On Monday, July 28, 2003, at 03:01 PM, Quinn wrote: At 16:42 -0700 21/7/03, raldrich@mac.com wrote: I would, however, like to find some documentation for "cluster_pagein" and "cluster_pageout" - an overview of the unified buffer cache would be nice as well... If your file system supports caching, the cluster_xxx calls are the way to go. All you have to do is direct your VOP_READ/WRITE/PAGEIN/PAGEOUT calls to the corresponding cluster_xxx calls, implement VOP_CMAP and VOP_STRATEGY and all the hard work is done. This is how both HFS Plus and AppleShare work (at least, for AppleShare, in the non-cached case) and it makes the read/write path very easy. I've gotten cluster I/O working, and it works quite well - really simplified the read/write path, although I suspect that it's less optimal than my previous code (it did make debugging memory mapped I/O a lot easier). Thanks for the pointers, Ron Aldrich Software Architects, Inc. _______________________________________________ darwin-kernel mailing list | darwin-kernel@lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.