Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759822AbYFYQDX (ORCPT ); Wed, 25 Jun 2008 12:03:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754109AbYFYQDJ (ORCPT ); Wed, 25 Jun 2008 12:03:09 -0400 Received: from fxip-0047f.externet.hu ([88.209.222.127]:43830 "EHLO pomaz-ex.szeredi.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752559AbYFYQDI (ORCPT ); Wed, 25 Jun 2008 12:03:08 -0400 To: johnpol@2ka.mipt.ru CC: miklos@szeredi.hu, jens.axboe@oracle.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, torvalds@linux-foundation.org, akpm@linux-foundation.org, hugh@veritas.com, nickpiggin@yahoo.com.au In-reply-to: <20080625154713.GA18682@2ka.mipt.ru> (message from Evgeniy Polyakov on Wed, 25 Jun 2008 19:47:13 +0400) Subject: Re: [patch 1/2] mm: dont clear PG_uptodate in invalidate_complete_page2() References: <20080625124038.103406301@szeredi.hu> <20080625124121.839734708@szeredi.hu> <20080625131117.GA28136@2ka.mipt.ru> <20080625141654.GA4803@2ka.mipt.ru> <20080625154713.GA18682@2ka.mipt.ru> Message-Id: From: Miklos Szeredi Date: Wed, 25 Jun 2008 18:02:43 +0200 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1076 Lines: 25 > > > Is this nfs/fuse problem you described: > > > http://marc.info/?l=linux-fsdevel&m=121396920822693&w=2 > > > > Yes. > > I do not know fuse good enough, but it looks like if your patch only > fixes issue for pages which are in splice buffer during invalidation, > any subsequent call for splice() will get correct new data (at least > invoke readpage(s)), but in the description of error there was a > long timeout between reading and writing, so it was a fresh splice() > call, which suddenly started to return errors. > > Is it possible that by having uptodate bit in place, but page was > actually freed but not removed from all trees and so on, so it > masked read errors? This may be not the right conclusion though :) No, the mechanics of the problem are well understood. Only the solution is problematic :) Miklos -- 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/