netdev - Re: [PATCH net-next v1 1/6] lan743x: boost performance on cpu archs w/o dma cache snooping

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]

Date:   Fri,  5 Feb 2021 18:09:35 +0300
From:   Sergej Bauer <sbauer@...ckbox.su>
To:     thesven73@...il.com
Cc:     andrew@...n.ch, Markus.Elfring@....de, rtgbnm@...il.com,
        tharvey@...eworks.com, anders@...ningen.priv.no,
        sbauer@...ckbox.su,
        Bryan Whitehead <bryan.whitehead@...rochip.com>,
        UNGLinuxDriver@...rochip.com (maintainer:MICROCHIP LAN743X ETHERNET
        DRIVER), "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        netdev@...r.kernel.org (open list:MICROCHIP LAN743X ETHERNET DRIVER),
        linux-kernel@...r.kernel.org (open list)
Subject: Re: [PATCH net-next v1 1/6] lan743x: boost performance on cpu archs w/o dma cache snooping

On Friday, February 5, 2021 5:07:22 PM MSK you wrote:
> Hi Sergej,
> 
> On Fri, Feb 5, 2021 at 7:44 AM Sergej Bauer <sbauer@...ckbox.su> wrote:
> > Hi Sven
> > I can confirm great stability improvement after your patch
> > "lan743x: boost performance on cpu archs w/o dma cache snooping".
> > 
> > Test machine is Intel Pentium G4560 3.50GHz
> > lan743x with rejected virtual phy 'inside'
> 
> Interesting, so the speed boost patch seems to improve things even on
> Intel...
> 
> Would you be able to apply and test the multi-buffer patch as well?
> To do that, you can simply apply patches [2/6] and [3/6] on top of
> what you already have.
> 

Hi Sven
Tests after applying patches [2/6] and [3/6] are:
$ ifmtu eth7 500
$ sudo test_ber -l eth7 -c 1000 -n 1000000 -f500 --no-conf
...
number of sent packets      = 1000000
number of received packets  = 713288
number of lost packets = 286712
number of out of order packets = 0
number of bit errors   = 0
total errors detected  = 286712
bit error rate         = 0.286712
average speed: 427.8043 Mbit/s

$ ifmtu eth7 1500
$ sudo test_ber -l eth7 -c 1000 -n 1000000 -f500 --no-conf
...
number of sent packets      = 1000000
number of received packets  = 707869
number of lost packets = 292131
number of out of order packets = 0
number of bit errors   = 0
total errors detected  = 292131
bit error rate         = 0.292131
average speed: 431.0163 Mbit/s

$ sudo test_ber -l eth7 -c 1000 -n 1000000 -f1500 --no-conf
...
number of sent packets      = 1000000
number of received packets  = 1000000
number of lost packets = 0
number of out of order packets = 0
number of bit errors   = 0
total errors detected  = 0
bit error rate         = 0
average speed: 646.4932 Mbit/s

> Keeping in mind that Bryan has identified an issue with the above
> patch, which will get fixed in v2. So YMMV.
I'll perform tests with v2 as well.