Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 27 Sep 2012 11:07:26 -0400
From:      Mike Tancsa <mike@sentex.net>
To:        freebsd-questions <freebsd-questions@freebsd.org>
Subject:   Re: LSI 9750-4i (tws based cards)
Message-ID:  <50646BAE.20102@sentex.net>
In-Reply-To: <5050E2EC.8070309@sentex.net>
References:  <5050E2EC.8070309@sentex.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 9/12/2012 3:30 PM, Mike Tancsa wrote:
> Does anyone have any experience with these cards ? We are looking for a
> controller that has a little more gas than the twa based cards which
> have been very reliable and stable for us on FreeBSD.  I dont have any
> experience with 3ware/LSI's cards that use the tws driver.  Has anyone
> used them yet  ?

For the archives...


I ordered a 3ware 9750 4i card to test with and its quite fast!  There
is a small bug in the driver fixed now in HEAD as well as some cosmetic
changes.  But other than that it seems pretty solid.  The same
management interface as the twa and twe based cards.

I ran a test box using a kernel with INVARIANTS and WITNESS with the
card and 4 10k disks in raid 10.  The card seems pretty zippy for the
price.  RW performance does seem to take advantage of the faster disk speeds


0{3w9750}# dd if=/dev/zero of=/mnt/test bs=1024k count=9000
9000+0 records in
9000+0 records out
9437184000 bytes transferred in 39.859600 secs (236760629 bytes/sec)
0{3w9750}#
0{3w9750}# umount /mnt
0{3w9750}# mount /dev/da0 /mnt
0{3w9750}# dd if=/mnt/test of=/dev/null bs=1024k
9000+0 records in
9000+0 records out
9437184000 bytes transferred in 27.887930 secs (338396720 bytes/sec)
0{3w9750}#

For stress testing, I ran the disk.cfg component of
http://people.freebsd.org/~pho/stress/index.html

as well as random copies of dbench and bonnie as well as periodically
accessing the disk while the stress scripts ran for 72hrs.  The OS was
netbooted, RELENG9 AMD64

0{3w9750}# tw_cli "/c0 show"

Unit  UnitType  Status         %RCmpl  %V/I/M  Stripe  Size(GB)  Cache
AVrfy
------------------------------------------------------------------------------
u0    RAID-10   OK             -       -       256K    931.303   RiW
ON

VPort Status         Unit Size      Type  Phy Encl-Slot    Model
------------------------------------------------------------------------------
p0    OK             u0   465.76 GB SATA  0   -            WDC
WD5002AALX-00J3
p1    OK             u0   465.76 GB SATA  1   -            WDC
WD5002AALX-00J3
p2    OK             u0   465.76 GB SATA  2   -            WDC
WD5002AALX-00J3
p3    OK             u0   465.76 GB SATA  3   -            WDC
WD5002AALX-00J3

0{3w9750}#



For some reason the card defaults legacy interrupts.  Adding
hw.tws.enable_msi=1 to /boot/loader.conf fixes that


LSI 3ware device driver for SAS/SATA storage controllers, version:
10.80.00.003
tws0: <LSI 3ware SAS/SATA Storage Controller> port 0x4000-0x40ff mem
0xc2460000-0xc2463fff,0xc2400000-0xc243ffff irq 17
at device 0.0 on pci2
tws0: Using MSI
tws0: Controller details: Model 9750-4i, 8 Phys, Firmware FH9X
5.12.00.007, BIOS BE9X 5.11.00.006
(probe65:tws0:0:65:0): INQUIRY. CDB: 12 0 0 0 24 0
(probe65:tws0:0:65:0): CAM status: Invalid Target ID
(probe65:tws0:0:65:0): Error 22, Unretryable error
da0 at tws0 bus 0 scbus0 target 0 lun 0
da0: <LSI 9750-4i    DISK 5.12> Fixed Direct Access SCSI-5 device
da0: 6000.000MB/s transfers
da0: 953654MB (1953083392 512 byte sectors: 255H 63S/T 121573C)
tws0@pci0:2:0:0:        class=0x010400 card=0x000113c1 chip=0x101013c1
rev=0x05 hdr=0x00
    vendor     = '3ware Inc'
    device     = '9750 SAS2/SATA-II RAID PCIe'
    class      = mass storage
    subclass   = RAID
    bar   [10] = type I/O Port, range 32, base 0x4000, size 256, enabled
    bar   [14] = type Memory, range 64, base 0xc2460000, size 16384, enabled
    bar   [1c] = type Memory, range 64, base 0xc2400000, size 262144,
enabled
    cap 01[50] = powerspec 3  supports D0 D1 D2 D3  current D0
    cap 10[68] = PCI-Express 2 endpoint max data 128(4096) link x4(x8)
    cap 03[d0] = VPD
    cap 05[a8] = MSI supports 1 message, 64 bit enabled with 1 message
ecap 0001[100] = AER 1 1 fatal 0 non-fatal 0 corrected
ecap 0004[138] = unknown 1


In summary, we like the card on FreeBSD.  We make heavy use of the older
3ware cards in our company on various platforms, so our staff are
comfortable using the management tools to swap out dead drives.  We will
probably start to use these cards for customer builds in the future
where they need faster IO.

	---Mike



-- 
-------------------
Mike Tancsa, tel +1 519 651 3400
Sentex Communications, mike@sentex.net
Providing Internet services since 1994 www.sentex.net
Cambridge, Ontario Canada   http://www.tancsa.com/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?50646BAE.20102>