public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: Wilco Dijkstra <Wilco.Dijkstra@arm.com>,
	'GNU C Library' <libc-alpha@sourceware.org>
Subject: Re: [PATCH] Avoid RMW of flags2 outside lock (BZ #27842)
Date: Thu, 19 May 2022 12:47:39 -0300	[thread overview]
Message-ID: <7d120257-c779-ee56-932d-17738067f0ea@linaro.org> (raw)
In-Reply-To: <DB6PR0801MB187958646D1DCDD920EA00F183D09@DB6PR0801MB1879.eurprd08.prod.outlook.com>



On 19/05/2022 12:10, Wilco Dijkstra wrote:
> 
> Remove an unconditional RMW on flags2 in flockfile - if _IO_FLAGS2_NEED_LOCK
> is set, we are single-threaded and we can safely clear the flag. This fixes
> BZ #27842.
> 

I don't think this is correct because if the caller issues pthread_create 
after flockfile, funlockfile will not issues the correct operations.  I
have a fix that uses a different locking mechanism where the _IO_FLAGS2_NEED_LOCK
is removed by moving both the thread id and single-thread optimization to the
locks itself (on Linux tid has at maximum 30-bits, we can use 1 bits for the
single-thread optimization and 1 bits for congestion optimization).

The issue is now requires slight more memory operations to check if you
can use single-thread optimization, since the FILE lock is not always
present and you need to first check __flags.

I would say that with currency scheme where _IO_FLAGS2_NEED_LOCK is stick,
this is a benign data race (although still undesirable). 

> ---
> 
> diff --git a/stdio-common/flockfile.c b/stdio-common/flockfile.c
> index a5decb450f8d477e3105d02661282afeab58f88b..7ba9ab59082d2c1dfba7d5e9a91175c9dec7ec49 100644
> --- a/stdio-common/flockfile.c
> +++ b/stdio-common/flockfile.c
> @@ -22,7 +22,10 @@
>  void
>  __flockfile (FILE *stream)
>  {
> -  stream->_flags2 |= _IO_FLAGS2_NEED_LOCK;
> +  /* If we're single-threaded, turn off single-thread optimizations
> +     when locking a file.  Avoid updating flags2 if multi-threaded.  */
> +  if (!_IO_need_lock (stream))
> +    stream->_flags2 |= _IO_FLAGS2_NEED_LOCK;
>    _IO_lock_lock (*stream->_lock);
>  }
>  weak_alias (__flockfile, flockfile);

  parent reply	other threads:[~2022-05-19 15:47 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-19 15:10 Wilco Dijkstra
2022-05-19 15:25 ` Andreas Schwab
2022-05-19 15:47 ` Adhemerval Zanella [this message]
2022-05-19 16:26   ` Wilco Dijkstra
2022-05-19 16:57     ` Adhemerval Zanella
2022-05-20 10:34       ` Wilco Dijkstra

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=7d120257-c779-ee56-932d-17738067f0ea@linaro.org \
    --to=adhemerval.zanella@linaro.org \
    --cc=Wilco.Dijkstra@arm.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).