(clip)
>
> Does this noatime patch really break POSIX conformance? I.e. does it really
> not update the atimes at all? If it doesn't, wouldn't it be better to put
> the atime updates in the kernel's disk-write cache and not flush the cache
> if there are only atime updates to be flushed?
>
(clip)
As usual, my suggestion is to restructure all sorts of things I don't
understand....
Maybe you could store atimes in a compressed sense; Make a struct with just
atimes and inode numbers, and pack them into a page of memory. You could
stuff a lot of atimes away before they took up much space.
Even assuming you put -half- of the memory to pointers for some kind of data
structure (much more than necessary), you could look at 7,000 files and
still only have 1 meg of atimes 'constipated away'.
__kmb203@psu.edu_________________________Debian__1.1___Linux__2.0.0___
You have new mail in /dev/null