netdev - Re: [PATCH 2/3] virtio: indirect ring entries (VIRTIO_RING_F_INDIRECT

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: <200905172346.14498.rusty@rustcorp.com.au>
Date:	Sun, 17 May 2009 23:46:12 +0930
From:	Rusty Russell <rusty@...tcorp.com.au>
To:	Avi Kivity <avi@...hat.com>
Cc:	Mark McLoughlin <markmc@...hat.com>, dlaor@...hat.com,
	netdev@...r.kernel.org, Dor Laor <dor@...hat.com>,
	virtualization@...ts.linux-foundation.org
Subject: Re: [PATCH 2/3] virtio: indirect ring entries (VIRTIO_RING_F_INDIRECT_DESC)

On Sun, 17 May 2009 03:57:01 pm Avi Kivity wrote:
> Rusty Russell wrote:
> > +static void adjust_threshold(struct vring_virtqueue *vq,
> > +			     unsigned int out, unsigned int in)
> > +{
> > +	/* There are really two species of virtqueue, and it matters here.
> > +	 * If there are no output parts, it's a "normally full" receive queue,
> > +	 * otherwise it's a "normally empty" send queue. */
>
> This comment is true for networking, but not for block. ++overkill with
> a ->adjust_threshold op.

No, it's true for block.  It has output parts, so we should reduce threshold 
when it's full.  Network recvq is an example which should reduce threshold 
when it's empty.

->adjust_threshold is better as an arg to vring_new_virtqueue, but it's still 
not clear what the answer would be.

Rusty.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html