31 Jan
2022
31 Jan
'22
1:43 p.m.
From: Paul Koning
When did Unix first get partitions?
'Partitions' the mechanism, or partitions the term for the mechanism?
The former appeared about V5:
https://minnie.tuhs.org/cgi-bin/utree.pl?file=V5/usr/sys/dmr/rp.c
when an RP03 was added; pre-V7, UNIX filesystems were limited to 2^16 blocks,
so the 406*10*20 blocks of an RP03 had to be split up into partitions (called
'sections' or 'pseudo-disks' in the documentation) to make all of it
useable.
The latter? No idea...
Partitions may have appeared in DOS/Windows for much the same reason; with 32
KB clusters, FAT16 filesystems were limited to 2GB. I distinctly recall
having to use partitions when I bought a 13GB hard drive for my Windows 95
machine (FAT32 only came in with Windows 95 OSR2).
Noel
31 Jan
31 Jan
2:21 p.m.
On Mon, 31 Jan 2022, Noel Chiappa via cctalk wrote:
Partitions may have appeared in DOS/Windows for much
the same reason; with 32
KB clusters, FAT16 filesystems were limited to 2GB. I distinctly recall
having to use partitions when I bought a 13GB hard drive for my Windows 95
machine (FAT32 only came in with Windows 95 OSR2).
PC-DOS/MS-DOS early partitioning
When PC-DOS/MS-DOS started to support hard drives (XT/5160, March 8,1983),
with PC/MS-DOS 2.00, the limit was 32MB. NOT 2GB.
That limit called for breaking up drives as small as 40MB into multiple
partitions.
That limit lasted until MS-DOS 3.31 / PC-DOS 4.00 After that, the limit
was bumped up to 2GB. (Probably would have been 4GB if they had used an
UNSIGNED 32 bit number, and given up the option of having negative file
and drive sizes)
The XT/5160 shipped with a 10MB drive, but some people even split that up.
("we're going to share the drive between all of us, so there will be a
partition for DOS and programs, and each person will get a personal
partition for your data."
The XT/5160 used a rebranded Xebec hard disk controller, that included
support for 2 drives, of 5?, 10, 16, and 26MB. It was set for
10MB, but the other sizes could be accessed by adding jumpers to solder
pads on the board. Other sizes could be done with software kludges.
FDISK.COM, for partitioning, was included in PC-DOS 2.00 .
The 5160 shipped with the hard disk already formatted.
However low level format, to redo it, or for a third party drive, required
"Advanced Diagnostics", or third party solutions.
"Advanced Diagnostics" was part of the "Hardware Maintenance And
Service" and was on a 160K floppy AND on a cassette! (for machines without
floppy)
The regular "Diagnostics" that shipped with the machine (both 5150 and
5160) did NOT have hard disk low level format.
Third party controllers usually came with a low-level format utility, or
provision in the hard disk controller's ROM at C800h
--
Grumpy Ol' Fred cisin at xenosoft.com
3:28 p.m.
On 2022-01-31 18:21, Fred Cisin via cctalk wrote:
On Mon, 31 Jan 2022, Noel Chiappa via cctalk wrote:
The other thing you could do is alter the hard disk bios to support the
disk you wanted to use, many years ago I did that for a gentleman that
had a few 42GB disks, and he gave me one of the drives for my efforts.?
I also recall patching AT BIOS to support drive that didn't match the
geometry of drives in the drive table.
Paul.
Partitions may have appeared in DOS/Windows for
much the same reason;
with 32
KB clusters, FAT16 filesystems were limited to 2GB. I distinctly recall
having to use partitions when I bought a 13GB hard drive for my
Windows 95
machine (FAT32 only came in with Windows 95 OSR2).
PC-DOS/MS-DOS early partitioning
When PC-DOS/MS-DOS started to support hard drives (XT/5160, March
8,1983), with PC/MS-DOS 2.00, the limit was 32MB.? NOT 2GB.
That limit called for breaking up drives as small as 40MB into
multiple partitions.
That limit lasted until MS-DOS 3.31 / PC-DOS 4.00? After that, the
limit was bumped up to 2GB.? (Probably would have been 4GB if they had
used an UNSIGNED 32 bit number, and given up the option of having
negative file and drive sizes)
The XT/5160 shipped with a 10MB drive, but some people even split that
up. ("we're going to share the drive between all of us, so there will
be a partition for DOS and programs, and each person will get a
personal partition for your data."
