Re: [RESEND] [PATCH] VFS: make file->f_pos access atomic on 32bit arch

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Wed, 2008-10-08 at 10:32 +0200, Eric Dumazet wrote:

> About dup() syscall, it wont help, since old and new descriptor points to
> the same "struct file", definitly sharing file position, since first Unixes.
> 
> To quote the fine manual :
> 
>        After successful return of dup or dup2, the old and new descriptors may
>        be used interchangeably. They share locks, file position  pointers  and
>        flags;  for example, if the file position is modified by using lseek on
>        one of the descriptors, the position is also changed for the other.

Ah, ok. I'll try to remember for next time I write a multi-threaded user
app (which given the size of my kernel todo list won't be any time soon
I guess ;-)

> pread()/pwrite() are used my multi-threaded applications that want to share
> a single "struct file".

Yeah, pread/pwrite is good.

>  Or they must use some form of synchronization around
> regular read()/write()/lseek() calls.
> 
> There is no generic f_pos race, only buggy applications.

Agreed.