Re: [PATCH] fs: push i_mutex and filemap_write_and_wait down into->fsync() handlers

[Josef Bacik]

On 06/29/2011 02:09 PM, Christoph Hellwig wrote:
>> -int blkdev_fsync(struct file *filp, int datasync)
>> +int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
>>  {
>>  	struct inode *bd_inode = filp->f_mapping->host;
>>  	struct block_device *bdev = I_BDEV(bd_inode);
>> @@ -389,14 +389,10 @@ int blkdev_fsync(struct file *filp, int datasync)
>>  	 * i_mutex and doing so causes performance issues with concurrent
>>  	 * O_SYNC writers to a block device.
>>  	 */
>> -	mutex_unlock(&bd_inode->i_mutex);
>> -
> 
> Now that i_mutex on entry isn't held the comment above can be removed.
> 

The comment is still useful since it states why we don't take the
i_mutex at all, so I think it's still valuable.  Thanks,

Josef
--
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/