From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin LaHaise Subject: Re: [PATCH][WIP v1] aio: experimental use of threads, demonstration of cancel method Date: Thu, 14 Feb 2013 16:03:09 -0500 Message-ID: <20130214210309.GD25505@kvack.org> References: <20130213221632.GA25312@kvack.org> <20130214193333.GG22221@lenny.home.zabbo.net> <20130214205328.GB25505@kvack.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Kent Overstreet , linux-aio@kvack.org, linux-fsdevel@vger.kernel.org, Linux Kernel To: Zach Brown Return-path: Content-Disposition: inline In-Reply-To: <20130214205328.GB25505@kvack.org> Sender: owner-linux-aio@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Thu, Feb 14, 2013 at 03:53:28PM -0500, Benjamin LaHaise wrote: > If the kiocb has been cancelled, this code ensures that the helper thread > has received the SIGSEGV. This ensures that the signal can then be flushed > by flush_signals(). If the kiocb has not been cancelled, it can no longer > be cancelled after the cmpxchg() has executed. Whoops, SIGSEGV will certainly need special handling. The cancel code should really be using SIGKILL. I will add that to the list of things to fix. -ben -- "Thought is the essence of where you are now." -- To unsubscribe, send a message with 'unsubscribe linux-aio' in the body to majordomo@kvack.org. For more info on Linux AIO, see: http://www.kvack.org/aio/ Don't email: aart@kvack.org