Re: [PATCH 08/16] fsverity: kick off hash readahead at data I/O submission time

[Eric Biggers <ebiggers@kernel.org>]

On Mon, Jan 26, 2026 at 11:26:55AM -0800, Darrick J. Wong wrote:
> > +/**
> > + * generic_readahead_merkle_tree() - generic ->readahead_merkle_tree helper
> > + * @inode:	inode containing the Merkle tree
> > + * @index:	0-based index of the first page to read ahead in the inode
> > + * @nr_pages:	number of data pages to read ahead
> 
> On second examination: *data* pages?  I thought @index/@index are the
> range of merkle tree pages to read into the pagecache?
> 
> I'd have thought the kerneldoc would read "number of merkle tree pages
> to read ahead".

Agreed, please make it clear that this function is working on Merkle
tree pages.

> > +	for (level = 0; level < params->num_levels; level++) {
> > +		unsigned long level_start = params->level_start[level];
> > +		unsigned long next_start_hidx = start_hidx >> params->log_arity;
> > +		unsigned long next_end_hidx = end_hidx >> params->log_arity;
> > +		pgoff_t long start_idx = (level_start + next_start_hidx) >>
> > +				params->log_blocks_per_page;
> > +		pgoff_t end_idx = (level_start + next_end_hidx) >>
> > +				params->log_blocks_per_page;
> 
> The pgoff_t usage here makes the unit analysis a bit easier, thanks.

"pgoff_t long" should be just "pgoff_t".

(It compiles anyway because it expands to "unsigned long long"...)

- Eric