Re: [PATCH v11 2/3] NFSD: Implement NFSD_IO_DIRECT for NFS WRITE

[Christoph Hellwig <hch@infradead.org>]

On Mon, Nov 10, 2025 at 10:42:48AM -0500, Mike Snitzer wrote:
> On Mon, Nov 10, 2025 at 01:12:33AM -0800, Christoph Hellwig wrote:
> > On Fri, Nov 07, 2025 at 05:16:27PM -0500, Mike Snitzer wrote:
> > > Those ends lend themselves to benefitting from more capable RMW
> > > if/when needed.
> > 
> > How are ends of a larger I/O inhently more benefits from cached
> > RMW then say a single tiny I/O?
> 
> Any unaligned WRITE, no matter the size of the WRITE, will benefit
> from RMW of its subpage ends if streaming unaligned WRITEs.

Yes.

> I saw below that even saying "unaligned WRITE" leaves you annoyed.

No, it doesn't.

> Every time I say stuff like "the entire reason for the patchset" I set
> myself up for a rebuttal from you.  You just did the same thing.  I'm
> going to try to be accomodating and cool but please lets dial back
> this weirdly caustic dynamic that is forming yet again.
> 
> Why is it that you cannot acknowledge that it isn't an all or nothing
> situation?

I'm actually arguing for that, not you.  You make claims of "the entire
reason", not me.


> Restating an important point underpinning my desire to handle this
> niche unaligned streaming WRITE case efficiently: it removes the one
> case I know of where NFSD_IO_DIRECT cannot compete with
> NFSD_IO_BUFFERED. Whereby making NFSD_IO_DIRECT a compelling
> alternative default IO mode.  To leave it to use DONTCACHE would be
> purposely compromising NFSD_IO_DIRECT to be less effective.
> 
> I'm also saying: there is very limited downside to using cached
> buffered for these subpage end segments.  But IF I'm proven wrong and
> they do show themselves to be a problem in the future then a code
> change will be needed.

So spell them out in a comment, and leave nuggets for future readers.
And explain why they don't apply for a single segment this is not
page aligned by either being smaller than a page or straddling two
pages.

> 
> > Also to go back to my previous mail:  How can an I/O that is aligned
> > in file offset and length, but not in the backing memory ever going
> > to benefit from this caching?
> 
> It wouldn't, because it would take the no_dio path and the entire IO
> would be issued using DONTCACHE.

Yes, I see that right now it does not.  But I want to hear from you
why you think it is worth treating different.  Because there is no
rationale that a small unaligned write won't benefit from it.  In fact
all conventional wisdom suggest it will benefit even more.  And I'm
trying to explain that for a few days now, but somehow it doesn't make
it across.

> 
> That I just had to point that out shows how absurd all this contrarian
> hand-wringing has gotten.

Can you please top these attacks?

> 
> > > Or is this something we make tunable?
> > > NFSD_IO_DIRECT_DONTCACHE_UNALIGNED?
> > 
> > Throwing in yet another tunable while not understanding the problem
> > at hand is about the worst possible outcome.
> 
> Just because you don't understand something: that isn't my failing.
> So your need to be the expert here is misplaced.  Please show you
> understand the problem before you project lack of understanding onto
> me.

Well, if you do understand it you failed to explain it, which means
there it no _collective_ understanding.  But to me this thread suggests
you don't understand it either, although I'd rather get to the bottom
of it then getting stuck on who understands what.