netdev - RE: [PATCH 1/2] stmmac: introduce flag to dynamically disable TX offload for rockchip devices

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: <1554305721.1328.0@gmail.com>
Date:   Wed, 03 Apr 2019 18:35:21 +0300
From:   "Leonidas P. Papadakos" <papadakospan@...il.com>
To:     Jose Abreu <jose.abreu@...opsys.com>
Cc:     Robin Murphy <robin.murphy@....com>,
        Philipp Tomsich <philipp.tomsich@...obroma-systems.com>,
        Heiko Stübner <heiko@...ech.de>,
        Christoph Müllner 
        <christoph.muellner@...obroma-systems.com>,
        Maxime Coquelin <mcoquelin.stm32@...il.com>,
        Alexandre Torgue <alexandre.torgue@...com>,
        netdev@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Klaus Goger <klaus.goger@...obroma-systems.com>
Subject: RE: [PATCH 1/2] stmmac: introduce flag to dynamically disable TX
 offload for rockchip devices



Στις Τετ, 3 Απρ, 2019 at 10:55 ΠΜ, ο/η Jose Abreu 
<jose.abreu@...opsys.com> έγραψε:
> From: Leonidas P. Papadakos <papadakospan@...il.com>
> Date: Tue, Apr 02, 2019 at 23:48:16
> 
>>  I can confirm that snps,no-pbl-x8, currently set by uboot for the 
>> fdt
>>  basically fixes the tx-checksumming issue.
>>  I get 700~800 Mbps myself but that might come down to the setup.
>>  You could say it affects performance slightly, as I saw in a 
>> previous
>>  email.
>> 
>>  Without the setting The rate consistently drops to 0.
>>  ethtool reports tx-checksumming is on in both cases
>> 
>>  Pretty cool!
>>  Does that give you guys an idea on how to tackle it "officially"?
>> 
>> 
>> 
> 
> Thank you all for testing !
> 
> It's been on my todo list to cook up a patch that limits PBL 
> according to
> FIFO size but I didn't manage to have the time yet so for now it would
> be better if you fix it by submitting the DT bindings changes.
> 
> Thanks,
> Jose Miguel Abreu

If snps,no-pbl-x8 does indeed have a performace hit, should the 
workaround be to
a) turn tx checksumming off or
b) snps,no-pbl-x8 set?

I'll test to see the difference