Re: [PATCH 10/10] mm: per device dirty threshold

[Miklos Szeredi]

> > No, we _start_ writeback for 1.5*ratelimit_pages pages, but do not
> > wait for those writebacks to finish.
> > 
> > So for a slow device and a fast writer, dirty+writeback can indeed
> > increase beyond the dirty threshold.
> > 
> 
> Nope, try it.
> 
> If a process dirties 1000 pages it'll then go into balance_dirty_pages()
> and start writeback against 1,500 pages.  When we hit dirty_ratio that
> process will be required to write back 1,500 pages for each eight pages
> which it dirtied.  We'll quickly reach the stage where there are no longer
> 1,500 pages to be written back and the process will block in
> balance_dirty_pages() until the dirty+writeback level subsides.

OK.  I was confused by this:

static long ratelimit_pages = 32;

and didn't realize, that that 32 is totally irrelevant.

So I'm still right, that for N dirty pages, the writer is allowed to
dirty N/1500*8 more dirty pages, but I agree, that this isn't really
an issue.

Miklos
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/