On Tue, Jun 10, 2014 at 10:05:56PM +0200, Mark Kettenis wrote:
> > Date: Tue, 10 Jun 2014 21:55:04 +0200
> > From: Otto Moerbeek <o...@drijf.net>
> >
> > On Tue, Jun 10, 2014 at 09:52:23PM +0200, Mark Kettenis wrote:
> >
> > > > Date: Tue, 10 Jun 2014 21:34:56 +0200
> > > > From: Otto Moerbeek <o...@drijf.net>
> > > >
> > > > On Fri, Jun 06, 2014 at 08:54:24PM +0200, Otto Moerbeek wrote:
> > > >
> > > > > This volumes feel pretty fast, so I suspect caching mode is OK. Still
> > > > > it is confusing to have a flag that doesn't reflect reality.
> > > > >
> > > > > I'm planning to upgrade the firmware next week. We'll see if that
> > > > > changes anything. BTW, al battery indicaters are healthy, no learning
> > > > > cycle going on or something like that.
> > > > >
> > > >
> > > > I just upgraded the perc firmware. After reboot, the volumes now
> > > > report they are in WB mode. We'll have to wait and see if it stays
> > > > that way.
> > >
> > > Which firmware version are you running now?
> >
> > mfi0: "PERC 6/i Integrated", firmware 6.3.3.0002, 256MB cache
>
> And the release notes for that firmware still have the blurb about
> RAID volumes always showing up as WT. Quality engineering!
>
The release note refers to the caching mode page of the SCSI target. That is,
as presented to the OS the disk appears to have write cache disabled. The
bioctl command is reaching down into the controller through a control path
to query what is actually configured. The rationale for this disconnected
implementation comes from other OS environments where software takes note of a
WB disk and pitches a fit without realizing that the WB cache is itself
battery backed. The (unfortunate) compromise is to lie about cache in the SCSI
domain.
