On 1/27/25 18:54, Jens Axboe wrote:
On 1/27/25 11:21 AM, Andres Freund wrote:
That's precisely what io-uring fixed buffers do :)
I looked at using them at some point - unfortunately it seems that there is
just {READ,WRITE}_FIXED not {READV,WRITEV}_FIXED. It's *exceedingly* common
for us to do reads/writes where source/target buffers aren't wholly
contiguous. Thus - unless I am misunderstanding something, entirely plausible
- using fixed buffers would unfortunately increase the number of IOs
noticeably.
Should have sent an email about that...
I guess we could add some heuristic to use _FIXED if it doesn't require
splitting an IO into too many sub-ios. But that seems pretty gnarly.
Adding Pavel, he's been working on support registered buffers with
readv/writev.
Yes, there are patches from last year out there, I'm going to pick
them up and, hopefully for 6.15. I'll keep you in the loop.
I dimly recall that I also ran into some around using fixed buffers as a
non-root user. It might just be the accounting of registered buffers as
mlocked memory and the difficulty of configuring that across
distributions. But I unfortunately don't remember any details anymore.
Should just be accounting.
--
Pavel Begunkov
