From mboxrd@z Thu Jan 1 00:00:00 1970 From: Oleg Drokin Subject: Re: Copy time comparison 2.4.20-pre6 <-> 2.4.19+data-logging (was:Compatibility of current 2.4.19.pending ...) Date: Fri, 13 Sep 2002 11:37:35 +0400 Message-ID: <20020913113735.A6022@namesys.com> References: <3D7FDEA5.7040806@mb.tu-ilmenau.de> <3D81076E.6090408@mb.tu-ilmenau.de> Mime-Version: 1.0 Return-path: list-help: list-unsubscribe: list-post: Errors-To: flx@namesys.com Content-Disposition: inline In-Reply-To: <3D81076E.6090408@mb.tu-ilmenau.de> List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Manuel Krause Cc: reiserfs-list Hello! On Thu, Sep 12, 2002 at 11:30:22PM +0200, Manuel Krause wrote: > >When I see the discussion about the new block-allocator, the testing > >directory etc. I assume a definite "No." (Please, please, correct me, if > ;-) You really didn't correct me so far. Mmh. :-( Datalogging is incompatible with stuff in testing directory. It won't apply cleanly to new block allocatior, but that's can be easily fixed. > >So, for the overnight change to pre-kernel 7, where are the new > >data-logging patches Hans even announced to lkml? > So I really got a 2.4.20-pre6 compiled with the > preempt-kernel-rml-2.4.20-pre5-1.patch and Olegs fix he sent to the list > yesterday. BTW, have the testing patches from -pre5 gone into -pre6? It > isn't clear to me. Testing patches were merged into -pre6 by Marcelo's mistake and -pre7 have these patches removed already. > Application startup timings of NS-7.0 and OOo-1.0.1 didn't increase. :-)) Sure, read code path did not changed at all. > I hope I don't need to comment on the values... Sure. datalogging and new file_write code are not solutions to same problem. These are solutions for different problems and can be used together (after merging of course) for even better performance, I hope ;) Bye, Oleg