When retrieving a requeued IO, don't setup data pointers again

This is a bug, if we had a residual data count we would be
retrying the full request again.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

diff --git a/io_u.c b/io_u.c
index 7ac31a0d297872fe82168e74cff63cfce1880dcb..faa1c5e36a8d84ac558d6867a7e796ec5a71d8fe 100644 (file)
--- a/io_u.c
+++ b/io_u.c
@@ -764,10 +764,9 @@ struct io_u *get_io_u(struct thread_data *td)
         * Set io data pointers.
         */
        io_u->endpos = io_u->offset + io_u->buflen;
-out:
        io_u->xfer_buf = io_u->buf;
        io_u->xfer_buflen = io_u->buflen;
-
+out:
        if (!td_io_prep(td, io_u)) {
                fio_gettime(&io_u->start_time, NULL);
                return io_u;