At 10:41 PM 7/22/2004 +0200, Eric Auer wrote:
Nice. One thing which I remember: XTree file manager had INT 3 (1 byte opcode) as replacement for INT 21, both to make the code smaller and make debugging harder. So that falls into the category (why NEWDES?).
NEWDES appears to be a case of the developer forgetting to throw the right compiler switch and leaving in a bunch of INT 3's for the release version. They don't appear to do anything either useful or nasty.
Yes! Me... Tyler and me figured out that there is some incompatibility between "LBAcache loaded into EMM386-generated UMBs" and "SCSI drive detection". Symptom: Scanning through drives 0x80..0x87 takes much longer than usual and in the end LBAcache will conclude that there are no harddisks.
Much of the port I/O must be emulated in V86 mode with EMM386 via exception, which is dog slow comparatively speaking. DMA I/O is emulated on INT 13 and INT 40h via HLT instruction which triggers code to copy buffers back and forth via a REP MOVSD. As a consequence disk access will be slower: imperceptibly to moderately depending on what's happening..
So Lynx does not work... Any hope for volunteers to compile Links / ELinks in DJGPP for DOS?
I'll take anything which is pre-packaged to work, or close to, for trial. Lynx assumes a dedicated individual will set it up for their machine; I'm not that dedicated.
GEOS Ensemble Lite - can you explain in what way it is FreeDOS incompatible?
Kicks out an error message, even if HIMEM/EMM386 not loaded. Error is it can't find a directory for fonts, though the directory is present. I'm assuming problem is Ensemble depends on an exact internal setup, possibly related to FindFirst/FindNext function, which is not 100% duplicated from FreeDOS vs. MS-DOS. Could be that all the extra DTA fields added in DOS 7.10, which FreeDOS reports as the current OS version, are checked. Maybe it works in 2035, as it appears I'm still using 2034, but I doubt it.
And MSD ...: The programs which included the evil "pretend DR DOS is not compatible" AARD test in Win3.1 were: MN, SETUP, HIMEM, SMARTDRV, MSD and WIN.
MSD is an interesting case. I can break in with the debugger where it goes into an infinite loop checking invalid memory for a zero terminator and if I force it out of that loop, then MSD memory displays do work. It's doing an LES BX from a stack frame, which is dependent on an earlier LES BX, which may be from an earlier one than that. Given the indirect indexing, I'm assuming it happens due to list of lists, which can have multiple levels of indirection or DOS UMB interface. Right before the infinite loop, MSD calls set allocation strat function to get at UMB's and does call list of lists function, so quite possibly the problem lies in one of those areas. And may be fixed in 2035, too.
------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click _______________________________________________ Freedos-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-devel
