Problem with Serverworks HT1000 serial ATA support - Tyan S3992 Motherboard

[Nicole Harrington]

--- Mike Tancsa <mike at sentex.net> wrote:
> On Tue, 6 Feb 2007 15:26:27 -0800 (PST), in
> sentex.lists.freebsd.questions you wrote:

> > If I run: bonnie++ -d /home -u root s1g
> >  The system reboots.
> >
> > I have tried this is both S-ATA and P-ATA
> emulation.
> >atapci0: <ServerWorks HT1000 SATA150 controller>
> port
> >0xb080-0xb087,0xb000-0xb003,0xac00-0xac07
> >,0xa880-0xa883,0xa800-0xa81f mem
> 0xff4fe000-0xff4fffff
> >irq 11 at device 14.0 on pci1
> >
> >S-ATA:
> >ad4: 70911MB <WDC WD740ADFD-00NLR1 20.07P20> at
> >ata2-master SATA150
> >
> > OR
> >P-ATA:
> >ad4: DMA limited to UDMA33, device found non-ATA66
> >cable
> >ad4: 70911MB <WDC WD740ADFD-00NLR1 20.07P20> at
> >ata2-master UDMA33

 
> I use RELENG_6 in i386 mode on this MB (two
> integrated em nics) and it
> works quite well

 Have you tried it with the tune for Diskd mentioned.
 
> atapci0: <ServerWorks HT1000 SATA150 controller>
> port
>
0xc080-0xc087,0xc000-0xc003,0xbc00-0xbc07,0xb880-0xb883,0xb800-0xb81f
> mem 0xff
> 3fe000-0xff3fffff irq 11 at device 14.0 on pci1
> ata2: <ATA channel 0> on atapci0
> ata3: <ATA channel 1> on atapci0
> ata4: <ATA channel 2> on atapci0
> ata5: <ATA channel 3> on atapci0
> atapci1: <ServerWorks HT1000 UDMA100 controller>
> port
> 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xffa0-0xffaf at
> device 2.1 on
> pci0
> ata0: <ATA channel 0> on atapci1
> ata1: <ATA channel 1> on atapci1
> 
> ad4: 76319MB <Seagate ST380811AS 3.AAE> at
> ata2-master SATA150
> 
> # atacontrol list
> ATA channel 0:
>     Master:      no device present
>     Slave:       no device present
> ATA channel 1:
>     Master:      no device present
>     Slave:       no device present
> ATA channel 2:
>     Master:  ad4 <ST380811AS/3.AAE> Serial ATA II
>     Slave:       no device present
> ATA channel 3:
>     Master:      no device present
>     Slave:       no device present
> ATA channel 4:
>     Master:      no device present
>     Slave:       no device present
> ATA channel 5:
>     Master:      no device present
>     Slave:       no device present
> 
> 
> However, update the BIOS and make sure you dont use
> PATA emulation on
> the SATA controller as that doesnt really work. 
> 
> 	---Mike
>
--------------------------------------------------------
> Mike Tancsa, Sentex communications
> http://www.sentex.net
> Providing Internet Access since 1994
> mike at sentex.net, (http://www.tancsa.com)

Hi Mike
 Yes, it does work quite well. But not if I use the
same tuning parameters used on systems running on Non
Serverworks chipsets. (Tyan S2881 or S2882 Board)

 I use these for supporting many heavy loaded Squid
servers. 

 The normal tweaks suggested for Squid, and work well
on Non Serverworks chipsets are:

/boot/loader.conf
kern.ipc.maxsockets=16424
kern.ipc.msgmnb=16384
kern.ipc.msgmni=41
kern.ipc.msgseg=2049
kern.ipc.msgssz=64
kern.ipc.msgtql=2048
 OR
kern.ipc.msgmnb=8192
kern.ipc.msgmni=40
kern.ipc.msgseg=512
kern.ipc.msgssz=64
kern.ipc.msgtql=1024

 However, with any setting of kern.ipc.msgtql (What
does it do anyway?) above 64 the server crashes with
the Bonnie test.

 The higher the setting the faster the crash.

 

 Nicole