Re: [PATCH v2] btrfs: don't limit direct reads to a single sector

[Qu Wenruo <wqu@suse.com>]

On 2022/6/21 14:40, Christoph Hellwig wrote:
> On Tue, Jun 21, 2022 at 02:35:55PM +0800, Qu Wenruo wrote:
>> In fact, one bio vector can point to multiple pages, as its bv_len can
>> be larger than PAGE_SIZE.
> 
> Yes, it can.  But it usually doesn't, as that requires contigous memory.

That's true.

> Without the large folio support now merges for xfs/iomap that is very
> unusual and tends to happen only soon after booting.

A little off-topic here, what did the extra XFS/iomap do here?

IIRC the multi-page bio vector is already there for years.

As long as the pages in page cache are contiguous, bio_add_page() will 
create such multi-page vector, without any extra support from fs.

>  At which point
> allocating the larger csums array is also not a problem as we can
> find contigous memory for that easily as well.  For direct I/O on the
> other hand the destination could be THPs or hugetlbs even when memory
> is badly fragmented.

My point here is just no need to align any limit.

Smash a good enough value here is enough, or we need to dig way deeper 
to see if the MAX_SECTORS based one is really correct or not.

Thanks,
Qu