write barriers - Was: Re: [RFC][PATCH] problem ofcont_prepare_write()

[Anton Altaparmakov]

Hi,

Sorry to pickup such a long thread but here goes anyway...

On Mon, 2004-11-22 at 15:43 -0800, Andrew Morton wrote:
> Anton Altaparmakov <aia21@xxxxxxxxx> wrote:
> >
> > I always have a flush_dcache_page(page) between the memset() and 
> > the SetPageUptodate() so I don't need the barrier, right?  Or does the 
> > flush_dcache_page() not imply ordering?
> 
> err, flush_dcache_page() might indeed provide a write barrier on all
> architectures which need write barriers.  Then again it might not ;) It's
> not intended that this be the case.

What about if the page is unmapped between the memset() and the
SetPageUptodate()?  Does that imply ordering?  I.e. do I need a write
barrier in code like this:

memset(page_address(page), 0, blah);
flush_dcache_page(page);
kunmap(page);
SetPageUptodate(page);

And a more general question:

If I am setting two variables in sequence and it is essential that if a
different cpu reads those variables it seems them updated in the same
order as they were written in the C code do I need a write barrier in
between the two?  For example:

ntfs_inode->allocated_size = 10;
ntfs_inode->initilized_size = 10;

Should another CPU see initialized_size = 10 but allocated_size < 10 the
ntfs driver will blow up in some places.  So does that mean I need a
write barrier, between the two?

If yes, do I still need it if I wrap the two settings (and all accesses)
with a spin lock?  And in particular with a rw-spinlock?  For example:

write_lock_irqsave(&ntfs_inode->size_lock, flags);
ntfs_inode->allocated_size = 10;
ntfs_inode->initilized_size = 10;
write_unlock_irqrestore(&ntfs_inode->size_lock, flags);

Do I still need a write barrier or does the spinlock imply it already?

Thanks a lot in advance and apologies for the stupid(?) questions...

Best regards,

        Anton
-- 
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Unix Support, Computing Service, University of Cambridge, CB2 3QH, UK
Linux NTFS maintainer / IRC: #ntfs on irc.freenode.net
WWW: http://linux-ntfs.sf.net/ & http://www-stu.christs.cam.ac.uk/~aia21/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/