[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <117689.1718956457@warthog.procyon.org.uk>
Date: Fri, 21 Jun 2024 08:54:17 +0100
From: David Howells <dhowells@...hat.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: dhowells@...hat.com, Sagi Grimberg <sagi@...mberg.me>,
Jakub Kicinski <kuba@...nel.org>,
Matthew Wilcox <willy@...radead.org>,
kernel test robot <oliver.sang@...el.com>, oe-lkp@...ts.linux.dev,
lkp@...el.com, netdev@...r.kernel.org
Subject: Re: [PATCH] net: micro-optimize skb_datagram_iter
Eric Dumazet <edumazet@...gle.com> wrote:
> But the trace involves af_unix, not GRO ?
Ah. splice()/vmsplice() can feed multipage folios into the socket. Possibly
those can be in highmem.
David
Powered by blists - more mailing lists