Date: Thu, 19 Aug 1999 10:50:34 +0200 From: Laszlo Molnar To: Eli Zaretskii Cc: Martin Stromberg , DJGPP-WORKERS , peuha AT cc DOT helsinki DOT fi, pausmith AT nortelnetworks DOT com Subject: Re: Make 3.78 is in pretest Message-ID: <19990819105034.E14930@crater.eth.ericsson.se> References: <199908181226 DOT OAA24678 AT spica DOT lu DOT erisoft DOT se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.6i In-Reply-To: ; from Eli Zaretskii on Wed, Aug 18, 1999 at 05:13:42PM +0300 Reply-To: djgpp-workers AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk On Wed, Aug 18, 1999 at 05:13:42PM +0300, Eli Zaretskii wrote: > > This gives some clues on what Eli said. The exit code is _supposed_ to be > > in the upper eigth bits in perl. > If this is ``standard Perl'', then I'd say the DJGPP port has a bug, > because the command that deletes a file had its status NOT shifted 8 bits > to the left; see my description of problems with features/errors test. > > Laszlo, can you please comment on this? Seems like a bug to me. The exit code is only set correctly when perl's system() is used. In the case of backticks popen() is used by perl and it doesn't set the exit code correctly. I've found that this affects some other ports too. I'll fix it, and release an updated perl port with djgpp 2.03. Laszlo