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] [thread-next>] [day] [month] [year] [list]
Message-ID: <cd0b3dec-af3f-af69-50b7-6ca6f7256900@gmx.de>
Date:   Mon, 18 Feb 2019 19:05:42 +0100
From:   Simon Huelck <simonmail@....de>
To:     Jose Abreu <jose.abreu@...opsys.com>,
        Martin Blumenstingl <martin.blumenstingl@...glemail.com>
Cc:     Emiliano Ingrassia <ingrassia@...genesys.com>,
        Gpeppe.cavallaro@...com, alexandre.torgue@...com,
        linux-amlogic@...ts.infradead.org, netdev@...r.kernel.org
Subject: Re: stmmac / meson8b-dwmac

Am 18.02.2019 um 18:05 schrieb Jose Abreu:
> On 2/18/2019 4:51 PM, Simon Huelck wrote:
>> Hi,
>>
>> no im testing vanilla mainline kernel and against 4.14. where
>> performance was ok. but turning off EEE via ethtool should have same
>> results than removal of that patch.
>>
>> But, since it was mainlined recently , not long ago, i think this was
>> tested ??
> The thing is that phy_init_eee() is called by stmmac and this
> function does not take into account the broken modes if it's
> called too early which will cause driver to enable LPI in MAC.
>
> And anyway, if you have lpi values in the ethtool stats then EEE
> is being enabled by stmmac.
>
> Please try the change I suggested.
>
> Thanks,
> Jose Miguel Abreu

Hi,



disabling EEE doesnt help ( did it via the entry in the .dtb / .dts ),
the results are the same. I can confirm the LPI counters are zero or one
only after the test.....

C:\Users\Simon\Downloads\iperf-2.0.9-win64>iperf -c 10.10.11.1 -i1  -d
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size:  208 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to 10.10.11.1, TCP port 5001
TCP window size:  208 KByte (default)
------------------------------------------------------------
[  4] local 10.10.11.100 port 56114 connected with 10.10.11.1 port 5001
[  5] local 10.10.11.100 port 5001 connected with 10.10.11.1 port 47866
[ ID] Interval       Transfer     Bandwidth
[  4]  0.0- 1.0 sec  20.1 MBytes   169 Mbits/sec
[  5]  0.0- 1.0 sec  75.8 MBytes   635 Mbits/sec
[  4]  1.0- 2.0 sec  18.1 MBytes   152 Mbits/sec
[  5]  1.0- 2.0 sec  69.0 MBytes   579 Mbits/sec
[  4]  2.0- 3.0 sec  15.6 MBytes   131 Mbits/sec
[  5]  2.0- 3.0 sec  70.7 MBytes   593 Mbits/sec
[  4]  3.0- 4.0 sec  15.9 MBytes   133 Mbits/sec
[  5]  3.0- 4.0 sec  70.3 MBytes   590 Mbits/sec
[  4]  4.0- 5.0 sec  16.4 MBytes   137 Mbits/sec
[  5]  4.0- 5.0 sec  69.5 MBytes   583 Mbits/sec
[  4]  5.0- 6.0 sec  16.5 MBytes   138 Mbits/sec
[  5]  5.0- 6.0 sec  69.0 MBytes   579 Mbits/sec
[  4]  6.0- 7.0 sec  18.2 MBytes   153 Mbits/sec
[  5]  6.0- 7.0 sec  69.8 MBytes   585 Mbits/sec
[  4]  7.0- 8.0 sec  15.6 MBytes   131 Mbits/sec
[  5]  7.0- 8.0 sec  70.0 MBytes   587 Mbits/sec
[  4]  8.0- 9.0 sec  16.2 MBytes   136 Mbits/sec
[  5]  8.0- 9.0 sec  69.6 MBytes   583 Mbits/sec
[  4]  9.0-10.0 sec  17.1 MBytes   144 Mbits/sec
[  4]  0.0-10.0 sec   170 MBytes   142 Mbits/sec
[  5]  9.0-10.0 sec  69.1 MBytes   580 Mbits/sec
[  5]  0.0-10.0 sec   703 MBytes   589 Mbits/sec



regards,

Simon


View attachment "ethtool_test2.txt" of type "text/plain" (5373 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