pHILE+ and FAT filesystems
newsg...@wallfamily.us
We are currently using pHILE format, but I need to implement FAT
for compatibility with PC's, and so far no attempt to format a
device has produced something that pcmount_vol() recognizes.
I've tried both FAT12 and FAT16, as a partitioned drive and as
a superfloppy, but in every case the pcmount_vol() call reads
the first block of the device (block 0), and fails with this error:
#define E_VALIEN 0x2007 /* Wrong volume format */
It might help if I could get a dump of the first few blocks from
someone who has this working...
Thanks...
Dietmar Schindler
Here is the first block of a recognized (CompactFlash) disk:
00A09490 FA BE 00 7C BF 00 7A B9 00 01 FC 0E 1F 0E 07
F3 ...|..z.........
00A094A0 A5 EA 16 7A 00 00 BB BE 7B 33 C9 80 3F 80 75 06 ...z....
{3..?.u.
00A094B0 FE C5 8B F3 EB 07 80 3F 00 75 02 FE C1 83 C3
10 .......?.u......
00A094C0 81 FB FE 7B 72 E5 83 F9 04 74 0B 81 F9 03 01 74 ...
{r....t.....t
00A094D0 0A BB A6 7A EB 2C BB 87 7A EB 27 8B 4C 02 8B
14 ...z.,..z.'.L...
00A094E0 B8 01 02 BB 00 7C CD 13 73 05 BB BE 7A EB 13
2E .....|..s...z...
00A094F0 A1 FE 7D 3D 55 AA 74 05 BB BE 7A EB 05 EA 00
7C ..}=U.t...z....|
00A09500 00 00 2E 8A 07 3C 00 74 0C 53 BB 07 00 B4 0E
CD .....<.t.S......
00A09510 10 5B 43 EB ED EB FE 4E 6F 20 62 6F 6F 74 61 62 .
[C....No bootab
00A09520 6C 65 20 70 61 72 74 69 74 69 6F 6E 20 69 6E 20 le
partition in
00A09530 74 61 62 6C 65 00 49 6E 76 61 6C 69 64 20 50 61
table.Invalid Pa
00A09540 72 74 69 74 69 6F 6E 20 74 61 62 6C 65 00 49 6E rtition
table.In
00A09550 76 61 6C 69 64 20 6F 72 20 64 61 6D 61 67 65 64 valid
or damaged
00A09560 20 42 6F 6F 74 61 62 6C 65 20 70 61 72 74 69 74
Bootable partit
00A09570 69 6F 6E 00 53 79 73 74 65 6D 53 6F 66 74 20 42
ion.SystemSoft B
00A09580 6F 6F 74 20 50 61 72 74 69 74 69 6F 6E 20 53 65 oot
Partition Se
00A09590 63 74 6F 72 00 00 00 00 00 00 00 00 00 00 00 00
ctor............
00A095A0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A095B0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A095C0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A095D0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A095E0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A095F0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09600 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09610 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09620 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09630 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09640 00 00 00 00 00 00 00 00 00 00 00 00 00 00 80
01 ................
00A09650 01 00 06 03 E0 E7 20 00 00 00 E0 F3 01 00 00
00 ...... .........
00A09660 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09670 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 ................
00A09680 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55
AA ..............U.
Let me know if you need more block dumps.
pROBE+> qv
pSOS+/PPC V2.5.4
pROBE+/PPC PS V3.5.1
pROBE+/PPC CE V3.5.1
pROBE+/PPC QS V3.5.1
pROBE+/PPC DI V3.5.1
pHILE+/PPC FS V4.0.0
pREPC+/PPC V2.5.2
pNA+/PPC V4.2.0.9
pMONT+/PPC V1.5.3
newsg...@wallfamily.us
the os_de_read() call was swapping bytes, so the signature was reading
as aa55 instead of 55aa.
Can I ask how you created this boot block? I can only seem to get
"superfloppy" format working (no partitions - the whole device is the
filesystem), and your dump looks like a standard boot block, not
superfloppy.
Perhaps the "boot" flag is critical... I'll have to give that a
try...
- spw
Dietmar Schindler
> Can I ask how you created this boot block? I can only seem to get
> "superfloppy" format working (no partitions - the whole device is the
> filesystem), and your dump looks like a standard boot block, not
> superfloppy.
We use the flash disks as they come preformatted or reformatted under
windows. Only FAT16 format seems to work. We had some problems in the
driver initialization distinguishing between partitioned and
unpartitioned disks, but now both work. At initialization (when the
device is opened) the driver builds a table of each partition's start
sector and number of sectors (with only one table entry in case of a
superfloppy covering the whole disk); when the device is read or
written, depending on the minor device number, the b_blockno from the
struct ioparms is offset by the corresponding start sector.
I hope this helps.
spw
- spw