Re: [PATCH] btrfs: add debug check for extent_io range alignment

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Apr 30, 2013 at 04:29:44PM +0200, David Sterba wrote:
> The 'end' value must exactly cover the end of the interval, which means
> one byte less than the expected block alignment, or in case of a file
> smaller than one block, one byte less than the inode size.

   If that's actually the case, then the error message needs to use
"]" not ")" to close the interval. Using ")" would imply that the end
value is one past the end of the interval in question.

   Hugo.

> Signed-off-by: David Sterba <dsterba@xxxxxxx>
> ---
>  fs/btrfs/extent_io.c |   27 +++++++++++++++++++++++++++
>  1 files changed, 27 insertions(+), 0 deletions(-)
> 
> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
> index c4949f1..6f6beda 100644
> --- a/fs/btrfs/extent_io.c
> +++ b/fs/btrfs/extent_io.c
> @@ -76,10 +76,29 @@ void btrfs_leak_debug_check(void)
>  		kmem_cache_free(extent_buffer_cache, eb);
>  	}
>  }
> +
> +#define btrfs_debug_check_extent_io_range(inode, start, end)		\
> +	__btrfs_debug_check_extent_io_range(__func__, (inode), (start), (end))
> +static inline void __btrfs_debug_check_extent_io_range(const char *caller,
> +		struct inode *inode, u64 start, u64 end)
> +{
> +	u64 isize = i_size_read(inode);
> +
> +	if (end >= PAGE_SIZE && (end % 2) == 0 && end != isize - 1) {
> +		printk_ratelimited(KERN_DEBUG
> +		    "btrfs: %s: ino %llu isize %llu odd range [%llu,%llu)\n",
> +				caller,
> +				(unsigned long long)btrfs_ino(inode),
> +				(unsigned long long)isize,
> +				(unsigned long long)start,
> +				(unsigned long long)end);
> +	}
> +}
>  #else
>  #define btrfs_leak_debug_add(new, head)	do {} while (0)
>  #define btrfs_leak_debug_del(entry)	do {} while (0)
>  #define btrfs_leak_debug_check()	do {} while (0)
> +#define btrfs_debug_check_extent_io_range(c, s, e)	do {} while (0)
>  #endif
>  
>  #define BUFFER_LRU_MAX 64
> @@ -509,6 +528,8 @@ int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
>  	int err;
>  	int clear = 0;
>  
> +	btrfs_debug_check_extent_io_range(tree->mapping->host, start, end);
> +
>  	if (delete)
>  		bits |= ~EXTENT_CTLBITS;
>  	bits |= EXTENT_FIRST_DELALLOC;
> @@ -664,6 +685,8 @@ static void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
>  	struct extent_state *state;
>  	struct rb_node *node;
>  
> +	btrfs_debug_check_extent_io_range(tree->mapping->host, start, end);
> +
>  	spin_lock(&tree->lock);
>  again:
>  	while (1) {
> @@ -756,6 +779,8 @@ __set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
>  	u64 last_start;
>  	u64 last_end;
>  
> +	btrfs_debug_check_extent_io_range(tree->mapping->host, start, end);
> +
>  	bits |= EXTENT_FIRST_DELALLOC;
>  again:
>  	if (!prealloc && (mask & __GFP_WAIT)) {
> @@ -976,6 +1001,8 @@ int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
>  	u64 last_start;
>  	u64 last_end;
>  
> +	btrfs_debug_check_extent_io_range(tree->mapping->host, start, end);
> +
>  again:
>  	if (!prealloc && (mask & __GFP_WAIT)) {
>  		prealloc = alloc_extent_state(mask);

-- 
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
  PGP key: 65E74AC0 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
         --- Is it true that "last known good" on Windows XP ---         
                            boots into CP/M?                             

Attachment: signature.asc
Description: Digital signature


[Index of Archives]     [Linux Filesystem Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux