linux-kernel - Re: [PATCH 1/2] vhost test module

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20101202231818.GA12362@redhat.com>
Date:	Fri, 3 Dec 2010 01:18:18 +0200
From:	"Michael S. Tsirkin" <mst@...hat.com>
To:	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Cc:	virtualization@...ts.linux-foundation.org, kvm@...r.kernel.org,
	rusty@...tcorp.com.au, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] vhost test module

On Thu, Dec 02, 2010 at 03:13:03PM -0800, Paul E. McKenney wrote:
> On Thu, Dec 02, 2010 at 09:47:09PM +0200, Michael S. Tsirkin wrote:
> > On Thu, Dec 02, 2010 at 11:26:16AM -0800, Paul E. McKenney wrote:
> > > On Thu, Dec 02, 2010 at 09:11:30PM +0200, Michael S. Tsirkin wrote:
> > > > On Thu, Dec 02, 2010 at 11:00:37AM -0800, Paul E. McKenney wrote:
> > > > > On Mon, Nov 29, 2010 at 07:09:01PM +0200, Michael S. Tsirkin wrote:
> > > > > > This adds a test module for vhost infrastructure.
> > > > > > Intentionally not tied to kbuild to prevent people
> > > > > > from installing and loading it accidentally.
> > > > > > 
> > > > > > Signed-off-by: Michael S. Tsirkin <mst@...hat.com>
> > > > > 
> > > > > On question below.
> > > > > 
> > > > > > ---
> > > > > > 
> > > > > > diff --git a/drivers/vhost/test.c b/drivers/vhost/test.c
> > > > > > new file mode 100644
> > > > > > index 0000000..099f302
> > > > > > --- /dev/null
> > > > > > +++ b/drivers/vhost/test.c
> > > > > > @@ -0,0 +1,320 @@
> > > > > > +/* Copyright (C) 2009 Red Hat, Inc.
> > > > > > + * Author: Michael S. Tsirkin <mst@...hat.com>
> > > > > > + *
> > > > > > + * This work is licensed under the terms of the GNU GPL, version 2.
> > > > > > + *
> > > > > > + * test virtio server in host kernel.
> > > > > > + */
> > > > > > +
> > > > > > +#include <linux/compat.h>
> > > > > > +#include <linux/eventfd.h>
> > > > > > +#include <linux/vhost.h>
> > > > > > +#include <linux/miscdevice.h>
> > > > > > +#include <linux/module.h>
> > > > > > +#include <linux/mutex.h>
> > > > > > +#include <linux/workqueue.h>
> > > > > > +#include <linux/rcupdate.h>
> > > > > > +#include <linux/file.h>
> > > > > > +#include <linux/slab.h>
> > > > > > +
> > > > > > +#include "test.h"
> > > > > > +#include "vhost.c"
> > > > > > +
> > > > > > +/* Max number of bytes transferred before requeueing the job.
> > > > > > + * Using this limit prevents one virtqueue from starving others. */
> > > > > > +#define VHOST_TEST_WEIGHT 0x80000
> > > > > > +
> > > > > > +enum {
> > > > > > +	VHOST_TEST_VQ = 0,
> > > > > > +	VHOST_TEST_VQ_MAX = 1,
> > > > > > +};
> > > > > > +
> > > > > > +struct vhost_test {
> > > > > > +	struct vhost_dev dev;
> > > > > > +	struct vhost_virtqueue vqs[VHOST_TEST_VQ_MAX];
> > > > > > +};
> > > > > > +
> > > > > > +/* Expects to be always run from workqueue - which acts as
> > > > > > + * read-size critical section for our kind of RCU. */
> > > > > > +static void handle_vq(struct vhost_test *n)
> > > > > > +{
> > > > > > +	struct vhost_virtqueue *vq = &n->dev.vqs[VHOST_TEST_VQ];
> > > > > > +	unsigned out, in;
> > > > > > +	int head;
> > > > > > +	size_t len, total_len = 0;
> > > > > > +	void *private;
> > > > > > +
> > > > > > +	private = rcu_dereference_check(vq->private_data, 1);
> > > > > 
> > > > > Any chance of a check for running in a workqueue?  If I remember correctly,
> > > > > the ->lockdep_map field in the work_struct structure allows you to create
> > > > > the required lockdep expression.
> > > > 
> > > > We moved away from using the workqueue to a custom kernel thread
> > > > implementation though.
> > > 
> > > OK, then could you please add a check for "current == custom_kernel_thread"
> > > or some such?
> > > 
> > > 							Thanx, Paul
> > 
> > It's a bit tricky. The way we flush out things is by an analog of
> > flush_work. So just checking that we run from workqueue isn't
> > right need to check that we are running from one of the specific work
> > structures that we are careful to flush.
> > 
> > I can do this by passing the work structure pointer on to relevant
> > functions but I think this will add (very minor) overhead even when rcu
> > checks are disabled. Does it matter? Thoughts?
> 
> Would it be possible to set up separate lockdep maps, in effect passing
> the needed information via lockdep rather than via the function arguments?
> 
> 							Thanx, Paul

Possibly, I don't know enough about this but will check.
Any examples to look at?

-- 
mST
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/