netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Dan Jurgens <danielj@nvidia.com>
Cc: netdev@vger.kernel.org, jasowang@redhat.com, pabeni@redhat.com,
	virtualization@lists.linux.dev, parav@nvidia.com,
	shshitrit@nvidia.com, yohadt@nvidia.com,
	xuanzhuo@linux.alibaba.com, eperezma@redhat.com, jgg@ziepe.ca,
	kevin.tian@intel.com, kuba@kernel.org, andrew+netdev@lunn.ch,
	edumazet@google.com
Subject: Re: [PATCH net-next v11 08/12] virtio_net: Use existing classifier if possible
Date: Wed, 19 Nov 2025 01:35:21 -0500	[thread overview]
Message-ID: <20251119013423-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <cb64732c-294e-49e7-aeb5-f8f2f082837e@nvidia.com>

On Wed, Nov 19, 2025 at 12:26:23AM -0600, Dan Jurgens wrote:
> On 11/18/25 3:55 PM, Michael S. Tsirkin wrote:
> > On Tue, Nov 18, 2025 at 08:38:58AM -0600, Daniel Jurgens wrote:
> >> Classifiers can be used by more than one rule. If there is an existing
> >> classifier, use it instead of creating a new one.
> 
> >> +	struct virtnet_classifier *tmp;
> >> +	unsigned long i;
> >>  	int err;
> >>  
> >> -	err = xa_alloc(&ff->classifiers, &c->id, c,
> >> +	xa_for_each(&ff->classifiers, i, tmp) {
> >> +		if ((*c)->size == tmp->size &&
> >> +		    !memcmp(&tmp->classifier, &(*c)->classifier, tmp->size)) {
> > 
> > note that classifier has padding bytes.
> > comparing these with memcmp is not safe, is it?
> 
> The reserved bytes are set to 0, this is fine.

I mean the compiler padding.  set to 0 where?

> > 
> > 
> >> +			refcount_inc(&tmp->refcount);
> >> +			kfree(*c);
> >> +			*c = tmp;
> >> +			goto out;
> >> +		}
> >> +	}
> >> +
> >> +	err = xa_alloc(&ff->classifiers, &(*c)->id, *c,
> >>  		       XA_LIMIT(0, le32_to_cpu(ff->ff_caps->classifiers_limit) - 1),
> >>  		       GFP_KERNEL);
> >>  	if (err)
> > 
> > what kind of locking prevents two threads racing in this code?
> 
> The ethtool calls happen under rtnl_lock.
> 
> > 
> > 
> >> @@ -6932,29 +6945,30 @@ static int setup_classifier(struct virtnet_ff *ff, struct virtnet_classifier *c)
> >>  		      (*c)->size);
> >>  	if (err)
> >>  		goto err_xarray;
> >>  
> >> +	refcount_set(&(*c)->refcount, 1);
> > 
> > 
> > so you insert uninitialized refcount? can't another thread find it
> > meanwhile?
> 
> Again, rtnl_lock.
> 
> 
> >>  
> >>  	err = insert_rule(ff, eth_rule, c->id, key, key_size);
> >>  	if (err) {
> >>  		/* destroy_classifier will free the classifier */
> > 
> > will free is no longer correct, is it?
> 
> Clarified the comment.
> 
> > 
> >> -		destroy_classifier(ff, c->id);
> >> +		try_destroy_classifier(ff, c->id);
> >>  		goto err_key;
> >>  	}
> >>  
> >> -- 
> >> 2.50.1
> > 


