From: Kurt Kanzenbach <kurt@linutronix.de>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>,
libc-alpha@sourceware.org
Subject: Re: [PATCH 2/4] nptl: Use FUTEX_LOCK_PI2 when available
Date: Fri, 13 Aug 2021 08:26:46 +0200 [thread overview]
Message-ID: <87czqhyi55.fsf@kurt> (raw)
In-Reply-To: <3991be9a-2a41-c211-a4ce-f7efdad67cc5@linaro.org>
[-- Attachment #1: Type: text/plain, Size: 1945 bytes --]
On Thu Aug 12 2021, Adhemerval Zanella wrote:
> On 12/08/2021 03:42, Kurt Kanzenbach wrote:
>> Hi Adhemerval,
>>
>> thanks for the respin of this series. Just some minor comments below.
>>
>> On Wed Aug 11 2021, Adhemerval Zanella wrote:
>>> This patch uses the new futex PI operation provided by Linux v5.14
>>> when it is required.
>>>
>>> The futex_lock_pi64() is moved to futex-internal.c (since it used on
>>> two different places and its code size might be large depending of the
>>> kernel configuration) and clockid is added as an argument.
>>>
>>> Co-authored-by: Kurt Kanzenbach <kurt@linutronix.de>
>>> ---
>>> nptl/futex-internal.c | 63 +++++++++++++++++++++++++++++++
>>> nptl/pthread_mutex_lock.c | 3 +-
>>> nptl/pthread_mutex_timedlock.c | 3 +-
>>> sysdeps/nptl/futex-internal.h | 54 +-------------------------
>>> sysdeps/nptl/lowlevellock-futex.h | 1 +
>>> 5 files changed, 70 insertions(+), 54 deletions(-)
>>>
>>> diff --git a/nptl/futex-internal.c b/nptl/futex-internal.c
>>> index e74647a9d4..58605b2fca 100644
>>> --- a/nptl/futex-internal.c
>>> +++ b/nptl/futex-internal.c
>>> @@ -140,3 +140,66 @@ __futex_abstimed_wait_cancelable64 (unsigned int* futex_word,
>>> abstime, private, true);
>>> }
>>> libc_hidden_def (__futex_abstimed_wait_cancelable64)
>>> +
>>> +int
>>> +__futex_lock_pi64 (int *futex_word, clockid_t clockid,
>>> + const struct __timespec64 *abstime, int private)
>>> +{
>>> + int err;
>>
>> Is the clockid check not needed?
>>
>> if (! lll_futex_supported_clockid (clockid))
>> return EINVAL;
>
> The ___pthread_mutex_clocklock64() already checks it (as other functions that uses
> a clockid_t). I also tested for invalid clocks on tst-mutexpi10.c.
>
> And I think that the check on __futex_abstimed_wait_common() is superfluous..
OK, I see.
Thanks,
Kurt
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 861 bytes --]
next prev parent reply other threads:[~2021-08-13 6:26 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-11 20:01 [PATCH 0/4] Add CLOCK_MONOTONIC support for PI mutexes Adhemerval Zanella
2021-08-11 20:01 ` [PATCH 1/4] Linux: Add FUTEX_LOCK_PI2 Adhemerval Zanella
2021-08-11 20:01 ` [PATCH 2/4] nptl: Use FUTEX_LOCK_PI2 when available Adhemerval Zanella
2021-08-12 6:42 ` Kurt Kanzenbach
2021-08-12 12:40 ` Adhemerval Zanella
2021-08-13 6:26 ` Kurt Kanzenbach [this message]
2021-08-11 20:01 ` [PATCH 3/4] support: Add xpthread_mutex_pi_support_monotonic Adhemerval Zanella
2021-08-16 8:01 ` Florian Weimer
2021-08-19 12:49 ` Adhemerval Zanella
2021-08-11 20:01 ` [PATCH 4/4] nptl: Add CLOCK_MONOTONIC support for PI mutexes 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=87czqhyi55.fsf@kurt \
--to=kurt@linutronix.de \
--cc=adhemerval.zanella@linaro.org \
--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).