Date: Fri, 28 Jun 1996 17:16:00 +0200 Message-Id: <199606281516.RAA12853@tyr.diku.dk> From: Morten Welinder To: djgpp-workers AT delorie DOT com Subject: [DEMMER AT brain1 DOT lstm DOT ruhr-uni-bochum DOT de: select.c] I don't have the time for this, for the time being. somebody record it. Morten ------- Start of forwarded message ------- Return-Path: DEMMER AT brain1 DOT lstm DOT ruhr-uni-bochum DOT de From: "Tom Demmer" Organization: Lehrstuhl Stroemungsmechanik, RUB To: terra AT diku DOT dk Date: Thu, 27 Jun 1996 09:26:27 GMT-1 Subject: select.c Reply-to: Demmer AT LStM DOT Ruhr-Uni-Bochum DOT De Priority: normal Morten, I think in select.c is a bug: If I figure out the code correctly, it goes like this: - Check all fds from 0 to fdmax if they are ready, etc - After that, check via the __file_rec all descriptors that claim to be not ready. The first step handles File System Extension correctly, the second goes wrong for FSEs that got a fd from __FSEXT_alloc_fd. They are connected to DOS's nul device, which is always ready. So, even if a FSE says the file is not ready, the second test will overrule its answer. CIao Tom ****************************************************************** * Thomas Demmer * Phone : +49 234 700 6434 * * Universitaetsstr. 150 * or you try * * Lehrstuhl fuer Stroemungsmechanik * +49 234 700 2896 * * D-44780 Bochum * and MIGHT get someone who * * * knows where I am * * Demmer AT LStM DOT Ruhr-Uni-Bochum DOT De * Fax : +49 234 709-4162 * * http://www.lstm.ruhr-uni-bochum.de/~demmer * ****************************************************************** Nuclear weapons can wipe out life on earth, if used properly -- Talking Heads ------- End of forwarded message -------