lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Tue, 16 Aug 2011 09:04:59 -0500
From:	Eric Sandeen <sandeen@...hat.com>
To:	Tao Ma <tm@....ma>
CC:	linux-ext4@...r.kernel.org, "Theodore Ts'o" <tytso@....edu>
Subject: Re: [PATCH] ext4: Create helper function for EXT4_IO_END_UNWRITTEN
 and i_aiodio_unwritten.

On 8/16/11 2:06 AM, Tao Ma wrote:
> From: Tao Ma <boyu.mt@...bao.com>
> 
> EXT4_IO_END_UNWRITTEN flag set and the increase of i_aiodio_unwritten should
> be done simultaneously since ext4_end_io_nolock always clear the flag and
> decrease the counter in the same time.
> 
> We have found some bugs that the flag is set while leaving i_aiodio_unwritten
> unchanged. So this patch just tries to creat a helper function to wrap them
> to avoid any future bug. The idea is inspired by Eric.

Although I like it a little less now that I see it in practice, sorry.   ;)

This:
                /*
                 * Flag the inode(non aio case) or end_io struct (aio case)
                 * that this IO needs to conversion to written when IO is
                 * completed
                 */
		if (!ext4_set_io_unwritten_flag(inode, io))
			ext4_set_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);

just reads weirdly to me.

It encapsulates all the steps, but it really isn't intuitive to read, I think.

"If we (can't? don't?) set the io unwritten flag, set the dio unwritten flag
on the inode."

I wonder if changing "io" to "endio" would make it a little more obvious.

I'm not sure, maybe it's ok.  What do others think?

I also wonder about this, but I get lost in this code.  It was originally:

                if (io && !(io->flag & EXT4_IO_END_UNWRITTEN)) {
                        io->flag = EXT4_IO_END_UNWRITTEN;
                        atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
                } else
                        ext4_set_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);

That means that if it is AIO ("if (io)"), but the flag is already set
(io->flag & EXT4_IO_END_UNWRITTEN), we don't re-set the flag or increase
the count, which is fine - but is it then correct to set the state on the
inode?

thanks,
-Eric

> Cc: Eric Sandeen <sandeen@...hat.com>
> Cc: "Theodore Ts'o" <tytso@....edu>
> Signed-off-by: Tao Ma <boyu.mt@...bao.com>
> ---
> Ted,
> 	This patch is based on the patch with subject "ext4: Resolve the hang
> of direct i/o read in handling EXT4_IO_END_UNWRITTEN.". I meant it for the next
> merge window, but if you think it is also ok for 3.1, go ahead.
> 
>  fs/ext4/ext4.h    |   11 +++++++++++
>  fs/ext4/extents.c |   10 ++--------
>  fs/ext4/inode.c   |    5 +----
>  fs/ext4/page-io.c |    6 ++----
>  4 files changed, 16 insertions(+), 16 deletions(-)
> 
> diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
> index e717dfd..514f670 100644
> --- a/fs/ext4/ext4.h
> +++ b/fs/ext4/ext4.h
> @@ -1247,6 +1247,17 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
>  		 ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count));
>  }
>  
> +static inline int ext4_set_io_unwritten_flag(struct inode *inode,
> +					     struct ext4_io_end *io_end)
> +{
> +	if (io_end && !(io_end->flag & EXT4_IO_END_UNWRITTEN)) {
> +		io_end->flag |= EXT4_IO_END_UNWRITTEN;
> +		atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
> +		return 1;
> +	}
> +
> +	return 0;
> +}
>  /*
>   * Inode dynamic state flags
>   */
> diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
> index 57cf568..c0f8655 100644
> --- a/fs/ext4/extents.c
> +++ b/fs/ext4/extents.c
> @@ -3190,10 +3190,7 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
>  		 * that this IO needs to conversion to written when IO is
>  		 * completed
>  		 */
> -		if (io && !(io->flag & EXT4_IO_END_UNWRITTEN)) {
> -			io->flag = EXT4_IO_END_UNWRITTEN;
> -			atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
> -		} else
> +		if (!ext4_set_io_unwritten_flag(inode, io))
>  			ext4_set_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);
>  		if (ext4_should_dioread_nolock(inode))
>  			map->m_flags |= EXT4_MAP_UNINIT;
> @@ -3572,10 +3569,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
>  		 * that we need to perform conversion when IO is done.
>  		 */
>  		if ((flags & EXT4_GET_BLOCKS_PRE_IO)) {
> -			if (io && !(io->flag & EXT4_IO_END_UNWRITTEN)) {
> -				io->flag = EXT4_IO_END_UNWRITTEN;
> -				atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
> -			} else
> +			if (!ext4_set_io_unwritten_flag(inode, io))
>  				ext4_set_inode_state(inode,
>  						     EXT4_STATE_DIO_UNWRITTEN);
>  		}
> diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
> index 40c0b10..128493b 100644
> --- a/fs/ext4/inode.c
> +++ b/fs/ext4/inode.c
> @@ -2673,10 +2673,7 @@ static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate)
>  	 * but being more careful is always safe for the future change.
>  	 */
>  	inode = io_end->inode;
> -	if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) {
> -		io_end->flag |= EXT4_IO_END_UNWRITTEN;
> -		atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
> -	}
> +	ext4_set_io_unwritten_flag(inode, io_end);
>  
>  	/* Add the io_end to per-inode completed io list*/
>  	spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags);
> diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
> index 78839af..edd2f7b 100644
> --- a/fs/ext4/page-io.c
> +++ b/fs/ext4/page-io.c
> @@ -334,10 +334,8 @@ submit_and_retry:
>  	if ((io_end->num_io_pages >= MAX_IO_PAGES) &&
>  	    (io_end->pages[io_end->num_io_pages-1] != io_page))
>  		goto submit_and_retry;
> -	if (buffer_uninit(bh) && !(io_end->flag & EXT4_IO_END_UNWRITTEN)) {
> -		io_end->flag |= EXT4_IO_END_UNWRITTEN;
> -		atomic_inc(&EXT4_I(inode)->i_aiodio_unwritten);
> -	}
> +	if (buffer_uninit(bh))
> +		ext4_set_io_unwritten_flag(inode, io_end);
>  	io->io_end->size += bh->b_size;
>  	io->io_next_block++;
>  	ret = bio_add_page(io->io_bio, bh->b_page, bh->b_size, bh_offset(bh));

--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists