| PLEX86 | ||
|
Linnut screen shots 10761
Linnut screen shots 10762 Yes, but why does it need to be in the kernel? Since you seem to like marketing bullpoo, here's a blurb on... Yes you miss something. Not everything that is treated like a file is actually a bunch of inodes on a disk. It is quite possible to use the file descriptor mechanism, and thus the already existing vfs code, to address regions in a memory cache. Of course, sometimes those two mechanisms address practically the same thing, like when an actual file on disk is cached in memory. In both cases there are valid reasons to re-use the filesystem code. The one that springs to mind immediately is that new code is untested code, and therefore adapting existing, already debugged code, to a new purpose leads to a more stable system. Linnut screen shots 10763 OK poked his little head through the XP firewall and said: You have some real... Now, when comparing, which programmers like to re-use existing code, and which programmers just *love* to implement new APIs as completely new code? How many times have we heard 'total rewrite!' coming out of Redmond as if it's the next best thing since sliced bread?1 Mart 1 Of course, the obverse, trying to maintain old crufty code instead of rewriting it, is just as bad. -- "We will need a longer wall when the revolution comes." --- AJS, quoting an uncertain source.
|
||||||||