netdev - Re: [PATCH net-next 06/11] net: page_pool: avoid calling no-op externals when possible

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <313c4834-bcee-1182-7094-4feef800c312@intel.com>
Date: Thu, 18 May 2023 15:26:15 +0200
From: Alexander Lobakin <aleksander.lobakin@...el.com>
To: Christoph Hellwig <hch@....de>
CC: "David S. Miller" <davem@...emloft.net>, Eric Dumazet
	<edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo Abeni
	<pabeni@...hat.com>, Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
	Magnus Karlsson <magnus.karlsson@...el.com>, Michal Kubiak
	<michal.kubiak@...el.com>, Larysa Zaremba <larysa.zaremba@...el.com>, "Jesper
 Dangaard Brouer" <hawk@...nel.org>, Ilias Apalodimas
	<ilias.apalodimas@...aro.org>, <netdev@...r.kernel.org>,
	<intel-wired-lan@...ts.osuosl.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net-next 06/11] net: page_pool: avoid calling no-op
 externals when possible

From: Christoph Hellwig <hch@....de>
Date: Wed, 17 May 2023 10:14:58 +0200

> So while this looks ok, Eric Dumazet had looked into some optimization
> for this touching the core code, and promised me to come up with an
> even better version a while ago.  Eric, what's the state of your
> optimizations for no-op DMA syncs?
> 

Hmm, his last proposals were for avoiding indirect calls when IOMMU is
on, but in fact we don't need to synchronize stuff -- DMA IOMMU on
x86_64 also usually doesn't synchronize anything, but you don't know
that prior to doing an indirect call and dma_need_sync() won't help.
I was thinking of adding .dma_need_sync() callback to DMA ops, which
could also be called once only on page allocation, like in this patch.

Also want to hear how it goes for Eric :)

Thanks,
Olek