Why shm fs (Was [2.3.51pre2] wrong shm_statfs in ipc/shm.c)

From: Christoph Rohland (hans-christoph.rohland@sap.com)
Date: Fri Mar 10 2000 - 11:02:03 EST


Hi Rui,

Rui Sousa <rsousa@grad.physics.sunysb.edu> writes:
> And why do we need another virtual file system?
> I don't remember seeing any discussion about it and
> I'm just curious...

There was some discussion about it and I posted the shm fs patch on
lkm some weeks before.

The shm fs makes posix shm possible. The API for posix shm is much
more fitting in the Unix way then SYSV IPC. It also is much more
flexible.

In short you have the following calls for posix shm:
- shm_open gives you a file descriptor to a shm object
- you can manipulate this file descriptor with ftruncate, fstat and
  mmap like any other file descriptor
- shm_unlink removes the object

shm_open and shm_unlink directly map to open/unlink in the shm fs.
Also the internal handling fits now much better into the rest of the
mm code.

Greetings
                Christoph

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:18 EST