linux-kernel - Re: 2.6.38: rt2800usb: high latency (1000ms)?

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: <4DA68354.4060603@gmail.com>
Date:	Thu, 14 Apr 2011 09:17:08 +0400
From:	Igor Plyatov <plyatov@...il.com>
To:	Justin Piszcz <jpiszcz@...idpixels.com>
CC:	Daniel Halperin <dhalperi@...washington.edu>,
	Ivo Van Doorn <ivdoorn@...il.com>, linux-usb@...r.kernel.org,
	linux-wireless@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: 2.6.38: rt2800usb: high latency (1000ms)?

Hi!

> On Wed, 13 Apr 2011, Daniel Halperin wrote:
>
>> On Wed, Apr 13, 2011 at 10:57 AM, Justin Piszcz 
>> <jpiszcz@...idpixels.com> wrote:
>>> Hi,
>>>
>>> When powersave is enabled, it is very jumpy, I've used satellite comms
>>> before
>>> and (~600ms-1200ms was more smooth) as it did not jump around as 
>>> much.  The
>>> application is just a standalone desktop with minimal activity for the
>>> majority
>>> of the time, maybe thats why..
>>>
>>> With powersave disabled, I now see 0% packet loss (802.11n) and low 
>>> ping
>>> times, this looks like the proper solution for the wireless USB 
>>> device I
>>> am using.  By the way, is it possible/are there wireless USB devices 
>>> out
>>> there
>>> that support wake on wireless lan (WOWL?
>>>
>>> Your ping command with power off:
>>> 1408 bytes from server (192.168.1.2): icmp_req=539 ttl=64 time=1.07 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=540 ttl=64 time=1.31 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=541 ttl=64 time=1.07 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=542 ttl=64 time=1.26 ms
>>>
>>> Your ping command with power on:
>>> 1408 bytes from server (192.168.1.2): icmp_req=649 ttl=64 time=1.80 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=650 ttl=64 time=1.85 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=651 ttl=64 time=2.86 ms
>>> 1408 bytes from server (192.168.1.2): icmp_req=652 ttl=64 time=1.46 ms
>>>
>>> You are correct, if there is a lot of traffic, its good, but if the 
>>> system
>>> is relatively idle and all that's going on is an SSH session, there is
>>> horrible
>>> latency.
>>
>> Gotcha. I might still look around in the network stack and/or driver
>> and see what the time constants are. For instance:
>>
>> (1) What is the AP's beacon period and DTIM? Typical values are 100
>> TUs for beacons (102.4 ms) and 2 for DTIM (2 beacons per power-save
>> wakeup) which should imply a mean of 100 and max of 200 ms delay even
>> on pings.
> I am using a WNDR3700 with default settings in terms of beacons/etc.  No
> issues with any device (laptop, computer, etc (in windows)), I have two
> wireless USB adapters (bought two) and in Windows, no problems, I 
> don't think
> it is the WNDR3700.  As far as linux/wpa-supplicant, using default 
> settings.
>
>
>>
>> (2) How long does the client wait after waking up to go back to sleep?
>> It should be at least a few seconds. For ssh, then, you should see
>> something like a 100-200 ms delay for the first key and then nothing
>> at all unless you stop typing for a bit.
> It lags with each word I type, it is terrible.  If I run something like
> dmesg or ps auxww, the entire session freezes for 5-10 seconds before it
> comes back.

Yes, it's terrible. I have the same problems with ping and ssh with 
latest GIT
linux on my ARM platform and two different USB Wi-Fi adapters based on
RT3070 chipset ("D-Link DWA-125" and "Qcom LR802UKN3").

>
>>
>> I'm SSHing over a Wi-Fi link that uses power save right this second,
>> and have for years. It's not generally an issue, I suspect something
>> worse is going on.
> Maybe the wireless usb adapters do not function well in Linux with 
> power save
> on.
>
> I bought them awhile ago, they had the highest reviews, and in Windows,
> they did do 10-15MiB/s, in Linux, I see ~4.6MiB/s (but that was with 
> power save on) about the same, 4.5MiB/s.
> http://www.amazon.com/Medialink-Wireless-Adapter-802-11n-Compatible/dp/B002RM08RE 
>
>
> 100+0 records in
> 100+0 records out
> 104857600 bytes (105 MB) copied, 23.2803 s, 4.5 MB/s
>
> Justin.

Best regards!
--
Igor Plyatov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/