netdev - Re: [PATCH net-next] net: zerocopy: combine pages in zerocopy_sg_from

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] [day] [month] [year] [list]

Date:   Thu, 20 Aug 2020 16:13:33 -0700 (PDT)
From:   David Miller <davem@...emloft.net>
To:     edumazet@...gle.com
Cc:     netdev@...r.kernel.org, eric.dumazet@...il.com, willemb@...gle.com
Subject: Re: [PATCH net-next] net: zerocopy: combine pages in
 zerocopy_sg_from_iter()

From: Eric Dumazet <edumazet@...gle.com>
Date: Thu, 20 Aug 2020 08:43:59 -0700

> Currently, tcp sendmsg(MSG_ZEROCOPY) is building skbs with order-0 fragments.
> Compared to standard sendmsg(), these skbs usually contain up to 16 fragments
> on arches with 4KB page sizes, instead of two.
> 
> This adds considerable costs on various ndo_start_xmit() handlers,
> especially when IOMMU is in the picture.
> 
> As high performance applications are often using huge pages,
> we can try to combine adjacent pages belonging to same
> compound page.
> 
> Tested on AMD Rome platform, with IOMMU, nominal single TCP flow speed
> is roughly doubled (~55Gbit -> ~100Gbit), when user application
> is using hugepages.
> 
> For reference, nominal single TCP flow speed on this platform
> without MSG_ZEROCOPY is ~65Gbit.
> 
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>

Applied, the refcounitng in these kinds of patchs is always fun to
audit :-)