>Bryan Henderson wrote: >>>You want to *use* the kernel pagecache as much as you can. >> >> >> No, I really don't. Not always. I can think of only 2 reasons to >> maximize my use of the kernel pagecache: 1) saves me duplicating code; 2) >> allows me to share resources (memory and disk bandwidth come to mind) with >> others in the same Linux system fairly. There are many cases where those >> two benefits are outweighed by the benefits of using some other cache. If >> you're thinking of other benefits of using the pagecache, let's hear them. > >You forgot the third reason (benefit), though it doesn't directly >related to the page cache: read ahead. It greatly influences on the >performance and a direct I/O application has to reimplement this logic, >which generally isn't straightforward task.
Actually, that's the first reason I gave. The readahead that the page cache does can be duplicated, and sometimes far exceeded, by a special purpose user space cache. As you say, it isn't straightforward, so saving that duplication is a good reason to use the page cache. Using the page cache is clearly better than using no cache at all (we're including readahead and writebehind buffers in the term "cache") in nearly all cases. - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
