Date: Tue, 23 Feb 1999 14:50:26 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: "Justin T. Gibbs" <gibbs@narnia.plutotech.com> Cc: scsi@FreeBSD.org Subject: Re: dual-channel U2W scsi performance Message-ID: <14034.65025.537279.428755@grasshopper.cs.duke.edu> In-Reply-To: <199902230122.SAA11068@narnia.plutotech.com> References: <14033.59216.903659.321869@grasshopper.cs.duke.edu> <199902230122.SAA11068@narnia.plutotech.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Justin T. Gibbs writes:
> In article <14033.59216.903659.321869@grasshopper.cs.duke.edu> you wrote:
> >
> > Essentially, I'm wondering if I need to purchase another 2940UW for
> > each server, or if the above on-board dual-channel controller will
> > really be able to run 2 Ultra-Wide busses close to flat out.
>
> The only way to know for sure is to test it, and I have not performed
> such a test here. Based on the efficiency of past Adaptec PCI parts,
> I don't think it will have a problem saturating both channels assuming
> the rest of the system is up to the task and you have sufficient spare
> PCI bus bandwidth (no other big hogs).
>
> --
> Justin
I just tried, and I initially coouldn't get the second scsi channel
working at all. The end result is that the first channel appears to
probe the bus fine, if I move a chain which works just fine on the
first channel to the second channel, I see parity errors whenever a
disk is accessed when the bus is probed:
<..>
(probe0:ahc1:0:0:2): parity error during Message-In phase.
SEQADDR == 0x15d
SCSIRATE == 0x95
(probe0:ahc1:0:0:2): parity error during Data-In phase.
byte 3278
<.....>
da0: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da0: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da0: 4339MB (8887184 512 byte sectors: 255H 63S/T 553C)
(da1:ahc1:0:1:0): parity error during Command phase.
SEQADDR == 0x15d
SCSIRATE == 0x95
da1 at ahc1 bus 0 target 1 lun 0
<...>
The parity errors seem to disappear for reads if I disable parity
from SCSI select. However, I see parity errors on both channels
when I attempt to label the disks. This happens on both channels,
with parity set to disabled in scsi select for both channels:
login: (da0:ahc0:0:0:0): WRITE(06). CDB: a 0 0 0 10 0
(da0:ahc0:0:0:0): ABORTED COMMAND asc:47,0
(da0:ahc0:0:0:0): SCSI parity error field replaceable unit: 3
The performance is pretty good for just doing 8 dd if=/dev/rdaXc
of=/dev/null bs=64k from 8 cheetahs (about 9MB/sec per drive), or
about 72MB/sec overall. However, the parity error problems concern
me.
The kernel is built from 4.0-current sources cvsupped late last
week. The machine is a single 450Mhz Xeon sitting in a Tyan S1952DLU
Thunder X motherboard. Here's a full dmesg:
Copyright (c) 1992-1999 FreeBSD Inc.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
FreeBSD 4.0-CURRENT #4: Sat Feb 20 11:51:17 EST 1999
gallatin@grasshopper.cs.duke.edu:/freebsd/src/sys/compile/TPZ
Timecounter "i8254" frequency 1193182 Hz
Timecounter "TSC" frequency 451025157 Hz
CPU: Pentium II/Xeon/Celeron (451.03-MHz 686-class CPU)
Origin = "GenuineIntel" Id = 0x653 Stepping=3
Features=0x183fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,<b24>>
real memory = 134217728 (131072K bytes)
avail memory = 127447040 (124460K bytes)
Preloaded elf kernel "kernel" at 0xf02fb000.
