All of lore.kernel.org
 help / color / mirror / Atom feed
From: Linda Walsh <lkml@tlinx.org>
To: linux-ide@vger.kernel.org
Subject: Port Multiplier access with Sil 3124
Date: Sun, 08 Feb 2009 12:38:14 -0800	[thread overview]
Message-ID: <498F42B6.8030607@tlinx.org> (raw)

Is there something different, from normal disk-access, that I need to do
to access hard disks beyond '1', on a port-multiplier?

I thought I remembered reading the port multiplier support was
working for many SATA and SATA RAID controller capable chipsets,
including the Sil 3124.

I picked up a 2-Bay external SATA enclosure that I'm trying to access in
(what I thought) was the simplest mode: "JBOD".  However, when I boot,
I am only seeing the first hard disk.

Experimenting, I tried a single hard disk in both positions -- one
position let me see the disk directly (as though it was a direct,
str8-thru connection), the other position showed up detected by
the boot BIOS as a 7MB HD by some unrecognized vendor.   In
linux, I'm able to access and use the hard disk when it appears
'str8-thru', but linux sees nothing concerning the 7MB pseudo HD.

Is my expectation that the driver would simply recognize the
external enclosure by whatever I had the external enclosure set to,
too optimistic?  Do I need to run some special util to setup the disks in
JBOD mode?  I guess I thought I only needed to worry about
'special utils' if I was using the disk-pair in a RAID config (0/1)...

It seems there should be a linux util to manage the "container",
'sil57xx'  --  I take it is not used for RAID-only config?

My ultimate aim is to use it in a RAID-0, mirror config (my luck
with SATA disk drives has been abysmal, of late (*sigh*)).

Anyone with any real-world experience about when the 3Gb SAS
starts to become a bottleneck?  I know that theoretically, it could
support a hair over 350MB/s if there was no overhead, which would
reliably only support 2 hard disks at full speed (assuming ~120MB/s
max linear read speed/disk).  Does that jive with people's real-world 
experience?  I.e. port-multipliers can provide full throughput for
2-HD's but not likely 3? 

Should I be looking for an sil57xx program somewhere (the box contained
a mini-CD, but it looks like a driver for an older kernel (2.6.9).  Not 
so sure about it's usefulness in my setup.

Thanks,
-linda


             reply	other threads:[~2009-02-08 21:08 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-02-08 20:38 Linda Walsh [this message]
2009-02-09  9:26 ` Port Multiplier access with Sil 3124 Grant Grundler
2009-02-09 22:01   ` Linda Walsh
2009-02-09 14:35 ` Greg Freemyer
2009-02-09 20:09   ` Mark Lord
2009-02-09 22:22   ` Linda Walsh
2009-02-09 23:01     ` Greg Freemyer
2009-02-10  0:15       ` Jeff Garzik
2009-02-10  1:07         ` Linda Walsh
2009-02-10  3:48           ` Jeff Garzik
2009-02-11  2:09           ` Tejun Heo
2009-02-12  4:16             ` Robert Hancock
2009-02-10  3:25         ` Greg Freemyer
2009-02-11  2:12 ` Tejun Heo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=498F42B6.8030607@tlinx.org \
    --to=lkml@tlinx.org \
    --cc=linux-ide@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.