[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Y+VYfHcNdvez6M2a@jeremy-acer>
Date: Thu, 9 Feb 2023 12:33:00 -0800
From: Jeremy Allison <jra@...ba.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Stefan Metzmacher <metze@...ba.org>, Jens Axboe <axboe@...nel.dk>,
Linux API Mailing List <linux-api@...r.kernel.org>,
Samba Technical <samba-technical@...ts.samba.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Al Viro <viro@...iv.linux.org.uk>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>,
io-uring <io-uring@...r.kernel.org>
Subject: Re: copy on write for splice() from file to pipe?
On Thu, Feb 09, 2023 at 11:48:35AM -0800, Linus Torvalds via samba-technical wrote:
>
>So this is exactly *why* splicing from a file all the way to the
>network will then show any file changes that have happened in between
>that "splice started" and "network card got the data". You're supposed
>to use splice only when you can guarantee the data stability (or,
>alternatively, when you simply don't care about the data stability,
>and getting the changed data is perfectly fine).
Metze, correct me if I'm wrong but isn't this exactly the "file
is leased in SMB3" case ?
We already know if a file is leased, and so only use the splice calls
for I/O in that case, and fall back to the slower calls in the
non-leased case.
Powered by blists - more mailing lists