Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin-developers AT sources DOT redhat DOT com Date: Thu, 22 Mar 2001 16:12:41 -0500 From: Christopher Faylor To: Egor Duda , cygwin-apps AT cygwin DOT com Subject: Re: bug in ash (Was: State of the DLL, mark 2?) Message-ID: <20010322161241.A18144@redhat.com> Reply-To: cygwin-apps AT cygwin DOT com Mail-Followup-To: Egor Duda , cygwin-apps AT cygwin DOT com References: <20010319203650 DOT A30559 AT redhat DOT com> <907012433 DOT 20010320162719 AT logos-m DOT ru> <20010320091254 DOT F32706 AT redhat DOT com> <2710465067 DOT 20010320172451 AT logos-m DOT ru> <20010320101715 DOT K32706 AT redhat DOT com> <1983285237 DOT 20010321133832 AT logos-m DOT ru> <15325749005 DOT 20010322235224 AT logos-m DOT ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.11i In-Reply-To: <15325749005.20010322235224@logos-m.ru>; from deo@logos-m.ru on Thu, Mar 22, 2001 at 11:52:24PM +0300 On Thu, Mar 22, 2001 at 11:52:24PM +0300, Egor Duda wrote: >well, i've found the reason. moreover, i've found that as was once >patched to fix this bug, but later, patch was reverted. i wonder why. >did it have some ill side effects? Hmm. I vaguely remember this. I remember asking Corinna to look into this but neither of us had an actual test case that tickled the bug. I can't find any email from Corinna explaining why the bug was backed out but I wonder if I asked her to do this in a phone call or something. Have you tried doing a full configure/build with the ash as modified below? If so, maybe it is safe to check this in. >============================================================================= >Thu Oct 12 14:28:00 2000 Corinna Vinschen > > * memalloc.c: Revert previous patch until it's need is proved. > >Tue Oct 10 16:46:00 2000 Corinna Vinschen > > Patch suggested by Andy Mortimer : > * memalloc.c: Add member `refcnt' to struct stack_block. > (stalloc): Initialize refcnt when allocating a new block. > (setstackmark): Increment refcnt when reusing a stack block. > (popstackmark): Decrement refcnt when dropping usage of a stack block. > (growstackblock): Reallocate a block only if it's used only once. >========================================================================= I will wait for Corinna to come back from holiday so that she can offer the definitive opinion, though. Thanks for tracking this down. I was going to look into this myself this weekend. You saved me from some "enjoyable" ash hacking. :-) cgf