From: Jason Wang <jasowang@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>, peter.maydell@linaro.org
Cc: qemu-devel@nongnu.org
Subject: Re: [PULL 03/16] net: validate that ids are well formed
Date: Fri, 12 Mar 2021 21:26:27 +0800 [thread overview]
Message-ID: <c001654b-9802-7221-17e6-033bb10b89ff@redhat.com> (raw)
In-Reply-To: <2e3d2321-be67-ab5d-9511-64de1fcbd196@redhat.com>
On 2021/3/12 4:44 下午, Paolo Bonzini wrote:
> On 12/03/21 07:16, Jason Wang wrote:
>> From: Paolo Bonzini <pbonzini@redhat.com>
>>
>> When a network or network device is created from the command line or
>> HMP,
>> QemuOpts ensures that the id passes the id_wellformed check. However,
>> QMP skips this:
>>
>> $ qemu-system-x86_64 -qmp stdio -S -nic user,id=123/456
>> qemu-system-x86_64: -nic user,id=123/456: Parameter id expects an
>> identifier
>> Identifiers consist of letters, digits, -, ., _, starting with a
>> letter.
>>
>> $ qemu-system-x86_64 -qmp stdio -S
>> {"execute":"qmp_capabilities"}
>> {"return": {}}
>> {"execute":"netdev_add", "arguments": {"type": "user", "id":
>> "123/456"}}
>> {"return": {}}
>>
>> After:
>>
>> $ qemu-system-x86_64 -qmp stdio -S
>> {"execute":"qmp_capabilities"}
>> {"return": {}}
>> {"execute":"netdev_add", "arguments": {"type": "user", "id":
>> "123/456"}}
>> {"error": {"class": "GenericError", "desc": "Parameter "id"
>> expects an identifier"}}
>>
>> Validity checks should be performed always at the bottom of the call
>> chain,
>> because QMP skips all the steps above. Do this for the network
>> subsystem.
>>
>> Cc: Jason Wang <jasowang@redhat.com>
>> Reviewed-by: Eric Blake <eblake@redhat.com>
>> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>> ---
>> net/net.c | 12 ++++++++++++
>> 1 file changed, 12 insertions(+)
>>
>> diff --git a/net/net.c b/net/net.c
>> index 9c784da..d36729f 100644
>> --- a/net/net.c
>> +++ b/net/net.c
>> @@ -44,6 +44,7 @@
>> #include "qemu/cutils.h"
>> #include "qemu/config-file.h"
>> #include "qemu/ctype.h"
>> +#include "qemu/id.h"
>> #include "qemu/iov.h"
>> #include "qemu/qemu-print.h"
>> #include "qemu/main-loop.h"
>> @@ -1011,6 +1012,17 @@ static int net_client_init1(const Netdev
>> *netdev, bool is_netdev, Error **errp)
>> }
>> }
>> + /*
>> + * The id for -net has already been checked by QemuOpts and
>> + * could be automatically generated, in which case it is not
>> + * well-formed by design. HMP and QMP only call us with
>> + * is_netdev == true.
>> + */
>> + if (is_netdev && !id_wellformed(netdev->id)) {
>> + error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "id", "an
>> identifier");
>> + return -1;
>> + }
>> +
>> nc = qemu_find_netdev(netdev->id);
>> if (nc) {
>> error_setg(errp, "Duplicate ID '%s'", netdev->id);
>>
>
> Sorry, I sent v2 yesterday. This patch passed the tests at the time
> it was submitted, but now fails (because it does not work with -nic).
I don't see that. But I add a fixup in the pull request:
https://lists.gnu.org/archive/html/qemu-devel/2021-03/msg04237.html
If it doesn't make sense, I will drop this and send a new pull request
next week.
THanks
>
> Paolo
>
>
next prev parent reply other threads:[~2021-03-12 13:27 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-12 6:16 [PULL 00/16] Net patches Jason Wang
2021-03-12 6:16 ` [PULL 01/16] virtio-net: calculating proper msix vectors on init Jason Wang
2021-03-12 6:16 ` [PULL 02/16] net: Fix build error when DEBUG_NET is on Jason Wang
2021-03-12 6:16 ` [PULL 03/16] net: validate that ids are well formed Jason Wang
2021-03-12 8:44 ` Paolo Bonzini
2021-03-12 13:26 ` Jason Wang [this message]
2021-03-12 14:23 ` Paolo Bonzini
2021-03-12 6:16 ` [PULL 04/16] net: unbreak well-form id check for "-nic" Jason Wang
2021-03-12 6:16 ` [PULL 05/16] e1000: fail early for evil descriptor Jason Wang
2021-03-12 6:16 ` [PULL 06/16] net: introduce qemu_receive_packet() Jason Wang
2021-03-12 6:16 ` [PULL 07/16] e1000: switch to use qemu_receive_packet() for loopback Jason Wang
2021-03-12 6:16 ` [PULL 08/16] dp8393x: switch to use qemu_receive_packet() for loopback packet Jason Wang
2021-03-12 6:16 ` [PULL 09/16] msf2-mac: switch to use qemu_receive_packet() for loopback Jason Wang
2021-03-12 6:16 ` [PULL 10/16] sungem: " Jason Wang
2021-03-12 6:16 ` [PULL 11/16] tx_pkt: switch to use qemu_receive_packet_iov() " Jason Wang
2021-03-12 6:16 ` [PULL 12/16] rtl8139: switch to use qemu_receive_packet() " Jason Wang
2021-03-12 6:16 ` [PULL 13/16] pcnet: " Jason Wang
2021-03-12 6:16 ` [PULL 14/16] cadence_gem: " Jason Wang
2021-03-12 6:16 ` [PULL 15/16] lan9118: " Jason Wang
2021-03-12 6:16 ` [PULL 16/16] pvrdma: wean code off pvrdma_ring.h kernel header Jason Wang
2021-03-14 11:37 ` [PULL 00/16] Net patches Peter Maydell
2021-03-15 3:35 ` Jason Wang
2021-03-15 5:39 ` Thomas Huth
2021-03-15 8:07 ` Jason Wang
2021-03-15 9:52 ` Philippe Mathieu-Daudé
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=c001654b-9802-7221-17e6-033bb10b89ff@redhat.com \
--to=jasowang@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/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).