Subject: Re: fsync performance hit on 1.6.1
To: Greg A. Woods <firstname.lastname@example.org>
From: Christoph Hellwig <email@example.com>
Date: 07/09/2003 09:07:03
On Wed, Jul 09, 2003 at 03:01:30AM -0400, Greg A. Woods wrote:
> a shared memory "object". To quote the rationale from P1003.1-2001:
> On implementations where memory objects are implemented using the
> existing file system, the shm_open() function may be implemented
> using a macro that invokes open(), and the shm_unlink() function
> may be implemented using a macro that invokes unlink().
> There had to have been some pretty strange politics going on to have
> forced the creation of the POSIX shared memory objects API even when the
> old POSIX mmap() was already a well known option!
Umm, posix SHM _does_ use mmap. It just uses shm_open to get a suitable
fd, on Solaris and Linus that would be on tmpfs.