Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932897Ab3FRTF0 (ORCPT ); Tue, 18 Jun 2013 15:05:26 -0400 Received: from mail-we0-f180.google.com ([74.125.82.180]:55150 "EHLO mail-we0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932200Ab3FRTFY (ORCPT ); Tue, 18 Jun 2013 15:05:24 -0400 MIME-Version: 1.0 In-Reply-To: References: <1366026055-28604-1-git-send-email-octavian.purdila@intel.com> <20130612181440.GC6151@google.com> <20130612182430.GE14404@kvack.org> <20130612194005.GC8389@lenny.home.zabbo.net> Date: Tue, 18 Jun 2013 22:05:22 +0300 Message-ID: Subject: Re: [PATCH v3 next/akpm] aio: convert the ioctx list to radix tree From: Octavian Purdila To: Zach Brown Cc: Benjamin LaHaise , Kent Overstreet , Andrew Morton , linux-kernel@vger.kernel.org, linux-aio@kvack.org, linux-s390@vger.kernel.org, schwidefsky@de.ibm.com, "Kirill A. Shutemov" , Andi Kleen Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1573 Lines: 33 On Fri, Jun 14, 2013 at 5:20 PM, Octavian Purdila wrote: > On Wed, Jun 12, 2013 at 10:40 PM, Zach Brown wrote: >>> I've got an alternate approach for fixing this wart in lookup_ioctx()... >>> Instead of using an rbtree, just use the reserved id in the ring buffer >>> header to index an array pointing the ioctx. It's not finished yet, and >>> it needs to be tidied up, but is most of the way there. >> >> Yeah, that might work. >> >> Note that the patch wasn't using an rbtree, it was storing the pointer >> value in a *radix* which is why single lookups took so long. Presumably >> radix was used for RCU lookups. >> >> Your ring->id trick lets us use RCU with small ints instead of the >> context pointer. It might be worth using idr instead of rolling manual >> array code. It'd still be much faster than the list, but get rid of the >> large alloc, array walking, memcpy(), etc. >> > > I picked up Ben's patch and incorporated Zach's idea and the first > results look promising, as expected. I am going to do a full test with > the same workload I've used for rbtree and come back with the results > and the patch in a day or so. > Unfortunately, I still see performance degradation for the one ioctx case even when using IDR. I am using the same fio benchmark as before. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/