Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 11 Jan 1998 20:51:20 -0500
From:      dmaddox@scsn.net (Donald J. Maddox)
To:        John Kelly <jak@cetlink.net>
Cc:        "John S. Dyson" <toor@dyson.iquest.net>, hackers@FreeBSD.ORG
Subject:   Re: 16650 Support(?)
Message-ID:  <19980111205120.42545@scsn.net>
In-Reply-To: <34c0795e.7050337@mail.cetlink.net>; from John Kelly on Mon, Jan 12, 1998 at 02:07:04AM %2B0000
References:  <199801120007.TAA00316@dyson.iquest.net> <34c0795e.7050337@mail.cetlink.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jan 12, 1998 at 02:07:04AM +0000, John Kelly wrote:
> On Sun, 11 Jan 1998 19:07:35 -0500 (EST), "John S. Dyson"
> <toor@dyson.iquest.net> wrote:
> 
> >> The 650 support seems to be broken, so don't flag it as a 650.  Run it
> >> as a 550 and it should work fine.  You still get the benefit of the
> >> deeper FIFO, even when it's defined as a 550.  You don't get the auto
> >> CTS/RTS flow control, but that has questionable value anyway.
> >> 
> >I have a 16650 based card, and it appears to work well.  It would be interesting
> >to figure out why mine works, and others don't.
> >
> 
> Do you have it flagged as a 650 in your kernel, or as a 550?
> 
> I emailed you about the SIO 650 support a couple of months ago but I
> guess you were busy with other stuff.  There seem to be some changes
> in SIO for 650 support, attributed to you.  If that is true, can you
> describe the changes?
> 
> When I tried to use the 650 support in -current a couple of months ago
> I would always get interrupt-level buffer overflows.  But as long as
> it was defined as a 550 it worked (and still works) fine.

Interesting that you mention that...  Now that I have this thing actually
working (finally), I see this just moments ago:

Jan 11 20:25:40 rhiannon /kernel: sio2: 409 more tty-level buffer overflows (total 409)
Jan 11 20:25:40 rhiannon /kernel: sio2: 409 more tty-level buffer overflows (total 409)

It's still configured with flags=0x20000 for the moment.  Guess I'll drop
the flag and see what happens...




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980111205120.42545>