From mboxrd@z Thu Jan 1 00:00:00 1970 From: Junio C Hamano Subject: Re: support for large packs and 64-bit offsets Date: Mon, 09 Apr 2007 13:18:05 -0700 Message-ID: <7vwt0lxpxe.fsf@assigned-by-dhcp.cox.net> References: <11760951973172-git-send-email-nico@cam.org> <20070409171925.GS5436@spearce.org> <20070409174305.GU5436@spearce.org> <7vtzvpz5tu.fsf@assigned-by-dhcp.cox.net> <20070409195322.GB5436@spearce.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Nicolas Pitre , git@vger.kernel.org To: "Shawn O. Pearce" X-From: git-owner@vger.kernel.org Tue Apr 10 01:33:13 2007 Return-path: Envelope-to: gcvg-git@gmane.org Received: from vger.kernel.org ([209.132.176.167]) by lo.gmane.org with esmtp (Exim 4.50) id 1Hb0JY-0005zn-Px for gcvg-git@gmane.org; Mon, 09 Apr 2007 22:18:13 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752997AbXDIUSJ (ORCPT ); Mon, 9 Apr 2007 16:18:09 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753000AbXDIUSJ (ORCPT ); Mon, 9 Apr 2007 16:18:09 -0400 Received: from fed1rmmtao105.cox.net ([68.230.241.41]:45979 "EHLO fed1rmmtao105.cox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752999AbXDIUSH (ORCPT ); Mon, 9 Apr 2007 16:18:07 -0400 Received: from fed1rmimpo02.cox.net ([70.169.32.72]) by fed1rmmtao105.cox.net (InterMail vM.7.05.02.00 201-2174-114-20060621) with ESMTP id <20070409201807.XNIW25613.fed1rmmtao105.cox.net@fed1rmimpo02.cox.net>; Mon, 9 Apr 2007 16:18:07 -0400 Received: from assigned-by-dhcp.cox.net ([68.5.247.80]) by fed1rmimpo02.cox.net with bizsmtp id l8J51W00Y1kojtg0000000; Mon, 09 Apr 2007 16:18:05 -0400 In-Reply-To: <20070409195322.GB5436@spearce.org> (Shawn O. Pearce's message of "Mon, 9 Apr 2007 15:53:22 -0400") User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/21.4 (gnu/linux) Sender: git-owner@vger.kernel.org Precedence: bulk X-Mailing-List: git@vger.kernel.org Archived-At: "Shawn O. Pearce" writes: > Sure, fine. But I think you missed my point above - right now if > we move the SHA-1 table out of the .idx file I'm not sure we know > how to support the dumb clients *at all*. Even if they understand > the latest-and-greatest file formats... We do an incremental .keep pack for packs 100 objects or more. If .idx omits SHA-1 values but keeps the crc and offset, that would be around 12 bytes per object (but you may need an index into the real SHA-1 table in the .pack file, I dunno) so that would be 1200 bytes. If we duplicate SHA-1 table also in .idx that would make it 32-byte per object, totalling 3200 bytes, which admittedly is near 3-fold increase, but it may be worth if we want to avoid the hassle.