From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24213 invoked by alias); 19 Dec 2001 00:23:35 -0000 Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner@cygwin.com Received: (qmail 24170 invoked from network); 19 Dec 2001 00:23:35 -0000 Received: from unknown (HELO antiochus-fe0.ultra.net) (146.115.8.188) by sources.redhat.com with SMTP; 19 Dec 2001 00:23:35 -0000 Received: from enterprise-e.Starfleet.com (209-122-233-115.s1925.apx2.sbo.ma.dialup.rcn.com [209.122.233.115]) by antiochus-fe0.ultra.net (8.8.8/ult/n20340/mtc.v2) with ESMTP id TAA19449 for ; Tue, 18 Dec 2001 19:23:22 -0500 (EST) Message-Id: <4.3.1.2.20011218191628.0234ae38@pop.ma.ultranet.com> X-Sender: lhall@pop.ma.ultranet.com X-Mailer: QUALCOMM Windows Eudora Version 4.3.1 Date: Tue, 18 Dec 2001 16:29:00 -0000 To: cygwin@cygwin.com From: "Larry Hall (RFK Partners, Inc)" Subject: RE: cygwin1.dll (file version 1003.6.0.0) dup problem with gcc In-Reply-To: <718198F862F1D411B10F0002A50A4DB101A4F0E8@e90wwce3.dx.deere .com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-SW-Source: 2001-12/txt/msg00908.txt.bz2 At 02:37 PM 12/17/2001, Polley Christopher W wrote: > > > it. If you can, debugging Cygwin via gdb would be the best way to go. > > > Alternatively, if you can't do that, calling gcc indirectly through > > > strace in your IDE environment would provide some trace information > > > > > Here is a strace snippet right around the error's occurrence: > > 09:19:17 [main] make 384 dtable::build_fhandler: fd 3, fh 0x61550820 > 09:19:17 [main] make 384 dtable::build_fhandler: fd 4, fh 0x61550960 > 09:19:17 [main] make 384 fhandler_base::init: created new >fhandler_base for handle 0xD8 > 09:19:17 [main] make 384 fhandler_base::init: created new >fhandler_base for handle 0x154 > 09:19:17 [main] make 384 make_pipe: 0 = make_pipe ([3, 4], 16384, >0x10000) > 09:19:17 [main] make 384 _close: close (4) > 09:19:17 [main] make 384 fhandler_base::close: closing '(null)' >handle 0x154 > 09:19:17 [main] make 384 dtable::not_open: in not_open, >fds=0x61550128 fd=4 fds[fd]=1632962912 res=0 size=32 > 09:19:17 [main] make 384 _close: 0 = close (4) > 09:19:17 [main] make 384 fhandler_base::set_close_on_exec: set >close_on_exec for (null) to 1 > 09:19:17 [main] make 384 _fcntl: 0 = fcntl (3, 2, 0x1) > 09:19:17 [main] make 384 set_process_mask: old mask = 0, new mask = >1804007 > 09:19:17 [main] make 384 dtable::not_open: in not_open, >fds=0x61550128 fd=0 fds[fd]=1632961224 res=0 size=32 > 09:19:17 [main] make 384 dtable::build_fhandler: fd -1, fh >0x61550960 > 09:19:17 [main] make 384 fhandler_base::dup: in fhandler_base dup > 09:19:17 [main] make 384 fhandler_base::dup: dup(unknown disk file) >failed, handle 0, Win32 error 6 > 09:19:17 [main] make 384 seterrno_from_win_error: >/usr/src/cygwin/winsup/cygwin/fhandler.cc:926 errno 6 > 09:19:17 [main] make 384 geterrno_from_win_error: windows error 6 == >errno 9 > 09:19:17 [main] make 384 vfork: 0 = vfork() > 09:19:17 [main] make 384 vfork: exiting vfork, res 0 > > I'm not sure how far back is relevant to the vfork's setting up for >the make-file's "echo" command that occurs next vs. cleaning up from >checking dependencies (the previous task make was doing) > > the build_fhandler: fd=-1 is expected; that's a new dtable being >built, so aside from that the error occurring in fhandler_base::dup (which >we already knew) there's no new info. > > Larry, would posting a complete strace be useful? What might I look >for in the strace? It sounds to me like you've been debugging the code. strace output is sometimes a helpful resource when doing this but only to the extent that it leads you to the problem in the source. If you've spent the time poking around in the debugger, I'll wager you've amassed more information about the problem than strace will ever give you. I'd recommend continuing down the debugging avenue you're following. Don't let my comments about strace output sidetrack you. Larry Hall lhall@rfk.com RFK Partners, Inc. http://www.rfk.com 838 Washington Street (508) 893-9779 - RFK Office Holliston, MA 01746 (508) 893-9889 - FAX -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/