From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: Florian Weimer <fweimer@redhat.com>
Cc: Adhemerval Zanella via Libc-alpha <libc-alpha@sourceware.org>,
leonardo.macchia@gmail.com
Subject: Re: [PATCH 1/3] inet: Fix getnameinfo (NI_NOFQDN) race condition (BZ#28566)
Date: Thu, 11 Nov 2021 11:09:51 -0300 [thread overview]
Message-ID: <e7b14ba5-2b73-1431-dbc8-c3dc4f3d1253@linaro.org> (raw)
In-Reply-To: <87czn6yf0d.fsf@oldenburg.str.redhat.com>
On 11/11/2021 10:54, Florian Weimer wrote:
> * Adhemerval Zanella:
>
>> On 11/11/2021 05:16, Florian Weimer wrote:
>>> * Adhemerval Zanella via Libc-alpha:
>>>
>>>> diff --git a/inet/getnameinfo.c b/inet/getnameinfo.c
>>>> index 8380d85783..58ebbb1154 100644
>>>> --- a/inet/getnameinfo.c
>>>> +++ b/inet/getnameinfo.c
>>>> @@ -86,55 +86,75 @@ libc_freeres_ptr (static char *domain);
>>>> static char *
>>>> nrl_domainname (void)
>>>> {
>>>> + __libc_lock_define_initialized (static, lock);
>>>> + __libc_lock_lock (lock);
>>>>
>>>> + static bool not_first = false;
>>>> if (! not_first)
>>>
>>>> + done:
>>>> + scratch_buffer_free (&tmpbuf);
>>>> + not_first = true;
>>>
>>> This is missing the acquire/release pairing for the double-checked
>>> locking idiom. You can probably use the domain variable directly.
>>
>> But it is done now within the lock, different than current implementation
>> which does outside. I moved to be within the lock exactly to avoid the
>> double-checked locking idiom.
>
> Ah, sorry, I had missed that.
>
>> I think now that we might be moving to a more optimized lll_lock internally
>> using a acquire-load+CAS instead of just CAS we can get it without need
>> to code it explicitly.
>
> The double-checked locking idiom avoids the CAS after initialization.
> With the lll_lock change, an atomic read-modify-write operation still
> happens on the lock in all cases (prior to the eventual return to the
> caller).
I meant H.J internal lock optimization [1], where the lll_lock int this
case will be mostly a relaxed load instead of a CAS (since once 'domain'
is properly initialized, there is no need to take the lock).
[1] https://sourceware.org/pipermail/libc-alpha/2021-November/132899.html
next prev parent reply other threads:[~2021-11-11 14:09 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-10 18:58 [PATCH 0/3] Fixes for getnameinfo() with NI_NOFQDN Adhemerval Zanella
2021-11-10 18:58 ` [PATCH 1/3] inet: Fix getnameinfo (NI_NOFQDN) race condition (BZ#28566) Adhemerval Zanella
2021-11-11 8:16 ` Florian Weimer
2021-11-11 13:34 ` Adhemerval Zanella
2021-11-11 13:54 ` Florian Weimer
2021-11-11 14:09 ` Adhemerval Zanella [this message]
2021-11-11 14:12 ` Florian Weimer
2021-11-11 14:17 ` Adhemerval Zanella
2021-11-11 17:41 ` Florian Weimer
2021-11-10 18:58 ` [PATCH 2/3] inet: Remove strdupa from nrl_domainname() Adhemerval Zanella
2021-11-11 8:18 ` Florian Weimer
2021-11-11 13:36 ` Adhemerval Zanella
2021-11-11 14:00 ` Florian Weimer
2021-11-10 18:58 ` [PATCH 3/3] inet: Return EAI_MEMORY when nrl_domainname() fails to allocate memory Adhemerval Zanella
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=e7b14ba5-2b73-1431-dbc8-c3dc4f3d1253@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=fweimer@redhat.com \
--cc=leonardo.macchia@gmail.com \
--cc=libc-alpha@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).