  reply	other threads:[~2025-11-19  6:35 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-18 14:38 [PATCH net-next v11 00/12] virtio_net: Add ethtool flow rules support Daniel Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 01/12] virtio_pci: Remove supported_cap size build assert Daniel Jurgens
2025-11-19  7:38   ` Michael S. Tsirkin
2025-11-19 14:23     ` Dan Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 02/12] virtio: Add config_op for admin commands Daniel Jurgens
2025-11-19  7:36   ` Michael S. Tsirkin
2025-11-18 14:38 ` [PATCH net-next v11 03/12] virtio: Expose generic device capability operations Daniel Jurgens
2025-11-18 21:42   ` Michael S. Tsirkin
2025-11-19  3:27     ` Dan Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 04/12] virtio: Expose object create and destroy API Daniel Jurgens
2025-11-18 22:14   ` Michael S. Tsirkin
2025-11-19  3:29     ` Dan Jurgens
2025-11-19  6:39       ` Michael S. Tsirkin
2025-11-19  7:21         ` Dan Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 05/12] virtio_net: Query and set flow filter caps Daniel Jurgens
2025-11-18 22:06   ` Michael S. Tsirkin
2025-11-19  5:57     ` Dan Jurgens
2025-11-18 23:03   ` Michael S. Tsirkin
2025-11-19  4:27     ` Dan Jurgens
2025-11-19  7:53   ` Michael S. Tsirkin
2025-11-19 14:47     ` Dan Jurgens
2025-11-19  7:55   ` Michael S. Tsirkin
2025-11-18 14:38 ` [PATCH net-next v11 06/12] virtio_net: Create a FF group for ethtool steering Daniel Jurgens
2025-11-19  9:36   ` Michael S. Tsirkin
2025-11-19 17:14     ` Dan Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 07/12] virtio_net: Implement layer 2 ethtool flow rules Daniel Jurgens
2025-11-18 19:01   ` Michael S. Tsirkin
2025-11-19  6:07     ` Dan Jurgens
2025-11-19  9:20     ` Michael S. Tsirkin
2025-11-19  9:26   ` Michael S. Tsirkin
2025-11-19 16:51     ` Dan Jurgens
2025-11-18 14:38 ` [PATCH net-next v11 08/12] virtio_net: Use existing classifier if possible Daniel Jurgens
2025-11-18 21:55   ` Michael S. Tsirkin
2025-11-19  6:26     ` Dan Jurgens
2025-11-19  6:35       ` Michael S. Tsirkin [this message]
2025-11-19  7:18         ` Dan Jurgens
2025-11-19  7:23           ` Michael S. Tsirkin
2025-11-19  7:33             ` Dan Jurgens
2025-11-19  7:41               ` Michael S. Tsirkin
2025-11-19 15:45                 ` Dan Jurgens
2025-11-19 15:58                   ` Michael S. Tsirkin
2025-11-19  7:42   ` Michael S. Tsirkin
2025-11-19  9:01   ` Michael S. Tsirkin
2025-11-18 14:38 ` [PATCH net-next v11 09/12] virtio_net: Implement IPv4 ethtool flow rules Daniel Jurgens
2025-11-18 21:31   ` Michael S. Tsirkin
2025-11-19  7:03     ` Dan Jurgens
2025-11-19  7:06       ` Michael S. Tsirkin
2025-11-19  7:17         ` Dan Jurgens
2025-11-19  7:20           ` Michael S. Tsirkin
2025-11-19  9:18     ` Michael S. Tsirkin
2025-11-19 16:33       ` Dan Jurgens
2025-11-19 16:51         ` Michael S. Tsirkin
2025-11-19 16:59           ` Dan Jurgens
2025-11-18 14:39 ` [PATCH net-next v11 10/12] virtio_net: Add support for IPv6 ethtool steering Daniel Jurgens
2025-11-18 21:45   ` Michael S. Tsirkin
2025-11-19  7:35     ` Dan Jurgens
2025-11-19  7:44       ` Michael S. Tsirkin
2025-11-18 21:48   ` Michael S. Tsirkin
2025-11-19 16:04     ` Dan Jurgens
2025-11-18 14:39 ` [PATCH net-next v11 11/12] virtio_net: Add support for TCP and UDP ethtool rules Daniel Jurgens
2025-11-19  9:14   ` Michael S. Tsirkin
2025-11-19 16:07     ` Dan Jurgens
2025-11-18 14:39 ` [PATCH net-next v11 12/12] virtio_net: Add get ethtool flow rules ops Daniel Jurgens
2025-11-18 18:49   ` Michael S. Tsirkin
2025-11-19 16:24     ` Dan Jurgens
2025-11-18 22:39   ` Michael S. Tsirkin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251119013423-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=danielj@nvidia.com \
    --cc=edumazet@google.com \
    --cc=eperezma@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=jgg@ziepe.ca \
    --cc=kevin.tian@intel.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=parav@nvidia.com \
    --cc=shshitrit@nvidia.com \
    --cc=virtualization@lists.linux.dev \
    --cc=xuanzhuo@linux.alibaba.com \
    --cc=yohadt@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).