linux-kernel - Re: [PATCH net-next v1] net: stmmac: xgmac: increase length limit of descriptor ring

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 for Android: free password hash cracker in your pocket

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

Message-ID: <788c2744-e1c0-4338-9b86-9119d31841f1@lunn.ch>
Date: Fri, 21 Jun 2024 05:12:27 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Furong Xu <0x1207@...il.com>
Cc: Serge Semin <fancer.lancer@...il.com>,
	"David S. Miller" <davem@...emloft.net>,
	Alexandre Torgue <alexandre.torgue@...s.st.com>,
	Jose Abreu <joabreu@...opsys.com>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	Maxime Coquelin <mcoquelin.stm32@...il.com>,
	Joao Pinto <jpinto@...opsys.com>, Simon Horman <horms@...nel.org>,
	netdev@...r.kernel.org, linux-stm32@...md-mailman.stormreply.com,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	xfr@...look.com, rock.xu@....com
Subject: Re: [PATCH net-next v1] net: stmmac: xgmac: increase length limit of
 descriptor ring

> Heavy incoming traffic on some heavy load system, the max 1024 limit defined
> by DMA_MAX_RX_SIZE in drivers/net/ethernet/stmicro/stmmac/common.h is too
> few to achieve high throughput for XGMAC.
> With this patch, ethtool can set a new length than 1024

Please include some benchmark results to show the improvement.

But at some point, more buffers don't help you. If you are
consistently overloaded, you will still overflow the buffers.  So you
might want to look at where is the bottleneck and how do you
prioritise processing packets over whatever else is loading the
system.

Maybe this would help, if the bus is the problem:

https://www.spinics.net/lists/netdev/msg1006370.html

	Andrew