public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Darren Hart <dvhltc@us.ibm.com>
To: Michel Lespinasse <walken@google.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] futex: add FUTEX_SET_WAIT operation
Date: Mon, 23 Nov 2009 09:21:19 -0800	[thread overview]
Message-ID: <4B0AC48F.7020501@us.ibm.com> (raw)
In-Reply-To: <20091121023636.GA13763@google.com>

Michel Lespinasse wrote:
> On Thu, Nov 19, 2009 at 03:13:40PM -0800, Darren Hart wrote:
>>> This is not supposed to require any particular number of CPUs, so I am  
>>> concerned about the hang.
>>>
>>> How reproducible is this for you ? Do you know if the original test code 
>>> I sent hanged in the same way ?
>> I'm basically 2 for 2 on each version of the futex_wait_test. I haven't  
>> seen it run to completion yet. This is on a stock Ubuntu kernel  
>> (2.6.31-15-generic) on my core duo laptop (32 bit).
> 
> As we found out in private email conversation before, this was due to hitting
> resource allocation limits while creating the threads. The patch below
> (relative to your set_wait branch) should fix this.
> 
> Patch summary:
> * Added FUTEX_WAIT_BITSET/FUTEX_WAKE_BITSET definitions - I do have
>   old enough headers that this was a problem now that you use
>   inline functions in futextest.h

Great, thanks.

> * Changed futex_cmpxchg return type to int - I dont really like this change,
>   but otherwise gcc complains about the volatile keyword being ignored in
>   the returned function type. Feel free to ignore this if you like.

Hrm. Which version of gcc? I've tested with 4.1 and 4.4 with the -Wall 
option, neither complain for me.

> * futex_setwait_lock: changed to a more compact & faster implementation
>   (same algorithm but wrote the loop in a different way)
> * test harness: look at pthread_create return code; if thread creation fails
>   then join with existing threads and exit. Also moved the before/after
>   barrier logic into the test harness rather than the futex_[set]wait_test
>   functions.

I'll split these up and apply - with the exception of the futex_cmpxchg 
bit until we sort out what's going on. It would facilitate patch 
integration in the future if you could break things up functionally and 
use use "git format-patch" to generate the patches.

Again, thank you for the contributions.

-- 
Darren Hart
IBM Linux Technology Center
Real-Time Linux Team

  reply	other threads:[~2009-11-23 17:21 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-17  7:46 [PATCH] futex: add FUTEX_SET_WAIT operation Michel Lespinasse
2009-11-17  8:18 ` Ingo Molnar
2009-11-17  8:55   ` Peter Zijlstra
2009-11-17 16:16     ` Darren Hart
2009-11-18  3:37       ` Michel Lespinasse
2009-11-18  5:29         ` Darren Hart
2009-11-24 14:39         ` [PATCH 0/3] perf bench: Add new benchmark for futex subsystem Hitoshi Mitake
2009-11-24 14:39         ` [PATCH 1/3] perf bench: Add wrappers for atomic operation of GCC Hitoshi Mitake
2009-11-24 16:20           ` Darren Hart
2009-11-26  5:44             ` Hitoshi Mitake
2009-11-24 14:39         ` [PATCH 2/3] perf bench: Add new files for futex performance test Hitoshi Mitake
2009-11-24 16:33           ` Darren Hart
2009-11-26  5:53             ` Hitoshi Mitake
2009-11-26  5:56               ` [PATCH] futextest: Make locktest() in harness.h more general Hitoshi Mitake
2009-11-24 14:39         ` [PATCH 3/3] perf bench: Fix misc files to build files related to futex Hitoshi Mitake
2009-11-18 22:13       ` [PATCH] futex: add FUTEX_SET_WAIT operation Michel Lespinasse
2009-11-19  6:51         ` Darren Hart
2009-11-19 17:03         ` Darren Hart
     [not found]           ` <8d20b11a0911191325u49624854u6132594f13b0718c@mail.gmail.com>
2009-11-19 23:13             ` Darren Hart
2009-11-21  2:36               ` Michel Lespinasse
2009-11-23 17:21                 ` Darren Hart [this message]
2009-11-17 17:24     ` Ingo Molnar
2009-11-17 17:27       ` Darren Hart
2009-11-18  1:49       ` Hitoshi Mitake
2009-11-17  8:50 ` Peter Zijlstra
2009-11-17 15:24   ` Linus Torvalds
2009-11-18  4:21     ` Michel Lespinasse
2009-11-18  5:40       ` Darren Hart
2009-11-30 22:09   ` Darren Hart
2009-12-03  6:55   ` [PATCH] futex: add FUTEX_SET_WAIT operation (and ADAPTIVE) Darren Hart
2009-11-17 17:22 ` [PATCH] futex: add FUTEX_SET_WAIT operation Darren Hart
2009-11-18  3:29   ` Michel Lespinasse
2009-11-18  0:13 ` Darren Hart

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=4B0AC48F.7020501@us.ibm.com \
    --to=dvhltc@us.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=walken@google.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