|
From: | Ognyan Kulev |
Subject: | Re: [PATCH] ext2fs and large stores (> 1.5G) |
Date: | Wed, 30 Apr 2003 20:43:24 +0300 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3) Gecko/20030327 Debian/1.3-4 |
Marcus Brinkmann wrote:
I don't think it is possible to force GNU Mach to drop a page or reload it, once it is loaded it is in the page cache until GNU Mach decides to evict it on its own (due to memory pressure). Thomas, am I right?
It seems that I'll do some rewriting soon :-)
That's why both schemes proposed by Thomas and Roland did not change the meaning of a mapped page once it is mapped: In Thomas' model, ranges of the partition would be mapped, and there would be a cache of such mappings. If you need different areas of the partition, and the cache is full, you would drop a whole unused range by unmapping it and mapping the different area. In Roland's model, the whole metadata is collocated into a single (or several) mapped area, and the the mapping never changes.
The location of indirect blocks is unpredictable so I'll try "Thomas' model" for these. Mapping metadata at fixed location will remain as it is implemented now.
Regards -- Ognyan Kulev <ogi@fmi.uni-sofia.bg>, "\"Programmer\""
[Prev in Thread] | Current Thread | [Next in Thread] |