getting to 4K disk blocks in ZFS

[Steven Hartland]

----- Original Message ----- 
> From: "Justin T. Gibbs" <gibbs at scsiguy.com>
> To: "Willem Jan Withagen" <wjw at digiware.nl>
> Cc: "Steven Hartland" <killing at multiplay.co.uk>; <freebsd-stable at freebsd.org>; "Andriy Gapon" <avg at freebsd.org>; "Peter Wemm" 
> <peter at wemm.org>; "Aristedes Maniatis" <ari at ish.com.au>
> Sent: Saturday, September 20, 2014 12:07 AM
> Subject: Re: getting to 4K disk blocks in ZFS
>
>
> On Sep 11, 2014, at 5:32 PM, Willem Jan Withagen <wjw at digiware.nl> wrote:
>
> > On 11-9-2014 19:49, Peter Wemm wrote:
> >>> Another downside is 1/4th of uberblocks, 32 vs 128.
> >>> Also, automatic sector size detection works great for me and I've never had
> >>> a need to manually tweak ashift.
> >>
> >> Unfortunately, I have.  Same drive connected two different ways:
> >>
> >> da12 at mps1 bus 0 scbus1 target 11 lun 0
> >> da12: <ATA ST4000VN000-1H41 SC43> Fixed Direct Access SCSI-6 device
> >> da12: 600.000MB/s transfers
> >> da12: Command Queueing enabled
> >> da12: 3815447MB (7814037168 512 byte sectors: 255H 63S/T 486401C)
> >>
> >> ada1 at ahcich1 bus 0 scbus3 target 0 lun 0
> >> ada1: <ATA ST4000VN000-1H41 SC43> ATA-8 SATA 3.x device
> >> ada1: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 8192bytes)
> >> ada1: Command Queueing enabled
> >> ada1: 3815447MB (7814037168 512 byte sectors:  16H 63S/T 16383C)
> >> ada1: quirks=0x1<4K>
> >>
> >> The 4k flag is missing when it's on the sas controller.  The Ident strings are
> >> changed.
> >>
> >> This came up elsewhere recently.
> >
> > I reported the same fact for the new set of WD REDs I installed.
> > Seems that ada and da have different quirks tables...
> > So disks on SATA connectors on the motherboard are diagnosed as being 4Kb.
> > The disks on my twa don't get the quirk and are considered 512b
> >
> > —WjW
>
> I’m surprised that we have to constantly add quirks.  Are these drives really
> failing to report their ata params correctly?  Is there a reason we don’t
> currently utilize the ata params data (which is already fetched for trim/unmap
> detection) to also set lbppbe (logical block per physical block exponent) and
> lalba (lowest aligned lba)?  We may find that many of the existing quirks are
> unnecessary if we fix the probe code.

On the contary I've not found a single drive which reports 4k sectors on its
own, every single one that I've seen report 4k is because we've added a
quirk for it :(

    Regards
    Steve