Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 16 Nov 2009 17:59:35 +0200
From:      Alexander Motin <mav@FreeBSD.org>
To:        Dieter <freebsd@sopwith.solgatos.com>
Cc:        freebsd-hardware@freebsd.org
Subject:   Re: Adaptec 1405 on FreeBSD
Message-ID:  <4B0176E7.7080204@FreeBSD.org>
In-Reply-To: <1258363380.00183909.1258351203@10.7.7.3>
References:  <1258363380.00183909.1258351203@10.7.7.3>

next in thread | previous in thread | raw e-mail | index | archive | help
Dieter wrote:
>>> - SiI3124-based - fast and functional. It is actually PCI-X one, but
>>> there are many boards with built-in PCIe bridges.
> 
> Do any of these fit in a x1 slot?

I was surprised, but yes. And as expected, bus limits it performance
badly. But it is still works much faster then 3132 at the same bus.

>>> - First generation of SiI chips (SiI3114). They are quite old - SATA1
>>> and PCI, but they are long-time supported and they take all possible
>>> from PCI bus, and in 66MHz PCI-X slot can give even more. But I have
>>> heard some negative comments about them.
> 
> I have a 3512 card in a NetBSD box and it is slow but otherwise works fine.
> I have read various complaints about 1st gen SiL chips with FreeBSD,
> so do your homework before getting one for a FreeBSD box.  Perhaps the
> new drivers in 8.0 will fix these problems?

I haven't seen original chip errata to be sure, but ata(4) driver now
includes some workarounds for this chip. Same time, there is no such
quirks for 3114.

> 7.x doesn't support NCQ either.  :-(  I'm waiting impatiently for
> 8.0 to come out. as I need NCQ.  Speaking of which, I've read that
> some controller/disk combinations have problems with NCQ?  Is there
> a way to turn NCQ on/off on a per-disk basis?

It is already possible in HEAD to write quirks for specific device
models and firmware revisions. Also to control NCQ usage.

-- 
Alexander Motin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4B0176E7.7080204>