Re: mpt0: QUEUE FULL EVENT: Bus 0x00 Target 0x04 Depth 129

On Thu, 2007-10-11 at 13:30 +0200, Marian Hettwer wrote:
> Hi All,
> 
> I'm trying out -CURRENT on an IBM HS21 blade. So far the machine looks stable.
> Only thing that bothers me is the message as stated above.
> 
> db46-23-freebsd# uptime
> 11:25AM  up 55 mins, 5 users, load averages: 0.01, 0.90, 1.21
> db46-23-freebsd# dmesg | grep -c "QUEUE FULL EVENT"
> 111
> db46-23-freebsd# uname -a
> FreeBSD db46-23-freebsd 7.0-CURRENT FreeBSD 7.0-CURRENT #0: Thu Oct 11 11:09:46 CEST 2007     root_at_blowfish:/usr/obj/amd64/usr/src/sys/GENERIC  amd64
> db46-23-freebsd#
> 
> Is this something to worry about?
> 
> The LSI Logic controller in question is configured as a hardware RAID-0
> Informations about the controller:
> at boot time from dmesg:
> 
> mpt0: <LSILogic SAS/SATA Adapter> port 0x4000-0x40ff mem 0xdeffc000-0xdeffffff,0xdefe0000-0xdefeffff irq 16 at device 0.0 on pci2
> mpt0: [ITHREAD]
> mpt0: MPI Version=1.5.13.0
> mpt0: mpt_cam_event: 0x16
> mpt0: Unhandled Event Notify Frame. Event 0x16 (ACK not required).
> mpt0: mpt_cam_event: 0x12
> mpt0: Unhandled Event Notify Frame. Event 0x12 (ACK not required).
> mpt0: mpt_cam_event: 0x12
> mpt0: Unhandled Event Notify Frame. Event 0x12 (ACK not required).
> mpt0: mpt_cam_event: 0x16
> mpt0: Unhandled Event Notify Frame. Event 0x16 (ACK not required).
> mpt0: mpt_cam_event: 0x21
> mpt0: Unhandled Event Notify Frame. Event 0x21 (ACK not required).
> mpt0: mpt_cam_event: 0x21
> mpt0: Unhandled Event Notify Frame. Event 0x21 (ACK not required).
> da0 at mpt0 bus 0 target 0 lun 0
> 
> and from pciconf -lv
> 
> mpt0_at_pci0:2:0:0:        class=0x010000 card=0x03061014 chip=0x00561000 rev=0x02 hdr=0x00
>     vendor     = 'LSI Logic (Was: Symbios Logic, NCR)'
>     device     = 'SAS 3000 series, 4-port with 1064E -StorPort'
>     class      = mass storage
>     subclass   = SCSI
> 
> regards,
> Marian
> PS.: FWIW, Linux 2.6.20 had severe problems with this specific mpt controller - patches came with Linux 2.6.21
> Looks like the LSI Logic stuff isn't very well supported. On the other hand, there's now way around them when using IBM blades :(
> Same counts for the bce(4) NIC.
> 

IIRC both mpt(4) and bce(4) have vendor supported drivers, so I'd be
surprised if they didn't work well. Here at work, we had some issues
with a bce(4) device in about 20 of our Dell boxes running RELENG_6_2
and David Christensen (Broadcom employee) helped fix them and make them
work lovely (allowing us to cancel an order of ~20 PCIE em cards :)

Tom