[PATCH] nvmet-passthru: Cleanup nvmet_passthru_map_sg()

[Logan Gunthorpe]

On 2020-10-16 7:57 a.m., Christoph Hellwig wrote:
> On Thu, Oct 15, 2020 at 12:40:52PM -0600, Logan Gunthorpe wrote:
> blk_rq_map_sg maps the bio chain.  Take a look at the for_each_bio
> loop in __blk_bios_map_sg.

Ah, yes, ok... I originally had a loop that created a chain of bios but
used bio_add_page() instead of bio_add_pc_page(). This was combined with
a separate check for max_segments and max_hw_sectors. When I switched to
bio_add_pc_page() to drop the separate check, it no longer made sense to
do the chain because there was no way to tell if bio_add_pc_page()
failed because of bio_full() or because it exceeded
max_segments/hw_sectors. If the bio was full, we'd want to create the
chain, but if it exceeded the maximums we'd want to error out...

>> If you want me to send a patch to future proof the MDTS limit with
>> BIO_MAX_PAGES, I can do that, but it doesn't look like it will have any
>> effect right now unless big things change.
> 
> Yes, please do.

Will do.