Re: [PATCHSET 0/2] Turn single segment imports into ITER_UBUF

From: Linus Torvalds <torvalds@linux-foundation.org>
To: Jens Axboe <axboe@kernel.dk>
Cc: linux-fsdevel@vger.kernel.org, brauner@kernel.org,
	Al Viro <viro@zeniv.linux.org.uk>
Subject: Re: [PATCHSET 0/2] Turn single segment imports into ITER_UBUF
Date: Fri, 24 Mar 2023 14:14:58 -0700	[thread overview]
Message-ID: <CAHk-=wgjPAUwbQ9bf764x6xL8Ht56CGX79OLTG-fCS6u8yLaCA@mail.gmail.com> (raw)
In-Reply-To: <20230324204443.45950-1-axboe@kernel.dk>

On Fri, Mar 24, 2023 at 1:44 PM Jens Axboe <axboe@kernel.dk> wrote:
>
> We've been doing a few conversions of ITER_IOVEC to ITER_UBUF in select
> spots, as the latter is cheaper to iterate and hence saves some cycles.
> I recently experimented [1] with io_uring converting single segment READV
> and WRITEV into non-vectored variants, as we can save some cycles through
> that as well.
>
> But there's really no reason why we can't just do this further down,
> enabling it for everyone. It's quite common to use vectored reads or
> writes even with a single segment, unfortunately, even for cases where
> there's no specific reason to do so. From a bit of non-scientific
> testing on a vm on my laptop, I see about 60% of the import_iovec()
> calls being for a single segment.

I obviously think this is the RightThing(tm) to do, but it's probably
too late for 6.3 since there is the worry that somebody "knows" that
it's a IOVEC somewhere.

Even if it sounds unlikely, and wrong.

Adding Al, who tends to be the main iovec person.

Al, see

   https://lore.kernel.org/all/20230324204443.45950-1-axboe@kernel.dk/

for the series if you didn't already see it on fsdevel.

                  Linus