lkml.org 
[lkml]   [2000]   [Oct]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: New Benchmark tools, lookie looky........
On Tue, 17 Oct 2000, Malcolm Beattie wrote:

> Larry McVoy writes:
> > On Tue, Oct 17, 2000 at 09:21:00AM -0700, Andre Hedrick wrote:
> > > Expand 'traces' ... O-SCOPE analyizer?
> >
> > Insert a ring buffer into the disk sort entry point. Add a userland process
> > which reads this ring buffer and gets the actual requests in the actual order
> > they are sent to the drive[s]. Then take that data and write a simulator into
> > which you can plug in different algs. I have all this crud for SunOS if you
> > want it, including elevator.c, hacksaw.c, and inorder.c.
>
> I wrote a lightweight kernel->userland ring buffer device for Linux
> called bufflink and a block-request logger that uses it called
> reqlog. reqlog writes a structure
> struct reqlog {
> unsigned int major;
> unsigned int minor;
> unsigned long sector;
> long nr_sectors;
> };
> to the ring buffer when an ioctl is done to enable logging. The
> current patch isn't quite what you were suggesting since it does
> roughly
>
> add_request() {
> ...
> elevator_queue(req, tmp, latency, dev, current_request);
> + if (bl_reqlog && enable_reqlog) {
> + ...
> + bufflink_append(bl_reqlog, (unsigned char *)&rl, sizeof rl);
> + }
>
> if (queue_new_request)
> (dev->request_fn)();
> }
> but it would be easy to write the record (instead or as well)
> before the elevator_queue(). The patches are available from
> http://users.ox.ac.uk/~mbeattie/linux-kernel.html

Hi Malcolm,

Since what I am doing is building a raw request, I still get into the
queue as not to distrub things, but I can stomp all over the queue if
needed. So you are pointing to adding your ring buffer at the request
entry level or at the execution point?

Sorry for being so stupid, I never venture to high in the FS layer.

Cheers,

Andre Hedrick
The Linux ATA/IDE guy

-
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/

\
 
 \ /
  Last update: 2005-03-22 12:41    [from the cache]
©2003-2011 Jasper Spaans