Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 28 Jan 2011 19:12:57 +1030
From:      "Daniel O'Connor" <doconnor@gsoft.com.au>
To:        Hans Petter Selasky <hselasky@c2i.net>
Cc:        freebsd-usb@freebsd.org
Subject:   Re: libusb performance on 8.1
Message-ID:  <2E5206C4-9327-4270-8948-FD44C62C645D@gsoft.com.au>
In-Reply-To: <201101280858.05077.hselasky@c2i.net>
References:  <9CF6C32F-E230-446B-94FC-C57F0F02B0E4@gsoft.com.au> <201101221433.23194.hselasky@c2i.net> <6AD22899-0B00-483D-A01E-786029A82C9F@gsoft.com.au> <201101280858.05077.hselasky@c2i.net>

next in thread | previous in thread | raw e-mail | index | archive | help

On 28/01/2011, at 18:28, Hans Petter Selasky wrote:
> For this kind of applications ISOCHRONOUS transfers should be used. =
Then you=20
> can have a double buffer guard in the range 1-56ms, regardless of the =
buffer=20
> size the hardware uses.

Hmm, OK, I will have a look at changing it, although it is a bit of a =
pain because you can't stall an ISO EP and I'm using stall to indicate =
an error.

Is it possible to change the amount that is buffered? This is a =
specialised application so a custom kernel is no problem. Even a hint in =
the right direction would be greatly appreciated :)

> You could also try an XHCI controller, because the BULK buffering is =
done=20
> differently there.

OK thanks, I'll try that too.

>=20
>> I obviously don't need any more
>> throughput, however my application is very sensitive to latency, as I =
am
>> reading out of a fairly small FIFO and if it fills up my entire run =
has to
>> be aborted.
>=20
>=20
>=20
> --HPS
>=20

--
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C









Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2E5206C4-9327-4270-8948-FD44C62C645D>