Hi, not that I would not have other things that I would be more keen on, but did yet again some more FORMAT updates tonight. Check
http://www.coli.uni-sb.de/~eric/stuff/soft/by-others/ format-0.91o.zip (136136 bytes)... Changes: - progress indicator for scanning of existing FAT and for mirroring now uses percentage display (less output) - added another check for Win9x: Root directory chain starts at cluster 2 if Win9x suggests cluster 0 (nothing). I really hope that this finally allows FORMAT to create working FAT32 filesystems *even* under Win9x kernels... - no longer trying to find existing FAT32 format on floppy - faster mirror function: now using multi sector I/O useful for both floppy (slow seek) and FAT32 (huge partititons) - the "allow ESC at last moment if /D mode and floppy" is now suppressed by /Y (more consistent, /Y is "no prompts" mode) I hope that the better speed (floppy format detection and saving of unformat information in (default) safe format mode should be faster now) was worth the efforts. I wonder how fast FORMAT is now. You cannot really measure that in DOSEMU because DOSEMU is surrounded by powerful Linux disk caches :-). On the other hand, I have no real test harddisks around. If you need information on how to create a FAT32 diskimage for DOSEMU, feel free to contact me. In short: mkfatimage16, then use an hex editor (or xxd, joe and maybe dd and of course xxd -r) to modify the virtual MBR to select FAT32 LBA partition type (0b, instead of the default 06 "FAT16 > 32 MB") and finally format the thing with FORMAT in DOSEMU (it might help to zap the 55 aa magic of the virtual FAT16 boot sector before). However, the most exciting thing is of course the new Win98 compatibility patch! If you run FORMAT 0.91o under a Win9x kernel, FORMAT should now be able to sanitize 1. the FAT1x FAT size (must be 0) 2. the FAT1x root dir size (must be 0) 3. the FAT32 root dir position (must be 2 for us) 4. the FAT32 info sector position (must be 1 for us) 5. the backup offset (must be -1 if feature unused, not 0) 6. [whatever I forgot to mention], finally enabling it to create SANE FAT32 filesystems even under Win9x kernel. Thanks to Bernd for sending me a captured (SAVEBS-C) boot sector and DOSFSCK -v and FORMAT /D logs which allowed me to find (I hope) all values which FORMAT 0.91 had initialized the wrong way until now. Note that running FORMAT under a FreeDOS kernel has been working since 0.91k or similar, although the kernel had bugs before version 2033 which could lead to inefficient FAT32 parameters. If everything except 720k floppy lowlevel format works now, maybe THIS version gets the chance to become called 0.92 soon? At least it will be a nice version for FreeDOS beta9rc4, if you ask me :-). Testers, please test... Brian, if tests are okay, who will have the honour to declare this FORMAT to be 0.92 ? Should I stay maintainer of the 0.92 series? I suggest that yes - this would allow you to focus on FDISK, which should have quite a few bug reports checked. Eric. ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Freedos-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-devel
