Re: [PATCH linux-2.6.0-test10-mm1] filemap_fdatawait.patch

[Suparna Bhattacharya]

On Wed, Dec 31, 2003 at 02:10:42AM -0800, Andrew Morton wrote:
> Suparna Bhattacharya <suparna@xxxxxxxxxx> wrote:
> >
> >  > OK, but the thread which is running fdatawrite/fdatawait isn't interested
> >  > in that page, because it must have been dirtied _after_ this thread has
> >  > passed through filemap_fdatawrite(), yes?
> > 
> >  Not exactly. The page could actually have been dirtied _before_ this 
> >  thread passes through filemap_datawrite, but is just being parallely 
> >  written back by a background thread.
> 
> I still don't see it.  If the page was dirtied before this thread entered
> filemap_fdatawrite() then this thread will either start writeout
> immediately, or will wait on the writeout and start new writeout.  And
> the page lock avoids the timing window which you have mentioned.

>From filemap_fdatawait:

while (!list_empty(&mapping->locked_pages)) {
                struct page *page;
                                                                                
                page = list_entry(mapping->locked_pages.next,struct page,list);
                list_del(&page->list);
                if (PageDirty(page))
                        list_add(&page->list, &mapping->dirty_pages);
                else
                        list_add(&page->list, &mapping->clean_pages);
                                                                                
---->           if (!PageWriteback(page)) {
		                        if (++progress > 32) {
                                if (need_resched()) {
                                        spin_unlock(&mapping->page_lock);
                                        __cond_resched();
                                        goto restart;
                                }
                        }
---->                  continue;
                }
	
If PG_writeback is not set, it moves on to process the next page -- without
falling through to wait_on_page_writeback, isn't that so ? 
Another thread could have pulled the page off io_pages, put it on locked
pages, and not set PG_writeback as yet.

How does the page lock guard against this ? 
filemap_fdatawait seems to ignore the page lock.

Regards
Suparna

-- 
Suparna Bhattacharya (suparna@xxxxxxxxxx)
Linux Technology Center
IBM Software Lab, India

-
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/