This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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: [patch, fortran] Large transfer support for IO library

From: Janne Blomqvist <jblomqvi at cc dot hut dot fi>
To: Richard Henderson <rth at redhat dot com>, GNU GFortran <fortran at gcc dot gnu dot org>, GCC patches <gcc-patches at gcc dot gnu dot org>
Date: Fri, 30 Sep 2005 21:23:13 +0300
Subject: Re: [patch, fortran] Large transfer support for IO library
References: <20050927190634.GH2419@vipunen.hut.fi> <20050928201950.GI2419@vipunen.hut.fi> <20050929204920.GA17184@redhat.com>

On Thu, Sep 29, 2005 at 01:49:20PM -0700, Richard Henderson wrote:
> You can't rely on doing this in exactly two writes.  Or transfers
> less that SSIZE_MAX in exactly one write.  On some systems, signals
> do not restart system calls.

Indeed, indeed. Live and learn..

The attached patch fixes the problem pointed out by rth. Additionally,
it checks for EINTR and continues reading/writing in that case.

I added PR 16339 to the ChangeLog. I think that once this patch is
committed, we can close 16339 and 23363.

-- 
Janne Blomqvist

Attachment: ChangeLog
Description: Text document

Attachment: large_trans-3.diff
Description: Text document

References:
- [patch, fortran] Large transfer support for IO library
  - From: Janne Blomqvist
- Re: [patch, fortran] Large transfer support for IO library
  - From: Janne Blomqvist
- Re: [patch, fortran] Large transfer support for IO library
  - From: Richard Henderson

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