Re: [PATCH] fuse: make foffset alignment opt-in for optimum backend performance
From: Miklos Szeredi
Date: Thu Aug 29 2024 - 03:51:51 EST
On Fri, 5 Jul 2024 at 14:00, Jingbo Xu <jefflexu@xxxxxxxxxxxxxxxxx> wrote:
> I'm okay with resuing max_pages as the alignment constraint. They are
> the same in our internal scenarios. But I'm not sure if it is the case
> in other scenarios.
max_pages < alignment makes little sense.
max_pages = n * alignment could make sense, i.e. allow writes that are
whole multiples of the alignment.
I'm not against adding a separate alignment, but it could be just
uint8_t to take up less space in init_out. We could have done that
with max_stack_depth too. Oh well...
Thanks,
Miklos