Probing for devices on PCI bus 0:
chip0: <Host to PCI bridge (vendor=8086 device=71a0)> rev 0x00 on pci0.0.0
chip1: <PCI to PCI bridge (vendor=8086 device=71a1)> rev 0x00 on pci0.1.0
chip2: <Intel 82371AB PCI to ISA bridge> rev 0x02 on pci0.7.0
ide_pci0: <Intel PIIX4 Bus-master IDE controller> rev 0x01 on pci0.7.1
chip3: <Intel 82371AB Power management controller> rev 0x02 on pci0.7.3
ahc0: <Adaptec aic7896/97 Ultra2 SCSI adapter> rev 0x00 int a irq 10 on pci0.14.0
ahc0: aic7896/97 Wide Channel A, SCSI Id=7, 16/255 SCBs
ahc1: <Adaptec aic7896/97 Ultra2 SCSI adapter> rev 0x00 int a irq 10 on pci0.14.1
ahc1: aic7896/97 Wide Channel B, SCSI Id=7, 16/255 SCBs
mlanai0: <Myrinet PCI interface> rev 0x01 int a irq 11 on pci0.15.0
tpz0: Myrinet LANai 4.1 address 00:60:dd:e0:00:f1 (M2F-PCI32-10241)
tpz0: Duke TPZ/IP driver version $Id: if_tpz.c,v 1.12 1999/02/05 20:16:56 gallatin Exp $
fxp0: <Intel EtherExpress Pro 10/100B Ethernet> rev 0x05 int a irq 10 on pci0.16.0
fxp0: Ethernet address 00:a0:c9:e7:8e:9c
Probing for devices on PCI bus 1:
Probing for PnP devices:
Probing for devices on the ISA bus:
sc0 not found
atkbdc0 at 0x60-0x6f on motherboard
atkbd0 irq 1 on isa
psm0 not found
sio0 at 0x3f8-0x3ff irq 4 flags 0x10 on isa
sio0: type 16550A, console
sio1 at 0x2f8-0x2ff irq 3 on isa
sio1: type 16550A
ppc0 at 0x378 irq 7 on isa
ppc0: Generic chipset (NIBBLE-only) in COMPATIBLE mode
plip0: <PLIP network interface> on ppbus 0
ppi0: <generic parallel i/o> on ppbus 0
psm0 not found at 0x60
pcm0 not found
fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa
fdc0: FIFO enabled, 8 bytes threshold
fd0: 1.44MB 3.5in
wdc0 at 0x1f0-0x1f7 irq 14 flags 0xa0ff0ff on isa
wdc0: unit 0 (wd0): <QUANTUM FIREBALL EX5.1A>, LBA, DMA, 32-bit, multi-block-16, sleep-hack
wd0: 4892MB (10018890 sectors), 623 cyls, 255 heads, 63 S/T, 512 B/S
vga0 not found
npx0 flags 0x1 on motherboard
npx0: INT 16 interface
RTC BIOS diagnostic error 20<config_unit>
lo0 XXX: driver didn't set ifq_maxlen
cda0 at ahc0 bus 0 target 0 lun 0
da0: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da0: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da0: 4211MB (8625052 512 byte sectors: 255H 63S/T 536C)
da4 at ahc1 bus 0 target 0 lun 0
da4: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da4: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da4: 4339MB (8887184 512 byte sectors: 255H 63S/T 553C)
da7 at ahc1 bus 0 target 4 lun 0
da7: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da7: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da7: 4339MB (8887184 512 byte sectors: 255H 63S/T 553C)
da6 at ahc1 bus 0 target 3 lun 0
da6: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da6: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da6: 4339MB (8887184 512 byte sectors: 255H 63S/T 553C)
da5 at ahc1 bus 0 target 1 lun 0
da5: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da5: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da5: 4339MB (8887184 512 byte sectors: 255H 63S/T 553C)
da3 at ahc0 bus 0 target 3 lun 0
da3: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da3: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da3: 4211MB (8625052 512 byte sectors: 255H 63S/T 536C)
da2 at ahc0 bus 0 target 2 lun 0
da2: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da2: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da2: 4211MB (8625052 512 byte sectors: 255H 63S/T 536C)
da1 at ahc0 bus 0 target 1 lun 0
da1: <SEAGATE ST34501W 0018> Fixed Direct Access SCSI-2 device
da1: 40.0MB/s transfers (20.0MHz, offset 15, 16bit), Tagged Queueing Enabled
da1: 4211MB (8625052 512 byte sectors: 255H 63S/T 536C)
hanging root device to wd0s1a
ffs_mountfs: superblock updated for soft updates
ffs_mountfs: superblock updated for soft updates
(da0:ahc0:0:0:0): WRITE(06). CDB: a 0 0 0 10 0
(da0:ahc0:0:0:0): ABORTED COMMAND asc:47,0
(da0:ahc0:0:0:0): SCSI parity error field replaceable unit: 3
(da1:ahc0:0:1:0): WRITE(06). CDB: a 0 0 0 10 0
(da1:ahc0:0:1:0): ABORTED COMMAND asc:47,0
(da1:ahc0:0:1:0): SCSI parity error field replaceable unit: 3
(da6:ahc1:0:3:0): WRITE(06). CDB: a 0 0 0 10 0
(da6:ahc1:0:3:0): ABORTED COMMAND asc:47,0
(da6:ahc1:0:3:0): SCSI parity error field replaceable unit: 3
------------------------------------------------------------------------------
Andrew Gallatin, Sr Systems Programmer http://www.cs.duke.edu/~gallatin
Duke University Email: gallatin@cs.duke.edu
Department of Computer Science Phone: (919) 660-6590
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14034.65025.537279.428755>