16 Dec
2024
16 Dec
'24
2:48 a.m.
well after Unix had fledged, its developers at CSRC
found it necessary
and/or desirable to borrow back a Multics concept: they named it mmap().
As far as I know no Research version of Unix ever had mmap.
Multics had a segmented universal memory. A process incorporated
segments into its address space The universal memory was normally
addressed via a hierachical segment-name directory. With enhancement
to provide for multisegment "files", the directory could serve as a file
system and file I/O became data transfer between segments.
Unix originally imitated the Multics file system, but not the universal
memory. mmap(2) weakly imitates universal memory by allowing a process
to nominally incorporate a portion of a file into the process address space
at page-level granularity. However, an update is guaranteed to be visible
to the file and other processes only upon specific request.
Does anyone know whether there are implementations of mmap that
do transparent file sharing? It seems to me that should be possible by
making the buffer cache share pages with mmapping processes.
Doug
16 Dec
16 Dec
2:57 a.m.
New subject: SCCS, TeamWare, BitKeeper, and Git
On Sun, Dec 15, 2024 at 03:48:40PM -0500, Douglas McIlroy wrote:
Unix originally imitated the Multics file system, but
not the universal
memory. mmap(2) weakly imitates universal memory by allowing a process
to nominally incorporate a portion of a file into the process address space
at page-level granularity. However, an update is guaranteed to be visible
to the file and other processes only upon specific request.
That's not true with Sun's mmap(). It's coherent across process
boundaries
and it's in sync with read()/write(). This is because Sun got rid of the
buffer cache and _only_ did file IO to/from the page cache. You mapped
the actual page into your address space, if one process wrote it, the
other process will see it.
Does anyone know whether there are implementations of
mmap that
do transparent file sharing? It seems to me that should be possible by
making the buffer cache share pages with mmapping processes.
SunOS 4.0 did what I believe you are asking.
http://mcvoy.com/lm/papers/SunOS.vm_arch.pdf
http://mcvoy.com/lm/papers/SunOS.vm_impl.pdf
--
---
Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat
5:05 a.m.
New subject: mmap, was SCCS, TeamWare, BitKeeper, and Git
It appears that Douglas McIlroy <douglas.mcilroy(a)dartmouth.edu> said:
Unix originally imitated the Multics file system, but
not the universal
memory. mmap(2) weakly imitates universal memory by allowing a process
to nominally incorporate a portion of a file into the process address space
at page-level granularity. However, an update is guaranteed to be visible
to the file and other processes only upon specific request.
Does anyone know whether there are implementations of mmap that
do transparent file sharing? It seems to me that should be possible by
making the buffer cache share pages with mmapping processes.
These days they all do. The POSIX rationale says:
A memory object can be concurrently mapped into the address space of one or more
processes. The mmap( ) and munmap( ) functions allow a process to manipulate their
address
space by mapping portions of memory objects into it and removing them from it. When
multiple processes map the same memory object, they can share access to the underlying
data.
"Memory object" includes disk files.
There are MAP_SHARED and MAP_PRIVATE flags that say whether changes are written
through or copy-on-write to local private pages.
R's,
John
15
days inactive
15
days old
2 comments
3 participants
participants (3)
-
Douglas McIlroy -
John Levine -
Larry McVoy