From: Jason Wang <jasowang@redhat.com>
To: zhanghailiang <zhang.zhanghailiang@huawei.com>, qemu-devel@nongnu.org
Cc: zhangchen.fnst@cn.fujitsu.com, hongyang.yang@easystack.cn
Subject: Re: [Qemu-devel] [PATCH RFC v2 3/5] net/filter: Introduce a helper to add a filter to the netdev
Date: Mon, 1 Feb 2016 11:14:48 +0800 [thread overview]
Message-ID: <56AECDA8.6000402@redhat.com> (raw)
In-Reply-To: <1453883380-10532-4-git-send-email-zhang.zhanghailiang@huawei.com>
On 01/27/2016 04:29 PM, zhanghailiang wrote:
> We add a new helper function netdev_add_filter(), this function
> can help adding a filter object to a netdev.
> Besides, we add a is_default member for struct NetFilterState
> to indicate whether the filter is default or not.
>
> Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
> ---
> v2:
> -Re-implement netdev_add_filter() by re-using object_create()
> (Jason's suggestion)
> ---
> include/net/filter.h | 7 +++++
> net/filter.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 87 insertions(+)
>
> diff --git a/include/net/filter.h b/include/net/filter.h
> index af3c53c..ee1c024 100644
> --- a/include/net/filter.h
> +++ b/include/net/filter.h
> @@ -55,6 +55,7 @@ struct NetFilterState {
> char *netdev_id;
> NetClientState *netdev;
> NetFilterDirection direction;
> + bool is_default;
> bool enabled;
> QTAILQ_ENTRY(NetFilterState) next;
> };
> @@ -74,4 +75,10 @@ ssize_t qemu_netfilter_pass_to_next(NetClientState *sender,
> int iovcnt,
> void *opaque);
>
> +void netdev_add_filter(const char *netdev_id,
> + const char *filter_type,
> + const char *id,
> + bool is_default,
> + Error **errp);
> +
> #endif /* QEMU_NET_FILTER_H */
> diff --git a/net/filter.c b/net/filter.c
> index d08a2be..dc7aa9b 100644
> --- a/net/filter.c
> +++ b/net/filter.c
> @@ -214,6 +214,86 @@ static void netfilter_complete(UserCreatable *uc, Error **errp)
> QTAILQ_INSERT_TAIL(&nf->netdev->filters, nf, next);
> }
>
> +QemuOptsList qemu_filter_opts = {
> + .name = "default-filter",
> + .head = QTAILQ_HEAD_INITIALIZER(qemu_filter_opts.head),
> + .desc = {
> + {
> + .name = "qom-type",
> + .type = QEMU_OPT_STRING,
> + },{
> + .name = "id",
> + .type = QEMU_OPT_STRING,
> + },{
> + .name = "netdev",
> + .type = QEMU_OPT_STRING,
> + },{
> + .name = "status",
> + .type = QEMU_OPT_STRING,
> + },
> + { /* end of list */ }
> + },
> +};
> +
> +static void filter_set_default_flag(const char *id,
> + bool is_default,
> + Error **errp)
> +{
> + Object *obj, *container;
> + NetFilterState *nf;
> +
> + container = object_get_objects_root();
> + obj = object_resolve_path_component(container, id);
> + if (!obj) {
> + error_setg(errp, "object id not found");
> + return;
> + }
> + nf = NETFILTER(obj);
> + nf->is_default = is_default;
> +}
> +
> +void netdev_add_filter(const char *netdev_id,
> + const char *filter_type,
> + const char *id,
> + bool is_default,
> + Error **errp)
> +{
> + NetClientState *nc = qemu_find_netdev(netdev_id);
> + char *optarg;
> + QemuOpts *opts = NULL;
> + Error *err = NULL;
> +
> + /* FIXME: Not support multiple queues */
> + if (!nc || nc->queue_index > 1) {
> + return;
> + }
> + /* Not support vhost-net */
> + if (get_vhost_net(nc)) {
> + return;
> + }
> +
> + optarg = g_strdup_printf("qom-type=%s,id=%s,netdev=%s,status=%s",
> + filter_type, id, netdev_id, is_default ? "disable" : "enable"
Instead of this, I wonder maybe it's better to:
- store the default filter property into a pointer to string
- colo code may change the pointer to "filter-buffer,status=disable"
Then, there's no need for lots of codes above:
- no need a "is_default" parameter in netdev_add_filter which does not
scale consider we may want to have more property in the future
- no need to hacking like "qemu_filter_opts"
- no need to have a special flag like "is_default"
Thoughts?
