Re: [PATCH 0/9] some fixes for bugs spotted by valgrind

[Andi Kleen <andi@firstfloor.org>]

>         bh = btrfs_read_dev_super(fs_devices->latest_bdev);
>         if (!bh) {
>                 err = -EINVAL;
>                 goto fail_alloc;
>         }
> 
>         memcpy(&fs_info->super_copy, bh->b_data, sizeof(fs_info->super_copy));
>         memcpy(&fs_info->super_for_commit, &fs_info->super_copy,
>                sizeof(fs_info->super_for_commit));
>         brelse(bh);
> 
> But the way superblocks are written look racy.

FWIW for my ext4 checksumming work I had to add an extra lock to stabilize 
the super block (and the inodes) during checksumming.

Maybe something similar is needed here too.

-Andi
-- 
ak@linux.intel.com -- Speaking for myself only.