From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Thomas Monjalon <thomas@monjalon.net>, Matan Azrad <matan@nvidia.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Subject: Re: [dpdk-dev] [PATCH] ethdev: avoid unregistering a non-allocated callback
Date: Thu, 15 Jul 2021 11:06:42 +0200 [thread overview]
Message-ID: <3e53ea11-02c7-4137-917a-505ef0343740@intel.com> (raw)
In-Reply-To: <4457539.Pa9PMCDbyH@thomas>
On 7/14/2021 4:42 PM, Thomas Monjalon wrote:
> 14/07/2021 16:16, Matan Azrad:
>> From: Thomas Monjalon
>>> 13/07/2021 15:42, Matan Azrad:
>>>> From: Thomas Monjalon
>>>>> When registering a new event callback, if allocation fails, there is
>>>>> no need for unregistering the callback, because it is not registered.
>>>>>
>>>>> Fixes: 9ec0b3869d8d ("ethdev: allow event registration for all
>>>>> ports")
>>>>> Cc: stable@dpdk.org
>>>>>
>>>>> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
>>>>> ---
>>>>> --- a/lib/ethdev/rte_ethdev.c
>>>>> +++ b/lib/ethdev/rte_ethdev.c
>>>>> @@ -4649,8 +4649,6 @@ rte_eth_dev_callback_register(uint16_t
>>>>> } else {
>>>>> rte_spinlock_unlock(ð_dev_cb_lock);
>>>>> - rte_eth_dev_callback_unregister(port_id, event,
>>>>> - cb_fn, cb_arg);
>>>>
>>>> Please pay attention to the case of port_id=RTE_ETH_ALL where the user
>>> wants to register the event for all the ports.
>>>>
>>>> In this case, when a failure happens for one of the ports, this unregister call
>>> cleans the callback from all the ports.
>>>
>>> Yes I missed it. Now I better understand the intent, thanks.
>>>
>>> Next question: do we really want to rollback already registered ports?
>>> Anyway, if we are out of memory, I think it is better not doing more
>>> operations.
>>> There can be various opinions on this topic, please give yours.
>>
>> Sure,
>> I understand that memory error is serious,
>> Do you think it is a fatal error? If so, maybe we should use rte_exit?
>
> We don't call rte_exit in the lib, so the app can do whatever it wants.
>
+1
>> That way or others, I think the behavior should be a convention for all the file functions(at least).
>
> What do you mean "all the file functions"?
>
>> I tend to do cleanup on any error.
>
> I would like to hear opinions from others as well.
>
I also tend to do the cleanup, since API returns error I think application will
be right to think that no callback registered, partially registered callbacks on
error can be confusing.
next prev parent reply other threads:[~2021-07-15 9:06 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-13 13:17 Thomas Monjalon
2021-07-13 13:21 ` Andrew Rybchenko
2021-07-13 13:42 ` Matan Azrad
2021-07-14 11:42 ` Thomas Monjalon
2021-07-14 14:16 ` Matan Azrad
2021-07-14 14:42 ` Thomas Monjalon
2021-07-15 9:06 ` Ferruh Yigit [this message]
2022-09-26 14:36 ` Thomas Monjalon
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=3e53ea11-02c7-4137-917a-505ef0343740@intel.com \
--to=ferruh.yigit@intel.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=dev@dpdk.org \
--cc=matan@nvidia.com \
--cc=thomas@monjalon.net \
/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).