> + opts = qemu_opts_parse_noisily(&qemu_filter_opts,
> + optarg, false);
> + if (!opts) {
> + error_report("Failed to parse param '%s'", optarg);
> + exit(1);
> + }
> + g_free(optarg);
> + if (object_create(NULL, opts, &err) < 0) {
> + error_report("Failed to create object");
> + goto out_clean;
> + }
> + filter_set_default_flag(id, is_default, &err);
> +
> +out_clean:
> + qemu_opts_del(opts);
> + if (err) {
> + error_propagate(errp, err);
> + }
> +}
> +
> static void netfilter_finalize(Object *obj)
> {
> NetFilterState *nf = NETFILTER(obj);
next prev parent reply other threads:[~2016-02-01 3:15 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-27 8:29 [Qemu-devel] [PATCH RFC v2 0/5] Netfilter: Add each netdev a default filter zhanghailiang
2016-01-27 8:29 ` [Qemu-devel] [PATCH RFC v2 1/5] net/filter: Add a 'status' property for filter object zhanghailiang
2016-01-27 8:29 ` [Qemu-devel] [PATCH RFC v2 2/5] vl: Make object_create() public zhanghailiang
2016-02-01 3:05 ` Jason Wang
2016-02-01 6:19 ` Hailiang Zhang
2016-02-01 7:27 ` Jason Wang
2016-02-01 7:34 ` Hailiang Zhang
2016-02-01 10:41 ` Daniel P. Berrange
2016-02-01 10:44 ` Hailiang Zhang
2016-01-27 8:29 ` [Qemu-devel] [PATCH RFC v2 3/5] net/filter: Introduce a helper to add a filter to the netdev zhanghailiang
2016-02-01 3:14 ` Jason Wang [this message]
2016-02-01 6:13 ` Hailiang Zhang
2016-02-01 7:46 ` Jason Wang
2016-02-01 7:56 ` Hailiang Zhang
2016-02-01 8:05 ` Yang Hongyang
2016-02-01 8:21 ` Hailiang Zhang
2016-02-01 9:18 ` Jason Wang
2016-02-01 9:39 ` Hailiang Zhang
2016-02-01 9:49 ` Jason Wang
2016-02-01 10:41 ` Hailiang Zhang
2016-02-01 9:04 ` Jason Wang
2016-02-01 9:22 ` Hailiang Zhang
2016-02-01 9:42 ` Jason Wang
2016-02-01 10:40 ` Hailiang Zhang
2016-02-05 6:19 ` Jason Wang
2016-02-05 7:01 ` Hailiang Zhang
2016-02-05 7:40 ` Jason Wang
2016-02-05 8:29 ` Hailiang Zhang
2016-02-01 12:21 ` Hailiang Zhang
2016-02-01 10:43 ` Daniel P. Berrange
2016-02-01 10:57 ` Hailiang Zhang
2016-01-27 8:29 ` [Qemu-devel] [PATCH RFC v2 4/5] filter-buffer: Accept zero interval zhanghailiang
2016-01-27 8:29 ` [Qemu-devel] [PATCH RFC v2 5/5] net/filter: Add a default filter to each netdev zhanghailiang
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=56AECDA8.6000402@redhat.com \
--to=jasowang@redhat.com \
--cc=hongyang.yang@easystack.cn \
--cc=qemu-devel@nongnu.org \
--cc=zhang.zhanghailiang@huawei.com \
--cc=zhangchen.fnst@cn.fujitsu.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).