, 25 Aug 2014 11:11:56 -0400 From: Benjamin LaHaise To: "Elliott, Robert (Server Storage)" Cc: Dan Aloni , Linus Torvalds , "security@kernel.org" , "linux-aio@kvack.org" , "linux-kernel@vger.kernel.org" , Mateusz Guzik , Petr Matousek , Kent Overstreet , Jeff Moyer , "stable@vger.kernel.org" Subject: Re: Revert "aio: fix aio request leak when events are reaped by user space" Message-ID: <20140825151156.GE20070@kvack.org> References: <20140819163733.GA10132@gmail.com> <20140819165404.GD13858@kvack.org> <20140819171426.GA11811@gmail.com> <20140820004651.GJ13858@kvack.org> <20140822160111.GD20391@kvack.org> <20140822161502.GA30392@gmail.com> <20140822162630.GF20391@kvack.org> <94D0CD8314A33A4D9D801C0FE68B402958C22133@G9W0745.americas.hpqcorp.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <94D0CD8314A33A4D9D801C0FE68B402958C22133@G9W0745.americas.hpqcorp.net> Sender: linux-kernel-owner@vger.kernel.org List-ID: On Mon, Aug 25, 2014 at 03:06:12PM +0000, Elliott, Robert (Server Storage) wrote: > Using this version of the patch, I ran into this crash after 36 > hours of scsi-mq testing over the weekend. ... > > local_irq_save(flags); > kcpu = this_cpu_ptr(ctx->cpu); > kcpu->reqs_available += nr; > > while (kcpu->reqs_available >= ctx->req_batch * 2) { This crash is not particularly useful, as that version of the patch was buggy and could result in reqs_available getting corrupted (ie, it could overflow), which could lead to the system getting stuck for a significant amount of time in put_reqs_available(). In other words, the crash you reported matches the known problem with that patch. -ben -- "Thought is the essence of where you are now."