lkml.org 
[lkml]   [2008]   [May]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [2.6.26 PATCH, RESEND]: fs_stack/eCryptfs: fsstack_copy_* updates
In message <20080501122136.bc2215c9.akpm@linux-foundation.org>, Andrew Morton writes:
> On Thu, 1 May 2008 13:18:09 -0400
> Erez Zadok <ezk@cs.sunysb.edu> wrote:
>
> > In message <20080430101704.9cbd6384.akpm@linux-foundation.org>, Andrew Morton writes:
> > [...[
> > > Can we avoid having to think?
> > >
> > > void fsstack_copy_inode_size(struct inode *dst, const struct inode *src)
> > > {
> > > blkcnt_t i_blocks;
> > > loff_t i_size;
> > >
> > > i_size = i_size_read(src);
> > > spin_lock_32bit(&src->i_lock);
> > > i_blocks = src->i_blocks;
> > > spin_unlock_32bit(&src->i_lock);
> > >
> > > i_size_write(dst, i_size);
> > > spin_lock_32bit(&dst->i_lock)
> > > dst->i_blocks = i_blocks;
> > > spin_unlock_32bit(&dst->i_lock)
> > > }
> >
> > I can't find spin_[um]lock_32bit anywhere (checkd latest mmotm and linus's
> > tree). I therefore assume this was just your way of saying I should:
> >
> > #if BITS_PER_LONG == 32
> > spin_unlock(&dst->i_lock);
> > #endif
>
> Nope, it was my way of suggesting that you implement it ;)
> include/linux/spinlock.h would be a good place.

Sure, I can implement it and submit a patch. But I've got two questions
first:

1. i_size_read has a tri-state behaviour:

#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
// play with seqcount
#elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPT)
// wrap in preempt_disable/enable
#else
// just return value
#endif

Shouldn't the same be done for i_blocks, with matching i_blocks_read(),
i_blocks_write(), and adding an inode->i_blocks_seqcount field?


2. I've rewritten your suggested code a bit to reduce stack use. Modulo
having 32-bit spin_lock/unlock variants, do you see any problem with this
code below? My testing of it so far on 32/64-bit SMP/UMP have all
passed.

void fsstack_copy_inode_size(struct inode *dst, struct inode *src)
{
#if BITS_PER_LONG == 32
blkcnt_t i_blocks;

spin_lock(&src->i_lock);
i_blocks = src->i_blocks;
spin_unlock(&src->i_lock);
spin_lock(&dst->i_lock);
dst->i_blocks = i_blocks;
spin_unlock(&dst->i_lock);
#else
dst->i_blocks = src->i_blocks;
#endif
i_size_write(dst, i_size_read(src));
}

Thanks,
Erez.


\
 
 \ /
  Last update: 2008-05-02 01:47    [W:0.065 / U:0.976 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site