[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <608853.1692190847@warthog.procyon.org.uk>
Date: Wed, 16 Aug 2023 14:00:47 +0100
From: David Howells <dhowells@...hat.com>
To: David Laight <David.Laight@...LAB.COM>
Cc: dhowells@...hat.com, Al Viro <viro@...iv.linux.org.uk>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Jens Axboe <axboe@...nel.dk>, Christoph Hellwig <hch@...t.de>,
"Christian Brauner" <christian@...uner.io>,
Matthew Wilcox <willy@...radead.org>,
"Jeff Layton" <jlayton@...nel.org>,
"linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
"linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
"linux-mm@...ck.org" <linux-mm@...ck.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 2/2] iov_iter: Don't deal with iter->copy_mc in memcpy_from_iter_mc()
David Laight <David.Laight@...LAB.COM> wrote:
>
> Couldn't the relevant code directly call __copy_from_iter_mc() ?
> Or a version then checked iov_is_copy_mc() and then fell
> back to the standard function.
No, because the marked iterator is handed by the coredump code to
__kernel_write_iter() and thence on to who-knows-what driver - which will call
copy_from_iter() or some such. $DRIVER shouldn't need to know about
->copy_mc.
One thing I do wonder about, though, is what should happen if they call, say,
csum_and_copy_from_iter()? That doesn't have an _mc variant. Or what if they
extract the pages and operate directly on those?
David
Powered by blists - more mailing lists