Re: [PATCH 14/20] block: remove the nr_sects field in struct hd_struct

[Christoph Hellwig <hch@xxxxxx>]

On Fri, Nov 20, 2020 at 09:05:48PM +0100, Jan Kara wrote:
> The code is already switched to it AFAICT (the lock is really only used in
> the two places that write i_size). But the problem is that in theory two
> i_size_write() calls can race in a way that the resulting stored i_size is a
> mix of two stored sizes. Now I have hard time imagining how this could
> happen for a block device and if two reconfigurations of a block device
> could race like that we'd have a large problems anyway...

Now that you mention it, yes - i_size_write needs to be under i_rwsem
or an equivalent lock.  We could look into using i_rwsem also for block
device, but for now the spinlock seems to be doing fine.  Note that
in current mainline we only have such a lock protecting i_size of the
block_device inode, but none for the size in hd_struct.