SAS and SATA under IRIX

This is premature but I'm too excited to wait. There's a bunch of LSI 3041X-R cards floating around here for around $25. So what the heck, it's an LSI 1064 chip, can't hurt to try.

Came, looks like a fake with a used LSI chip in the middle - either that or LSI product quality has gone down seriously. Tested it in an x86 kompewter first just in case

Integral SCSI controller 1: Version QL12160, single ended
  Fabric Disk: node d04b561107262b port 0 lun 0 on SCSI controller 1 (unit 2)
  Disk drive / removable media: unit 3 on SCSI controller 1
  CDROM: unit 6 on SCSI controller 1
Integral SCSI controller 2: Version SAS/SATA LS1064
Integral SCSI controller 0: Version QL12160, low voltage differential
  Disk drive: unit 1 on SCSI controller 0 (unit 1)
  Disk drive: unit 2 on SCSI controller 0 (unit 2)

  PCI Adapter ID (vendor 0x14e4, device 0x1645) PCI slot 1
  PCI Adapter ID (vendor 0x1000, device 0x0050) PCI slot 2
  PCI Adapter ID (vendor 0x1077, device 0x1216) PCI slot 1
  PCI Adapter ID (vendor 0x10a9, device 0x0005) PCI slot 2
  PCI Adapter ID (vendor 0x3388, device 0x0021) PCI slot 3
  PCI Adapter ID (vendor 0x10a9, device 0x0003) PCI slot 4
  PCI Adapter ID (vendor 0x11c1, device 0x5802) PCI slot 5
  PCI Adapter ID (vendor 0x1033, device 0x0035) PCI slot 8
  PCI Adapter ID (vendor 0x1033, device 0x0035) PCI slot 8
  PCI Adapter ID (vendor 0x1033, device 0x00e0) PCI slot 8
  PCI Adapter ID (vendor 0x104c, device 0x8024) PCI slot 12

The hex 1000, device 0050 is the LSI sata/sas card. I don't have a drive to check it with yet but this could be a low-cost answer to the ide-sata-no-scsi-dvdrw-expensive scsi disk problem on the 3xxx platforms, anyhow.

Btw, when did they change reporting an IEEE1394 disk into being a "Fabric Disk" on the wrong controller ? Seems a little strange.

Stay tuned for the further adventures of Dick and Jane ....

Integral SCSI controller 3: Version IEEE1394 SBP2
  Disk drive / removable media: unit 2 on SCSI controller 3
Integral SCSI controller 2: Version SAS/SATA LS1064
  Disk drive: unit 0 on SCSI controller 2
Integral SCSI controller 0: Version QL12160, low voltage differential
  Disk drive: unit 1 on SCSI controller 0
  Disk drive: unit 2 on SCSI controller 0
Integral SCSI controller 1: Version QL12160, single ended
  CDROM: unit 6 on SCSI controller 1

fool 4% df -k
Filesystem             Type  kbytes     use     avail  %use Mounted on
/dev/root               xfs 35705764 29037816  6667948  82  /
/dev/dsk/dks0d2s7       xfs 71676204 25612604 46063600  36  /usr/people/username/disktwo
/dev/dsk/dks2d0s7       xfs 488324920 209052828 279272092  43  /usr/people/username/sata

Been a week now, didn't do to-the-death testing but pushed the disk pretty hard, seems reliable and don't see any noticeable speed differences in operation from the 15k scsi disks. Set ioconfig.conf aside and rebooted, Irix re-ordered my hardware inventory to make more sense. Kinda liked having a Fabric Disk for a while tho.

SATA CD-ROM has some problems tho :( That was one of my major interests in this whole exercise. Grrr.

Additional Notes

Hi, finally got around to installing an SAS3041X-R card in my Fuel, after futzing around getting it seated it worked fine as well. Stuffed a 200GB Western Digital IDE drive through an IDE->SATA bridge (picked up a couple for $10 each) and everything worked great, GUI tools found, initialized, and mounted the drive like it was a standard SCSI drive. A+++++++, would try again!

As long as has either the LSISAS1064 or LSISAS1068 chipset it should work - I didn't see a specific mention of the SAS3041XL-S nomenclature on LSI's site. You might want to confirm it's the correct height profile, the "L" in the product name might mean it's equipped with a Low-profile bracket? <though it looks like the retail version comes with both the low and normal height brackets>.

Dudley 1# diskperf -W -D -c4g /disk3/test
#---------------------------------------------------------
# Disk Performance Test Results Generated By Diskperf V1.2
#
# Test name     : Unspecified
# Test date     : Wed Mar 25 17:53:43 2009
# Test machine  : IRIX64 Dudley 6.5 07202013 IP35
# Test type     : XFS data subvolume
# Test path     : /disk3/test
# Request sizes : min=16384 max=4194304
# Parameters    : direct=1 time=10 scale=1.000 delay=0.000
# XFS file size : 4294967296 bytes
#---------------------------------------------------------
# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   43.75   34.36    7.03    7.50    3.74    1.69
      32768   45.96   40.45   11.83    7.99    6.89    3.25
      65536   44.76   22.84   18.39    8.32   11.94    5.92
     131072   44.74   44.70   18.61   17.41   13.75    9.82
     262144   46.63   24.65   25.04   17.79   20.22   14.52
     524288   46.13   25.54   28.42   28.32   25.65   21.33
    1048576   46.49   34.45   37.03   36.56   32.68   28.76
    2097152   47.32   37.19   44.00   44.39   39.70   34.08
    4194304   46.32   41.39   44.36   43.40   43.63   33.52
Dudley 2#