From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike McCarty Subject: Re: FORMAT floppy: Drive_IO error Date: Fri, 31 Mar 2006 03:30:04 -0600 Message-ID: <442CF69C.9020805@sbcglobal.net> References: <442B9C5C.1030209@sbcglobal.net> <442C57C7.4020306@pobox.com> <442C60EA.5020407@sbcglobal.net> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-msdos-owner@vger.kernel.org List-Id: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: linux-msdos@vger.kernel.org Bart Oldeman wrote: > On Thu, 30 Mar 2006, Mike McCarty wrote: > >> Alain M. wrote: >> >>> Make the test with the latest version. Eric made a lot of Fixes while >>> I was making tests. It is much better now, specialy in error conditions. >>> >>> But... I made oll tests in real dos... >>> >>> http://www.coli.uni-saarland.de/~eric/stuff/soft/by-others/ >>> >>> Alain >> >> >> Thanks. I downloaded the latest and tried again. It is very >> much faster. It also only found 8 bad sectors (hmm...). But >> the very same IO error occurred. >> >> So, where do I need to report the problem? > > > DOSEMU. It cannot and has never been able to format floppies, to be > precise, BIOS int13/ah=5 is not implemented. I don't believe so, for a number of reasons. First, the disc *was* written, and the % completion went up to 100 percent. Second, the FAT writing was stated to work. Third, the reported error was a WRITE, not a FORMAT TRACK. Fourth, I ran the same program on a pure MSDOS 6.0 machine, under the control of an interrupt capture and analysis program I wrote several years ago, and verified that the FORMAT TRACK calls were made during the time that was reporting % complete. The call which appeared to fail was indeed a BIOS WRITE. Fifth, with a disc which was verified to format properly under MSDOS with the FORMAT.EXE program (latest version) and with MSDOS FORMAT, the program was capable of formatting the disc under DOSEMU. Unfortunately, I was unable to use my interrupt capture program under DOSEMU, as it always reported not enough memory to load the program. Apparently, the BORLANDC link library code for loading/executing a program doesn't work with DOSEMU for some reason. Mike -- p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);} This message made from 100% recycled bits. You have found the bank of Larn. I can explain it for you, but I can't understand it for you. I speak only for myself, and I am unanimous in that!