As a test, here's the same experiment with a 30MB disk image. Again, it doesn't work with T2505 USB "Full" but does work with T2505 USB "Lite":
> $ qemu-img create -f qcow2 small.qcow2 30M ..booting from T2505 USB "Full": > C:\>fdisk 2 /auto > C:\>fdisk 2 /info > > Current fixed disk drive: 2 60480 sectors, geometry 60/016/63 > > Partition Status Mbytes System Usage Start CHS End > CHS > D: 1 4 A 29 FAT-16 <32M 100% 0/001/01 > 59/015/63 > > Largest continuous free space for primary partition = 0 MBytes ..rebooting and FORMAT: > C:\>format D: /D > > [DEBUG] FORMAT 0.92a, selected drive D: > [DEBUG] Sector buffer at 15E1:BD3C, track buffer at 15E1:673C > [DEBUG] DOS 7+, LOCKing drive > > WARNING: ALL DATA ON NON-REMOVABLE DISK > DRIVE D: WILL BE LOST! PLEASE CONFIRM! > Proceed with Format (YES/NO)?YES > [DEBUG] Checking whether UNFORMAT data can be saved. > [DEBUG] Test-reading boot sector. > +*** bad sector(s): 0 (code 0x201) on READ * > maybe FAT32? > **** bad sector(s): 0 (code 0x20C) on READ * > [DEBUG] Error code: 0x20C > Boot sector unreadable, disk not yet formatted > [DEBUG] DOS 7+, FORMAT-LOCKing drive > [DEBUG] FAT1x size: 59, using FAT16. > Disk size: 29 Mbytes, FAT16. *** > [DEBUG] Enabled access flags. > Cannot find existing format - not saving UNFORMAT data. > Please enter volume label (max. 11 chars): SMALL > QuickFormatting (only flushing metadata) > Warning: Resets bad cluster marks if any. > [DEBUG] File System Creation > [DEBUG] Boot Sector -> 0 > +**Drive_IO(WRITE 0, count 1 ) [FAT12/16] [drive D*] > > Critical error during DOS disk access > DOS driver error (hex): 01 > Description: unknown unit for driver > Program terminated. > [DEBUG] DOS 7+, UNLOCKing drive (by one level) > [DEBUG] DOS 7+, UNLOCKing drive (by one level) And trying again using T2505 USB "Lite": (recreated a new 30MB disk image) > C:\>fdisk 2 /auto > C:\>fdisk 2 /info > > Current fixed disk drive: 2 60480 sectors, geometry 60/016/63 > > Partition Status Mbytes System Usage Start CHS End > CHS > D: 1 4 A 29 FAT-16 <32M 100% 0/001/01 > 59/015/63 > > Largest continuous free space for primary partition = 0 MBytes ..and rebooting and FORMAT: > C:\>format D: /D > > [DEBUG] FORMAT 0.92a, selected drive D: > [DEBUG] Sector buffer at 15E1:BD3C, track buffer at 15E1:673C > [DEBUG] DOS 7+, LOCKing drive > > WARNING: ALL DATA ON NON-REMOVABLE DISK > DRIVE D: WILL BE LOST! PLEASE CONFIRM! > Proceed with Format (YES/NO)?YES > [DEBUG] Checking whether UNFORMAT data can be saved. > [DEBUG] Test-reading boot sector. > **** bad sector(s): 0 (code 0x810C) on READ * > maybe FAT32? > **** bad sector(s): 0 (code 0x20C) on READ * > [DEBUG] Error code: 0x20C > Boot sector unreadable, disk not yet formatted > [DEBUG] DOS 7+, FORMAT-LOCKing drive > [DEBUG] FAT1x size: 59, using FAT16. > Disk size: 29 Mbytes, FAT16. *** > [DEBUG] Enabled access flags. > Cannot find existing format - not saving UNFORMAT data. > Please enter volume label (max. 11 chars): > QuickFormatting (only flushing metadata) > Warning: Resets bad cluster marks if any. > [DEBUG] File System Creation > [DEBUG] Boot Sector -> 0 > [DEBUG] FAT Sectors: 1 to 118 -> > 0% 17% 35% 53% 71% 88% 89% 90% 91% 92% 93% 94% 95% 96% 97% > 98% > 99% 100% > [DEBUG] FAT1 Start -> 1 > [DEBUG] FAT2 Start -> 60 > [DEBUG] Root Directory Sectors: 119 to 150 ... done. > [DEBUG] Root, Label, Timestamp -> 119 > [DEBUG] File System Created > > QuickFormat complete. > [DEBUG] DOS 7+, UNLOCKing drive (by one level) > [DEBUG] DOS 7+, UNLOCKing drive (by one level) > > 30,933,504 bytes total disk space (disk size) > 30,855,168 bytes available on disk (free clusters) > > 2,048 bytes in each allocation unit. > 15,066 allocation units on disk. > > Volume Serial Number is 121F-1EEA On Sat, May 3, 2025 at 5:54 PM Jim Hall <jh...@freedos.org> wrote: > > Hi Eric > > On Sat, May 3, 2025 at 5:47 PM Eric Auer via Freedos-devel > <freedos-devel@lists.sourceforge.net> wrote: > > > > > > Hi Jim, > > > > can you check which differences can be found in the FORMAT > > log when formatting does succeed? If I understand you > > correctly, things depend on from which drive you boot > > and / or which KERNEL version you used? > > Not correct. The kernel version is the same between FreeDOS 1.4 and > FreeDOS T2505, so this isn't a kernel issue. > > FORMAT is the same version (0.92a) in both FreeDOS 1.4 and FreeDOS > T2505, so this isn't a FORMAT regression. > > > A FAT12 partition with 10k sectors is quite small, do you > > get similar problems for FAT16 < 32 MB partition type? Or > > is this a FAT12 specific issue, affecting others < 32 MB? > > I haven't tried other disk sizes (yet). I only needed a 5MB disk > image, so that's the one I used (and found the issue). > > > Does the problem depend on which version of the kernel > > and/or format you use, in other words, could this be a > > regression? Do things work with EDR-DOS or MS kernels? > > I haven't tried it with MS-DOS (and I don't have a copy of EDR-DOS) > but the FreeDOS kernel is the same version in FreeDOS 1.4 and T2505, > and the FORMAT version is the same in FreeDOS 1.4 and T2505. _______________________________________________ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel
