Messages in this thread | | | Date | Fri, 22 Dec 2000 17:52:28 -0200 (BRST) | From | Marcelo Tosatti <> | Subject | Re: [RFC] changes to buffer.c (was Test12 ll_rw_block error) |
| |
On Fri, 22 Dec 2000, Chris Mason wrote:
> > > On Thursday, December 21, 2000 22:38:04 -0200 Marcelo Tosatti > <marcelo@conectiva.com.br> wrote: > > >> Marcelo Tosatti writes: > >> > It seems your code has a problem with bh flush time. > >> > > >> > In flush_dirty_buffers(), a buffer may (if being called from kupdate) > >> > only be written in case its old enough. (bh->b_flushtime) > >> > > >> > If the flush happens for an anonymous buffer, you'll end up writing all > >> > buffers which are sitting on the same page (with > >> > block_write_anon_page), but these other buffers are not necessarily > >> > old enough to be flushed. > >> > > A quick benchmark shows there's room for improvement here. I'll play > around with a version of block_write_anon_page that tries to be more > selective when flushing things out.
There is one more nasty issue to deal with.
You only want to take into account the buffer flushtime if "check_flushtime" parameter is passed as true to flush_dirty_buffers (which is done by kupdate).
Thinking a bit more about the issue, I dont see any reason why we want to write all buffers of an anonymous page at sync_buffers/flush_dirty_buffers.
I think we could simply write the buffer with ll_rw_block() if the page which this buffer is sitting on does not have mapping->a_ops->writepage operation defined.
Chris?
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
| |