The XT/5160 used a rebranded Xebec hard disk controller, that included
support for 2 drives, of 5?, 10, 16, and 26MB.? It was set for 10MB,
but the other sizes could be accessed by adding jumpers to solder pads
on the board.? Other sizes could be done with software kludges.
FDISK.COM, for partitioning, was included in PC-DOS 2.00 .
The 5160 shipped with the hard disk already formatted. However low
level format, to redo it, or for a third party drive, required
"Advanced Diagnostics", or third party solutions.
"Advanced Diagnostics" was part of the "Hardware Maintenance And
Service" and was on a 160K floppy AND on a cassette! (for machines
without floppy)
The regular "Diagnostics" that shipped with the machine (both 5150 and
5160) did NOT have hard disk low level format.
Third party controllers usually came with a low-level format utility,
or provision in the hard disk controller's ROM at C800h
--
Grumpy Ol' Fred???????????? cisin at xenosoft.com 
10:19 p.m.
On Mon, Jan 31, 2022 at 02:21:19PM -0800, Fred Cisin via cctalk wrote:
[...]
That limit lasted until MS-DOS 3.31 / PC-DOS 4.00
After that, the limit
was bumped up to 2GB. (Probably would have been 4GB if they had used an
UNSIGNED 32 bit number, and given up the option of having negative file
and drive sizes)
Fun factoid: that 2GiB limit on *filesystem* size is actually due to the use
of an signed *eight* bit field, namely the number of sectors in a cluster.
This limits clusters to 64 sectors, or 32kiB. FAT16 supports a shade under
2**16 clusters, resulting in that 2GiB limit.
FAT's *file* size limitation is indeed due to a 32 bit field. The ISO 9660
standard offers an "interesting" solution to that, namely having multiple
directory entries for the same filename. So if you want to store files
larger than 4GiB on a CD-ROM, the filesystem won't hold you back.
1 Feb
1 Feb
11:09 a.m.
On Tue, 1 Feb 2022, Peter Corlett via cctalk wrote:
FAT's *file* size limitation is indeed due to a 32
bit field. The ISO 9660
standard offers an "interesting" solution to that, namely having multiple
directory entries for the same filename. So if you want to store files
larger than 4GiB on a CD-ROM, the filesystem won't hold you back.
Multiple DIRectory entries ("extents") for the same file was also how CP/M
handled "large" files.
And, how Windoze stores "long filenames", although that's just using up
DIRectory space for the filenames, with a pointer to the "real" entry.
31 Jan
31 Jan
2:24 p.m.
On 1/31/22 2:43 PM, Noel Chiappa via cctalk wrote:
Partitions may have appeared in DOS/Windows for much
the same reason;
with 32 KB clusters, FAT16 filesystems were limited to 2GB.
I think the same type of problem happened with the really old FAT-12 to
FAT-16 conversion.
Wasn't FAT-12 limited to something near 31 MB?
--
Grant. . . .
unix || die
2:56 p.m.
On Mon, 31 Jan 2022, Fred Cisin via cctalk wrote:
On Mon, 31 Jan 2022, Grant Taylor via cctalk wrote:
*chases Fred out of the house with a broom, screaming incoherently about
SI units*
g.
--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.
ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://scarlet.deltasoft.com - Get it _today_!
I think the same type of problem happened with
the really old FAT-12 to
FAT-16 conversion.
Wasn't FAT-12 limited to something near 31 MB?
32 MebiBytes - 1
3:19 p.m.
> On Mon, 31 Jan 2022, Grant Taylor via cctalk
wrote:
>> I think the same type of problem happened with the really old FAT-12 to
>> FAT-16 conversion.
>> Wasn't FAT-12 limited to something near 31 MB?
On Mon, 31 Jan 2022, Fred Cisin via cctalk wrote:
> 32 MebiBytes - 1
On Mon, 31 Jan 2022, geneb wrote:
*chases Fred out of the house with a broom, screaming
incoherently about SI
units*
Sorry that I was in the house with a broom, screaming incoherently about
SI units.
