From: Jason Wang <jasowang@redhat.com>
To: "Zhoujian (jay)" <jianjay.zhou@huawei.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>
Cc: "Huangweidong (C)" <weidong.huang@huawei.com>,
"mst@redhat.com" <mst@redhat.com>,
"wangxin (U)" <wangxinxin.wang@huawei.com>,
"Gonglei (Arei)" <arei.gonglei@huawei.com>,
"imammedo@redhat.com" <imammedo@redhat.com>,
"Liuzhe (Ahriy, Euler)" <liuzhe13@huawei.com>
Subject: Re: [Qemu-devel] [PATCH v5 2/4][RFC] tap: do not close fd if only vhost failed to initialize
Date: Thu, 11 Jan 2018 18:30:05 +0800 [thread overview]
Message-ID: <f90d96be-b58b-8ca1-d236-fe288c0a9b0c@redhat.com> (raw)
In-Reply-To: <B2D15215269B544CADD246097EACE7473AB36FE2@DGGEMM505-MBS.china.huawei.com>
On 2018年01月11日 11:54, Zhoujian (jay) wrote:
> Hi Jason,
>
>> -----Original Message-----
>> From: Jason Wang [mailto:jasowang@redhat.com]
>> Sent: Thursday, January 11, 2018 11:35 AM
>> To: Zhoujian (jay) <jianjay.zhou@huawei.com>; qemu-devel@nongnu.org
>> Cc: Huangweidong (C) <weidong.huang@huawei.com>; mst@redhat.com; wangxin (U)
>> <wangxinxin.wang@huawei.com>; Gonglei (Arei) <arei.gonglei@huawei.com>;
>> imammedo@redhat.com; Liuzhe (Ahriy, Euler) <liuzhe13@huawei.com>
>> Subject: Re: [Qemu-devel] [PATCH v5 2/4][RFC] tap: do not close fd if only
>> vhost failed to initialize
>>
>>
>>
>> On 2018年01月10日 15:39, Zhoujian (jay) wrote:
>>>>>> + *vhost_init_failed = true;
>>>> Why not simply check s->vhost_net after call net_init_tap_one()?
>>> s->vhost_net is always NULL if net_init_tap_one() failed, it can't
>> distinguish
>>> failure reasons.
>> On which condition net_init_tap_one() fail but s->vhost_net is set?
> No, it does not exist, so we could not use s->vhost_net to decide
> whether close the fd of tap when error occurred.
>
>
> Maybe the patch below will be much better to understand, please have a look.
>
> diff --git a/net/tap.c b/net/tap.c
> index 979e622..a5c5111 100644
> --- a/net/tap.c
> +++ b/net/tap.c
> @@ -642,7 +642,8 @@ static void net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
> const char *model, const char *name,
> const char *ifname, const char *script,
> const char *downscript, const char *vhostfdname,
> - int vnet_hdr, int fd, Error **errp)
> + int vnet_hdr, int fd, Error **errp,
> + bool *error_is_set_sndbuf)
> {
> Error *err = NULL;
> TAPState *s = net_tap_fd_init(peer, model, name, fd, vnet_hdr);
> @@ -651,6 +652,9 @@ static void net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
> tap_set_sndbuf(s->fd, tap, &err);
> if (err) {
> error_propagate(errp, err);
> + if (error_is_set_sndbuf) {
> + *error_is_set_sndbuf = true;
> + }
> return;
> }
>
> @@ -748,6 +752,7 @@ int net_init_tap(const Netdev *netdev, const char *name,
> Error *err = NULL;
> const char *vhostfdname;
> char ifname[128];
> + bool error_is_set_sndbuf = false;
>
> assert(netdev->type == NET_CLIENT_DRIVER_TAP);
> tap = &netdev->u.tap;
> @@ -783,7 +788,7 @@ int net_init_tap(const Netdev *netdev, const char *name,
>
> net_init_tap_one(tap, peer, "tap", name, NULL,
> script, downscript,
> - vhostfdname, vnet_hdr, fd, &err);
> + vhostfdname, vnet_hdr, fd, &err, NULL);
> if (err) {
> error_propagate(errp, err);
> return -1;
> @@ -835,7 +840,7 @@ int net_init_tap(const Netdev *netdev, const char *name,
> net_init_tap_one(tap, peer, "tap", name, ifname,
> script, downscript,
> tap->has_vhostfds ? vhost_fds[i] : NULL,
> - vnet_hdr, fd, &err);
> + vnet_hdr, fd, &err, NULL);
> if (err) {
> error_propagate(errp, err);
> goto free_fail;
> @@ -874,10 +879,13 @@ free_fail:
>
> net_init_tap_one(tap, peer, "bridge", name, ifname,
> script, downscript, vhostfdname,
> - vnet_hdr, fd, &err);
> + vnet_hdr, fd, &err, &error_is_set_sndbuf);
> if (err) {
> error_propagate(errp, err);
> - close(fd);
> + if (error_is_set_sndbuf || (tap->has_vhostforce &&
> + tap->vhostforce)) {
> + close(fd);
> + }
> return -1;
> }
> } else {
> @@ -913,10 +921,14 @@ free_fail:
> net_init_tap_one(tap, peer, "tap", name, ifname,
> i >= 1 ? "no" : script,
> i >= 1 ? "no" : downscript,
> - vhostfdname, vnet_hdr, fd, &err);
> + vhostfdname, vnet_hdr, fd, &err,
> + &error_is_set_sndbuf);
> if (err) {
> error_propagate(errp, err);
> - close(fd);
> + if (error_is_set_sndbuf || (tap->has_vhostforce &&
> + tap->vhostforce)) {
> + close(fd);
> + }
> return -1;
> }
> }
>
>
>
> PS: I think I do not express the meaning clearly... I can express it in
> Chinese in private if necessary
>
> Regards,
> Jay
That's kind of ugly.
I would suggest do all the correct thing in net_tap_init_one().
Thanks
next prev parent reply other threads:[~2018-01-11 10:30 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-09 16:40 [Qemu-devel] [PATCH v5 2/4][RFC] tap: do not close fd if only vhost failed to initialize Jay Zhou
2018-01-09 16:40 ` [Qemu-devel] [PATCH v5 3/4] vhost: fix memslot limit check Jay Zhou
2018-01-10 13:16 ` Igor Mammedov
2018-01-09 16:40 ` [Qemu-devel] [PATCH v5 4/4] vhost: used_memslots refactoring Jay Zhou
2018-01-10 4:18 ` [Qemu-devel] [PATCH v5 2/4][RFC] tap: do not close fd if only vhost failed to initialize Zhoujian (jay)
2018-01-10 6:01 ` Jason Wang
2018-01-10 7:39 ` Zhoujian (jay)
2018-01-11 3:34 ` Jason Wang
2018-01-11 3:54 ` Zhoujian (jay)
2018-01-11 10:30 ` Jason Wang [this message]
2018-01-11 12:31 ` Zhoujian (jay)
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=f90d96be-b58b-8ca1-d236-fe288c0a9b0c@redhat.com \
--to=jasowang@redhat.com \
--cc=arei.gonglei@huawei.com \
--cc=imammedo@redhat.com \
--cc=jianjay.zhou@huawei.com \
--cc=liuzhe13@huawei.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=wangxinxin.wang@huawei.com \
--cc=weidong.huang@huawei.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).