Expand on fill_device option since it apparently causes confusion
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
diff --git a/HOWTO b/HOWTO
index 68e17e9..e8fbd97 100644
--- a/HOWTO
+++ b/HOWTO
@@ -378,7 +378,11 @@
fill_device=bool Sets size to something really large and waits for ENOSPC (no
space left on device) as the terminating condition. Only makes
sense with sequential write. For a read workload, the mount
- point will be filled first then IO started on the result.
+ point will be filled first then IO started on the result. This
+ option doesn't make sense if operating on a raw device node,
+ since the size of that is already known by the file system.
+ Additionally, writing beyond end-of-device will not return
+ ENOSPC there.
blocksize=int
bs=int The block size used for the io units. Defaults to 4k. Values
diff --git a/fio.1 b/fio.1
index d372727..a6547ee 100644
--- a/fio.1
+++ b/fio.1
@@ -235,7 +235,9 @@
Sets size to something really large and waits for ENOSPC (no space left on
device) as the terminating condition. Only makes sense with sequential write.
For a read workload, the mount point will be filled first then IO started on
-the result.
+the result. This option doesn't make sense if operating on a raw device node,
+since the size of that is already known by the file system. Additionally,
+writing beyond end-of-device will not return ENOSPC there.
.TP
.BI filesize \fR=\fPirange
Individual file sizes. May be a range, in which case \fBfio\fR will select sizes