From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!not-for-mail From: Eli Zaretskii Newsgroups: gmane.emacs.devel Subject: Re: Pure space overflow (Cygwin) Date: Sat, 20 Oct 2012 10:34:07 +0200 Message-ID: <83d30dzgn4.fsf@gnu.org> References: <837gqm22ol.fsf@gnu.org> <50821050.3050000@cornell.edu> Reply-To: Eli Zaretskii NNTP-Posting-Host: plane.gmane.org X-Trace: ger.gmane.org 1350722078 25771 80.91.229.3 (20 Oct 2012 08:34:38 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Sat, 20 Oct 2012 08:34:38 +0000 (UTC) Cc: Angelo.Graziosi@roma1.infn.it, emacs-devel@gnu.org To: Ken Brown Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Sat Oct 20 10:34:45 2012 Return-path: Envelope-to: ged-emacs-devel@m.gmane.org Original-Received: from lists.gnu.org ([208.118.235.17]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1TPUW2-0007Ge-AJ for ged-emacs-devel@m.gmane.org; Sat, 20 Oct 2012 10:34:42 +0200 Original-Received: from localhost ([::1]:48656 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TPUVu-0006zh-Nr for ged-emacs-devel@m.gmane.org; Sat, 20 Oct 2012 04:34:34 -0400 Original-Received: from eggs.gnu.org ([208.118.235.92]:37038) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TPUVs-0006z7-5i for emacs-devel@gnu.org; Sat, 20 Oct 2012 04:34:33 -0400 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TPUVr-0005yj-0A for emacs-devel@gnu.org; Sat, 20 Oct 2012 04:34:32 -0400 Original-Received: from mtaout22.012.net.il ([80.179.55.172]:51323) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TPUVq-0005wB-OG for emacs-devel@gnu.org; Sat, 20 Oct 2012 04:34:30 -0400 Original-Received: from conversion-daemon.a-mtaout22.012.net.il by a-mtaout22.012.net.il (HyperSendmail v2007.08) id <0MC600300NNZH000@a-mtaout22.012.net.il> for emacs-devel@gnu.org; Sat, 20 Oct 2012 10:34:17 +0200 (IST) Original-Received: from HOME-C4E4A596F7 ([87.69.4.28]) by a-mtaout22.012.net.il (HyperSendmail v2007.08) with ESMTPA id <0MC60036UNT51B80@a-mtaout22.012.net.il>; Sat, 20 Oct 2012 10:34:17 +0200 (IST) In-reply-to: <50821050.3050000@cornell.edu> X-012-Sender: halo1@inter.net.il X-detected-operating-system: by eggs.gnu.org: Solaris 10 (beta) X-Received-From: 80.179.55.172 X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Original-Sender: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Xref: news.gmane.org gmane.emacs.devel:154459 Archived-At: > Date: Fri, 19 Oct 2012 22:45:36 -0400 > From: Ken Brown > CC: Angelo Graziosi , emacs-devel@gnu.org > > > I bumped the base value by 80000, so it should solve your problem. > > Perhaps Ken should look into tuning SYSTEM_PURESIZE_EXTRA (on > > conf_post.h) for Cygwin, since no other system seems to have this > > problem at this time. > > I could certainly do this, although there's no need at the moment after > your increase. Or are you suggesting that I revert your change and then > tune SYSTEM_PURESIZE_EXTRA for Cygwin? Reverting the change won't tell you enough detail to set the extra puresize correctly, because you don't know how close to the limit are Unix and GNU platforms, which is what the BASE_PURESIZE is set for. What you need is compare the values of pure-bytes-used in "emacs -Q" between Emacs compiled on Cygwin and on some Unix or GNU system. (Make sure both builds use the same configuration options and are done on platforms with the same size of basic data types and of EMACS_INT, which probably means a 32-bit Unix or GNU system.) Then subtract these two values and round up. Another alternative is to run "temacs -batch -l loadup" under GDB, set a breakpoint in pure_alloc, and record every allocation that comes from Cygwin-specific code; summing them up should give you the extra. But this is much more tedious, as you'd have to examine lots of allocations.