I think that the correct number is 33,554,432 bytes
File sizes were stored as the last 4 bytes of each DIRectory entry, with a
signed 32 bit number. So, a file could be from
-2,147,483,648 to 2,147,483,647
Unfortunately, they never successfully finished the implementation,
so copying a negative size file to the drive did not increase the free
space. Since they didn't make THAT work, perhaps they should have used an
UNSIGNED 32 bit number, to permit files up to 4,294,967,295
The 5160/XT original drive was 306 cylinders, with 4 heads, and 17 sectors
per track for 10,653,696
for 10.653696 "MegaBytes" 1,000,000 10^6
or 10.404 Marketing "MegaBytes" 1,024,000 10^3 * 2^10
or 10.16015625 MebiBytes 2^20
Seagate, and a few others, were smart enough to call that 10MB, so that
people forgave them a few bad tracks, and even bragged how great the drive
was because it was even more than the ad had called it!
--
Grumpy Ol' Fred cisin at xenosoft.com
8:53 p.m.
On Mon, 31 Jan 2022, Fred Cisin via cctalk wrote:
File sizes were stored as the last 4 bytes of each
DIRectory entry, with a
signed 32 bit number. So, a file could be from
-2,147,483,648 to 2,147,483,647
Unfortunately, they never successfully finished the implementation, so copying
a negative size file to the drive did not increase the free space. Since they
didn't make THAT work, perhaps they should have used an UNSIGNED 32 bit
number, to permit files up to 4,294,967,295
Consider the file seek operation however: with the file size limited to
the 31-bit positive range of the signed 32-bit data type you can move the
file offset backwards and forwards through the entire range with a single
operation using the same data type. It would not be possible if the data
type was unsigned and all the 32 bits used for the file size.
Maciej
4:30 p.m.
On Mon, Jan 31, 2022, 2:43 PM Noel Chiappa via cctalk <cctalk at classiccmp.org wrote:
SunOS 1.0 had on disk tables that described the partitions in 1982. These
were dynamic per disk as opposed to the hardwired in the os drivers that V5
and later had. I've not had luck with finding people to recall where this
feature came from...
Partitions may have appeared in DOS/Windows for much the same reason; with
From: Paul
Koning
When did Unix first get partitions?
'Partitions' the mechanism, or partitions the term for the mechanism?
The former appeared about V5:
https://minnie.tuhs.org/cgi-bin/utree.pl?file=V5/usr/sys/dmr/rp.c
when an RP03 was added; pre-V7, UNIX filesystems were limited to 2^16
blocks,
so the 406*10*20 blocks of an RP03 had to be split up into partitions
(called
'sections' or 'pseudo-disks' in the documentation) to make all of it
useable.
The latter? No idea...
32
KB clusters, FAT16 filesystems were limited to 2GB. I distinctly recall
having to use partitions when I bought a 13GB hard drive for my Windows 95
machine (FAT32 only came in with Windows 95 OSR2).
There were similar tables on DEC Rainbow disks which predated the MS DOS
port to the platform and was a different format. This was also 1981 or
1982. 3rd party disks had their own tables. I've not done the deep dive to
date it more specifically. Other CP/M raw disks from this era have similar
tables on them, but I have only what I've puzzled out, no docs.
Warner
1 Feb
1 Feb
6:32 a.m.
On Mon, Jan 31, 2022 at 4:43 PM Noel Chiappa via cctalk <
cctalk at classiccmp.org> wrote:
In the Unix V5 rp(4) man page the term "sections" is used. See page 222 of
https://www.tuhs.org/Archive/Distributions/Research/Dennis_v5/v5man.pdf
In "Setting Up Unix - Sixth Edition" the term "partitions" is used,
although the rp(4) man page still used "sections" in V6 and V7. See
https://www.tuhs.org/cgi-bin/utree.pl?file=V6/usr/doc/start/start
-ken
From: Paul
Koning
When did Unix first get partitions?
'Partitions' the mechanism, or partitions the term for the mechanism?
The former appeared about V5:
https://minnie.tuhs.org/cgi-bin/utree.pl?file=V5/usr/sys/dmr/rp.c
when an RP03 was added; pre-V7, UNIX filesystems were limited to 2^16
blocks,
so the 406*10*20 blocks of an RP03 had to be split up into partitions
(called
'sections' or 'pseudo-disks' in the documentation) to make all of it
useable.
The latter? No idea...
815
days inactive
816
days old
11 comments
9 participants
participants (9)
-
abuse＠cabal.org.uk -
cctalk＠gtaylor.tnetconsulting.net -
cisin＠xenosoft.com -
geneb＠deltasoft.com -
imp＠bsdimp.com -
jnc＠mercury.lcs.mit.edu -
kgober＠gmail.com -
macro＠orcam.me.uk -
phb.hfx＠gmail.com