Messages in this thread | | | Date | Sun, 2 Jul 2000 20:53:34 -0300 (BRT) | From | Marcelo Tosatti <> | Subject | Re: [PATCH] 2.2.17pre9 interactiveness under high IO |
| |
On 3 Jul 2000, Juan J. Quintela wrote:
> >>>>> "marcelo" == Marcelo Tosatti <marcelo@conectiva.com.br> writes: > > marcelo> On Mon, 3 Jul 2000, Jens Axboe wrote: > >> On Sat, Jul 01 2000, Marcelo Tosatti wrote: > >> > Hi, > >> > > >> > This silly patch avoids that by making ll_rw_block() ignore the IO if the > >> > request queue is full, causing the process to search for more freeable > >> > pages instead blocking. > >> > > >> > --- fs/buffer.c.orig Sat Jul 1 20:49:19 2000 > >> > +++ fs/buffer.c Sat Jul 1 20:28:36 2000 > >> > @@ -1517,7 +1517,7 @@ > >> > if (wait) > >> > __wait_on_buffer(p); > >> > } else if (buffer_dirty(p)) > >> > - ll_rw_block(WRITE, 1, &p); > >> > + ll_rw_block(WRITEA, 1, &p); > >> > } while (tmp != bh); > >> > > >> > do { > >> > >> This is wrong -- if the buffer is not queued because we are out of > >> free request slots, it has still already been locked at this point > >> (and _Req). > > marcelo> if (!req) { > marcelo> if (rw_ahead) > marcelo> goto end_io; > marcelo> req = __get_request_wait(max_req, bh->b_rdev); > marcelo> } > > marcelo> In case we fail to find a free request, bh->b_end_io is called, and > marcelo> it unlocks the buffer. > > Think what can happen if bh->b_end_io is set as _async handler, it > wolud unlock the buffer and more things that you don't want posibly > for it to do.
It never happens. We try to sync buffers which are on the page->buffers list, and those buffers use the sync handler. The async handler is only used for page IO.
- 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/
| |