Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com Date: Wed, 05 Dec 2001 15:33:50 +0100 From: "Yllman, Jens" Subject: Re: cygwin 1.3.6 + bash 2.05a-2 + command completion To: cygwin AT cygwin DOT com Cc: Robert Collins , "Yllman, Jens" Message-id: <3C0E304E.481984EB@uniweb.se> Organization: Uniweb AB MIME-version: 1.0 X-Mailer: Mozilla 4.78 [en] (WinNT; U) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT X-Accept-Language: en References: <3C0DE9BF DOT 42B4B14D AT uniweb DOT se> <03ee01c17d7e$1071a880$0200a8c0 AT lifelesswks> <3C0E0C98 DOT D7C7501C AT uniweb DOT se> <043501c17d85$26e70040$0200a8c0 AT lifelesswks> <3C0E156B DOT A76309D3 AT uniweb DOT se> <000d01c17d8b$45e47210$0200a8c0 AT lifelesswks> Hi, I tried to take a shot at this. But I did not manage to compile the cygwin source. I'm not sure how to do that. I also failed to compile the cygcheck program to find a possible missfeature in that program. And at the moment I don't have the time to find out what is going wrong with that. But as has shown is that it is the traversing of the path. That is trying to find files in paths that does not exist. I guess that it has something to do with opendir/readdir. If I have time tonight I'll try to compare diffrences between 1.3.5-3 and 1.3.6 source. But if I interperet the ChangeLog correctly the changes on the opendir/readdir part is quite large. Jens Yllman Robert Collins wrote: > > FWIW I think that cygwin is dying, not bash. > > However, as I haven't looked at the code.... > > Rob > === > ----- Original Message ----- > From: "Yllman, Jens" > To: "Robert Collins" > Cc: > Sent: Wednesday, December 05, 2001 11:39 PM > Subject: Re: cygwin 1.3.6 + bash 2.05a-2 + command completion > > > Hi, > > > > The ChangeLog reports the following. > > > > 2001-11-24 Christopher Faylor > > > > * fhandler_disk_file.cc (fhandler_cygdrive::readdir): Avoid reporting > > inaccessible drives. > > > > Maybe bash expect to get something like invalid drive or something. > > I've not checked bash. And I've not traced anything. Just this quick > > look at the ChangeLog. > > > > There is also other notes about fhandler_disk_file.cc in the > ChangeLog. > > > > * fhandler_disk_file.cc (fhandler_disk_file::opendir): Set nohandle > > flag on dummy fd. > > > > I don't know if anybody can look into it. I don't have any more time. > > > > About the reason for me testing with sh also and not only bash was to > > see that it realy is bash handling of tab that crashes and not the > stty > > settings or something like that. > > > > Jens Yllman > > > > Robert Collins wrote: > > > > > > === > > > ----- Original Message ----- > > > From: "Yllman, Jens" > > > > Hi, > > > > > > > > Yes, I have a /cygdrive/x in my path that is not always maped. > When I > > > > map it it works fine. > > > > > > Ok, in my mind that conforms the way to reproduce. I don't have time > at > > > the moment (sigh, I wish I did, I really do) to debug this myself > > > though. > > > > > > Rob ------------------------------------------------------------ Uniweb AB Phone: +46 8 626 42 00 P O Box 745 FAX: +46 8 626 42 01 S-191 27 SOLLENTUNA SWEDEN http://www.uniweb.se/ -- 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/