lkml.org 
[lkml]   [2006]   [Dec]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] incorrect error handling inside generic_file_direct_write
On Tue, 12 Dec 2006 12:22:14 +0300
Dmitriy Monakhov <dmonakhov@sw.ru> wrote:

> >> @@ -2041,6 +2041,14 @@ generic_file_direct_write(struct kiocb *
> >> mark_inode_dirty(inode);
> >> }
> >> *ppos = end;
> >> + } else if (written < 0) {
> >> + loff_t isize = i_size_read(inode);
> >> + /*
> >> + * generic_file_direct_IO() may have instantiated a few blocks
> >> + * outside i_size. Trim these off again.
> >> + */
> >> + if (pos + count > isize)
> >> + vmtruncate(inode, isize);
> >> }
> >>
> >
> > XFS (at least) can call generic_file_direct_write() with i_mutex not held.
> How could it be ?
>
> from mm/filemap.c:2046 generic_file_direct_write() comment right after
> place where i want to add vmtruncate()
> /*
> * Sync the fs metadata but not the minor inode changes and
> * of course not the data as we did direct DMA for the IO.
> * i_mutex is held, which protects generic_osync_inode() from
> * livelocking.
> */
>
> > And vmtruncate() expects i_mutex to be held.
> generic_file_direct_IO must called under i_mutex too
> from mm/filemap.c:2388
> /*
> * Called under i_mutex for writes to S_ISREG files. Returns -EIO if something
> * went wrong during pagecache shootdown.
> */
> static ssize_t
> generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,

yup, the comments are wrong.

> This means XFS generic_file_direct_write() call generic_file_direct_IO() without
> i_mutex held too?

Think so. XFS uses blockdev_direct_IO_own_locking(). We'd need to check
with the XFS guys regarding its precise operation and what needs to be done
here.

> >
> > I guess a suitable solution would be to push this problem back up to the
> > callers: let them decide whether to run vmtruncate() and if so, to ensure
> > that i_mutex is held.
> >
> > The existence of generic_file_aio_write_nolock() makes that rather messy
> > though.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2006-12-12 07:41    [W:0.193 / U:0.296 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site