X-Recipient: archive-cygwin AT delorie DOT com X-Spam-Check-By: sourceware.org Date: Fri, 1 Jul 2011 13:44:09 +0200 From: Corinna Vinschen To: cygwin AT cygwin DOT com Subject: Re: Vim segv'ing Message-ID: <20110701114409.GQ9552@calimero.vinschen.de> Reply-To: cygwin AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com References: <20110630072012 DOT GB9552 AT calimero DOT vinschen DOT de> <2BF01EB27B56CC478AD6E5A0A28931F202CFAFF7 AT A1DAL1SWPES19MB DOT ams DOT acs-inc DOT net> <20110630142353 DOT GH9552 AT calimero DOT vinschen DOT de> <20110630150502 DOT GJ9552 AT calimero DOT vinschen DOT de> <20110701083603 DOT GM9552 AT calimero DOT vinschen DOT de> <20110701094339 DOT GO9552 AT calimero DOT vinschen DOT de> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com On Jul 1 11:21, Andy Koppe wrote: > On 1 July 2011 10:43, Corinna Vinschen wrote: > > On Jul  1 10:36, Corinna Vinschen wrote: > >> FYI, I tracked it down to the place where the stack overwrite occurs. > >> This is most puzzeling.  When typing :wq!, the following chain of functions > >> is called: > >> > >>   nv_colon > >>    do_cmdline > >>     ex_exit > >>      do_write > >>       open                           <- Here it calls into the Cygwin DLL > >>        fhandler_base::open_with_arch > >>         fhandler_base::open_fs > >>          fhandler_base::open > >>           NtCreateFile               <--Here it calls into NTDLL.DLL > >> > >> The open call tries to open the backup file "/etc/hosts~", not the > >> symlink itself. > >> > >> In the optimized version of vim, the local variable "cap" in the > >> function nv_colon is kept in register $esi.  When do_cmdline is called, > >> $esi is pushed onto the stack.  Then everything goes its normal ways, > >> until NtCreateFile is called. > >> > >> And here's the puzzler: This call to NtCreateFile overwrites the 4 byte > >> stack slot in which the "cap" pointer is saved with the value 0x10c! > > What else has access to that? The signal thread? No, it's just the call stack of the main thread. cap is a local variable of a vim function and the location is just the one cap gets pushed to to be restored when returning from do_cmdline. It's so far away from the NtCreateFile call, it's no concern to anybody at that point. > > And it also does not happen if I build Cygwin with gcc 4.5.1 rather than > > with gcc 4.3.4.  Is it possible that a compiler bug is playing a role > > here?  I can't quite believe it, the effect is too specific. > > This sounds similar to some unexplained mintty crashes that Chris > Sutcliffe and I have seen with gcc 4.3.4, triggered by setup.exe > running postinstall scripts or vim accessing /etc/fstab. Unfortunately > we never managed to pin down the actual cause, inspite of a fair bit > of debugging and staring at the compiler's assembly output. The > crashes went away with changes that should have made no difference. > > See here: > http://code.google.com/p/mintty/source/detail?r=928 > http://code.google.com/p/mintty/source/detail?r=894 > http://groups.google.com/group/mintty-discuss/browse_thread/thread/7f1afcb74c61fe99 Hmm, that's really strange... Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple