This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: ofstream, fsync and data loss bugs

From: Paolo Carlini <paolo dot carlini at oracle dot com>
To: Joe Buck <Joe dot Buck at synopsys dot COM>
Cc: Pavel Sanda <ps at twin dot jikos dot cz>, "libstdc++ at gcc dot gnu dot org" <libstdc++ at gcc dot gnu dot org>
Date: Thu, 18 Mar 2010 00:45:18 +0100
Subject: Re: ofstream, fsync and data loss bugs
References: <20100317165853.GA3905@atrey.karlin.mff.cuni.cz> <4BA10F2D.3050505@oracle.com> <20100317183544.GA8071@atrey.karlin.mff.cuni.cz> <4BA12941.7000003@oracle.com> <4348dea51003171536y718b4673w95239258a4d7515d@mail.gmail.com> <20100317231116.GA30933@atrey.karlin.mff.cuni.cz> <20100317233315.GK22959@synopsys.com>

On 03/18/2010 12:33 AM, Joe Buck wrote:
> Are you suggesting putting an fsync() in every close()? That
> would be hugely expensive.
>   
Yeah... Actually, reminds me when novice C++ programmer are advised
*against* using << std::endl; instead << '\n';

But I agree something seems lacking in the ISO C++ specifications about
making the file descriptor available, by now it's really a FAQ. At some
point I think somebody even tried to discuss it within the ISO LWG, but
then the idea got lost...

Paolo.

References:
- ofstream, fsync and data loss bugs
  - From: Pavel Sanda
- Re: ofstream, fsync and data loss bugs
  - From: Paolo Carlini
- Re: ofstream, fsync and data loss bugs
  - From: Pavel Sanda
- Re: ofstream, fsync and data loss bugs
  - From: Paolo Carlini
- Re: ofstream, fsync and data loss bugs
  - From: Jonathan Wakely
- Re: ofstream, fsync and data loss bugs
  - From: Pavel Sanda
- Re: ofstream, fsync and data loss bugs
  - From: Joe Buck

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]