Message-ID: <522ED1A6.9020000@onera.fr>
Date: Tue, 10 Sep 2013 10:00:38 +0200
From: Paul Chavent <Paul.Chavent@onera.fr>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130801 Thunderbird/17.0.8
MIME-Version: 1.0
To: linux-usb@vger.kernel.org, gregkh@linuxfoundation.org, jhovold@gmail.com,
        fschaefer.oss@googlemail.com, jslaby@suse.cz, max@suse.de,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/5] Enable PPS reporting for USB serial devices
References: <1378742480-2146-1-git-send-email-paul.chavent@onera.fr> <20130910073105.GB5030@enneenne.com>
In-Reply-To: <20130910073105.GB5030@enneenne.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1545
Lines: 45

Hi.

At the bottom of the  description of the patch set, i warn about the 
latencies and jitter.

I did "real" test with a pps generated by a workstation (i haven't 
tested with a precise pps yet), so the the jitter don't mean anything. 
But the magnitude of difference between uart and usb pps isn't 
negligible (~400?s with uart compared to ~1ms with usb).

Do you think that, depending on the application, such clock instability 
is a flaw ?

I will run new tests with a precise pps source later.

I will also try to measure the latency of the pss signal through the 
usb, and see if it is constant. Perhaps we could introduce parameters to 
add offset to the pps signal used by the kernel consumer (the same way 
we can add the offset with PPS_FETCH ioctl) later.

Regards.

Paul.

On 09/10/2013 09:31 AM, Rodolfo Giometti wrote:
> On Mon, Sep 09, 2013 at 06:01:15PM +0200, Paul Chavent wrote:
>> Hi.
>>
>> This series enable the PPS reporting for USB serial devices.
>
> I have nothing against with this solution but consider that reporting
> a PPS signal through USB bus may add unknown delays that may vanish
> clock stability... as far as I know you should not use USB serial port
> with PPS. Have you some statistics about clock stability in a real
> implementation?
>
> Ciao,
>
> Rodolfo
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/