Messages in this thread | | | Date | Thu, 14 Dec 2000 19:28:04 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: Test12 ll_rw_block error. |
| |
On Thu, 14 Dec 2000, Russell Cattelan wrote: > > So one more observation in > filemap_sync_pte > > lock_page(page); > error = filemap_write_page(page, 1); > -> UnlockPage(page); > This unlock page was removed? is that correct?
Yes. The "writepage" thing changed: "struct file" disappeared (as I'm sure you also noticed), and the page writer is supposed to unlock the page itself. Which it may do at any time, of course.
There are some reasons to do it only after the IO has actually completed: this way the VM layer won't try to write it out _again_ before the first IO hasn't even finished yet, and the writing logic can possibly be simplified if you know that nobody else will be touching that page.
But that is up to you: you can do the UnlockPage before even returning from your "->writepage()" function, if you choose to do so.
Linus
- 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/
| |