Messages in this thread | | | Date | Sun, 27 Jun 1999 09:54:09 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: [patch] pagecache-2.3.9-H3, bmap & ext2fs cleanup patch |
| |
On Sun, 27 Jun 1999, Ingo Molnar wrote: > > > > this will obviously fail if holes are a little bit different and are not > > > uptodate anymore. > > > > "obviously fail" HOW? > > i'll have to change the above to do something like: > > was_hole = 0; > if (!buffer_mapped(bh)) { > if (buffer_uptodate(bh) && !buffer_dirty(bh)) > was_hole = 1;
Why?
Holes aren't special. They shouldn't be. Why do you CARE whether it was a hole or not?
You should not. I really don't see your point at all. Your changes make the path more complex for absolutely no gain that I can tell.
> eventually holes are _not_ kept uptodate anymore. They are not kept > uptodate in this case because i'm trying to do an optimization of 'late > clearing' of holes, ie. 'late uptodate' (in fact never) holes.)
I think your whole optimization is a pessimization, and makes the normal case go slower with no win that I can see.
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |