Is anyone else having problems with the AIC7XXX driver using a AHA-29160
controller? I believe my kernel is going into a unrecoverable spin-lock
when there is high SCSI load. I'm assuming this because the keyboard
lights still function and network still replies to pings.
In addition the 'st' driver returns with unrecoverable errors when
trying to tar to tape. This usually occurs after a few hundred
megabytes have been pushed through a device.
My setup is :
- AHA-29160N controller
- Internal 50 PIN / SCSI-2 port in use
- See below for connected drives
I tested the following kernels and they display similar behavior:
- v2.4.9
- v2.4.18
- v2.4.18 with updated 7XXX driver
- v2.4.19-pre10
Any insight is appreciated.
- Nick
DETAILS:
Jun 19 01:25:15 bombay kernel: scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.8
Jun 19 01:25:15 bombay kernel: <Adaptec 29160 Ultra160 SCSI adapter>
Jun 19 01:25:15 bombay kernel: aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs
Jun 19 01:25:15 bombay kernel:
Jun 19 01:25:15 bombay kernel: Vendor: IBM Model: DNES-318350 Rev: SAH0
Jun 19 01:25:15 bombay kernel: Type: Direct-Access ANSI SCSI revision: 03
Jun 19 01:25:15 bombay kernel: Vendor: IBM Model: DCAS-34330 Rev: S65A
Jun 19 01:25:15 bombay kernel: Type: Direct-Access ANSI SCSI revision: 02
Jun 19 01:25:15 bombay kernel: Vendor: PLEXTOR Model: CD-R PX-W124TS Rev: 1.07
Jun 19 01:25:15 bombay kernel: Type: CD-ROM ANSI SCSI revision: 02
Jun 19 01:25:15 bombay kernel: scsi0:A:0:0: Tagged Queuing enabled. Depth 253
Jun 19 01:25:15 bombay kernel: scsi0:A:2:0: Tagged Queuing enabled. Depth 253
Jun 19 01:25:15 bombay kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
Jun 19 01:25:15 bombay kernel: Attached scsi disk sdb at scsi0, channel 0, id 2, lun 0
Jun 19 01:25:15 bombay kernel: (scsi0:A:0): 20.000MB/s transfers (20.000MHz, offset 31)
Jun 19 01:25:15 bombay kernel: SCSI device sda: 35843670 512-byte hdwr sectors (18352 MB)
Jun 19 01:25:15 bombay kernel: sda: sda1 sda2 sda3
Jun 19 01:25:15 bombay kernel: (scsi0:A:2): 20.000MB/s transfers (20.000MHz, offset 15)
Jun 19 01:25:15 bombay kernel: SCSI device sdb: 8467200 512-byte hdwr sectors (4335 MB)
Jun 19 01:25:15 bombay kernel: sdb: unknown partition table
[root@bombay root]# more /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.2.8
aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs
Serial EEPROM:
0x439b 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a
0xc33a 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a 0xc33a
0xb8f4 0x7c5d 0x2807 0x0010 0x0300 0xffff 0xffff 0xffff
0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x0250 0x16b0
Channel A Target 0 Negotiation Settings
User: 40.000MB/s transfers (40.000MHz DT, offset 255)
Goal: 20.000MB/s transfers (20.000MHz, offset 31)
Curr: 20.000MB/s transfers (20.000MHz, offset 31)
Channel A Target 0 Lun 0 Settings
Commands Queued 245659
Commands Active 0
Command Openings 64
Max Tagged Openings 64
Device Queue Frozen Count 0
Channel A Target 1 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 2 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 20.000MB/s transfers (20.000MHz, offset 15)
Curr: 20.000MB/s transfers (20.000MHz, offset 15)
Channel A Target 2 Lun 0 Settings
Commands Queued 5
Commands Active 0
Command Openings 253
Max Tagged Openings 253
Device Queue Frozen Count 0
Channel A Target 3 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 40.000MB/s transfers (40.000MHz, offset 127)
Curr: 3.300MB/s transfers
Channel A Target 3 Lun 0 Settings
Commands Queued 1
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Channel A Target 4 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 5 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 6 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 7 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 8 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 9 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 10 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 11 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 12 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 13 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 14 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Channel A Target 15 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
I wonder if problems are due to my VIA PCI chipset? Here are some details:
Bus 0, device 10, function 0:
SCSI storage controller: Adaptec AIC-7892A U160/m (rev 2).
IRQ 12.
Master Capable. Latency=32. Min Gnt=40.Max Lat=25.
I/O at 0xa800 [0xa8ff].
Non-prefetchable 64 bit memory at 0xdf000000 [0xdf000fff].
Bus 0, device 0, function 0:
Host bridge: VIA Technologies, Inc. VT82C693A/694x [Apollo PRO133x] (rev 196).
Prefetchable 32 bit memory at 0xe7000000 [0xe7ffffff].
Bus 0, device 1, function 0:
PCI bridge: VIA Technologies, Inc. VT82C598/694x [Apollo MVP3/Pro133x AGP] (rev 0).
Master Capable. No bursts. Min Gnt=8.
Nick Papadonis <[email protected]> writes:
> Is anyone else having problems with the AIC7XXX driver using a AHA-29160
> controller? I believe my kernel is going into a unrecoverable spin-lock
> when there is high SCSI load. I'm assuming this because the keyboard
> lights still function and network still replies to pings.
>
> In addition the 'st' driver returns with unrecoverable errors when
> trying to tar to tape. This usually occurs after a few hundred
> megabytes have been pushed through a device.
>
> My setup is :
> - AHA-29160N controller
> - Internal 50 PIN / SCSI-2 port in use
> - See below for connected drives
>
> I tested the following kernels and they display similar behavior:
> - v2.4.9
> - v2.4.18
> - v2.4.18 with updated 7XXX driver
> - v2.4.19-pre10
>
> Any insight is appreciated.
>
> - Nick
On Wed, Jun 19, 2002 at 11:09:13AM -0400, Nick Papadonis wrote:
> Is anyone else having problems with the AIC7XXX driver using a AHA-29160
> controller? I believe my kernel is going into a unrecoverable spin-lock
> when there is high SCSI load. I'm assuming this because the keyboard
> lights still function and network still replies to pings.
A spinlock deadlock causes the machine to be *totally* unresponsive (aka,
no keyboard led changes and no pings) (at least, this is true of the scsi
subsystem spinlocks).
> In addition the 'st' driver returns with unrecoverable errors when
> trying to tar to tape. This usually occurs after a few hundred
> megabytes have been pushed through a device.
Hmmm...the tape drive isn't listed in your output below.
> My setup is :
> - AHA-29160N controller
> - Internal 50 PIN / SCSI-2 port in use
> - See below for connected drives
It appears that you have the two hard drives connected to this 50 pin
port. That's not good. The drives should be connected to the LVD port
via a good LVD cable with LVD terminator, the CD-ROM should be on the 50
pin port, and the tape (which I'm assuming is external and not a new LVD
tape drive but an older narrow tape drive using a 50 pin connecter) should
either be on the same cable as the CD-ROM (via an internal/external cable
adapter) or on the 50pin port on the back of the card (but only if the
port on the back of the card is not an LVD capable port, but is instead a
SE port wired together with the internal 50 pin port). Hooking them up
the way you have them is asking for trouble because your high end disks
are being forced to operate in a very degraded performance state.
> Channel A Target 0 Negotiation Settings
> User: 40.000MB/s transfers (40.000MHz DT, offset 255)
> Goal: 20.000MB/s transfers (20.000MHz, offset 31)
> Curr: 20.000MB/s transfers (20.000MHz, offset 31)
See, disk 1 is being forced all the way fown to narrow, ultra speeds.
> Channel A Target 2 Negotiation Settings
> User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
> Goal: 20.000MB/s transfers (20.000MHz, offset 15)
> Curr: 20.000MB/s transfers (20.000MHz, offset 15)
Same here, narrow ultra speeds. Both of these disks should likely be
running in wide ultra2 speeds or better (this second one might not do
ultra2, I couldn't tell by the model number, but I suspect it does).
> Channel A Target 3 Negotiation Settings
> User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
> Goal: 40.000MB/s transfers (40.000MHz, offset 127)
> Curr: 3.300MB/s transfers
Async transfers (the cd-rom). This should *not* be on the same cable as
the disks.
--
Doug Ledford <[email protected]> 919-754-3700 x44233
Red Hat, Inc.
1801 Varsity Dr.
Raleigh, NC 27606
On 2002.06.19 Nick Papadonis wrote:
>Is anyone else having problems with the AIC7XXX driver using a AHA-29160
>controller? I believe my kernel is going into a unrecoverable spin-lock
>when there is high SCSI load. I'm assuming this because the keyboard
>lights still function and network still replies to pings.
>
>In addition the 'st' driver returns with unrecoverable errors when
>trying to tar to tape. This usually occurs after a few hundred
>megabytes have been pushed through a device.
>
>My setup is :
> - AHA-29160N controller
> - Internal 50 PIN / SCSI-2 port in use
> - See below for connected drives
>
>I tested the following kernels and they display similar behavior:
> - v2.4.9
> - v2.4.18
> - v2.4.18 with updated 7XXX driver
> - v2.4.19-pre10
>
First of all, I have a setup like yours, and with 2 U160 drives hanged on the
LVD connector for the bus, and one zip and one CD on the 50-pin, everything
works fine:
annwn:/proc/scsi> cat scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: FUJITSU Model: MAJ3364MP Rev: 0115
Type: Direct-Access ANSI SCSI revision: 04
Host: scsi0 Channel: 00 Id: 01 Lun: 00
Vendor: FUJITSU Model: MAJ3364MP Rev: 0115
Type: Direct-Access ANSI SCSI revision: 04
Host: scsi0 Channel: 00 Id: 02 Lun: 00
Vendor: TOSHIBA Model: CD-ROM XM-6201TA Rev: 1030
Type: CD-ROM ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 00
Vendor: IOMEGA Model: ZIP 100 Rev: E.08
Type: Direct-Access ANSI SCSI revision: 02
annwn:/proc/scsi/aic7xxx> cat 0
Adaptec AIC7xxx driver version: 6.2.8
aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs
Channel A Target 0 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 160.000MB/s transfers (80.000MHz DT, offset 127, 16bit)
Curr: 160.000MB/s transfers (80.000MHz DT, offset 127, 16bit)
...
Channel A Target 1 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 160.000MB/s transfers (80.000MHz DT, offset 127, 16bit)
Curr: 160.000MB/s transfers (80.000MHz DT, offset 127, 16bit)
...
Channel A Target 2 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 10.000MB/s transfers (10.000MHz, offset 16)
Curr: 10.000MB/s transfers (10.000MHz, offset 16)
...
Channel A Target 4 Negotiation Settings
User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
Goal: 3.300MB/s transfers
Curr: 3.300MB/s transfers
>
>DETAILS:
>Jun 19 01:25:15 bombay kernel: scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.8
>Jun 19 01:25:15 bombay kernel: <Adaptec 29160 Ultra160 SCSI adapter>
>Jun 19 01:25:15 bombay kernel: aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs
>Jun 19 01:25:15 bombay kernel:
>Jun 19 01:25:15 bombay kernel: Vendor: IBM Model: DNES-318350 Rev: SAH0
>Jun 19 01:25:15 bombay kernel: Type: Direct-Access ANSI SCSI revision: 03
>Jun 19 01:25:15 bombay kernel: Vendor: IBM Model: DCAS-34330 Rev: S65A
>Jun 19 01:25:15 bombay kernel: Type: Direct-Access ANSI SCSI revision: 02
>Jun 19 01:25:15 bombay kernel: Vendor: PLEXTOR Model: CD-R PX-W124TS Rev: 1.07
>Jun 19 01:25:15 bombay kernel: Type: CD-ROM ANSI SCSI revision: 02
You are hangin things on the 50pin connector, which is SE, so you could never
get 160Mb/s (LVD, transmit on up and down), you could at most go at 80Mb/s.
>
>Channel A Target 0 Negotiation Settings
> User: 40.000MB/s transfers (40.000MHz DT, offset 255)
> Goal: 20.000MB/s transfers (20.000MHz, offset 31)
> Curr: 20.000MB/s transfers (20.000MHz, offset 31)
Hardware thinks the disk just can do Goal=20Mb/s. Why ?
It says nothing about 16 bits... Are you forcing it to narrow
with a jumper ??
>Channel A Target 2 Negotiation Settings
> User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
> Goal: 20.000MB/s transfers (20.000MHz, offset 15)
> Curr: 20.000MB/s transfers (20.000MHz, offset 15)
It recognizes the drive can do wide, but as the first device in the
chain forced narrow, goal is narrow.
>Channel A Target 3 Negotiation Settings
> User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
> Goal: 40.000MB/s transfers (40.000MHz, offset 127)
> Curr: 3.300MB/s transfers
Same as above....
Move the fast drive to the LVD connector, and chech for jumper settings on
that drive.
--
J.A. Magallon \ Software is like sex: It's better when it's free
mailto:[email protected] \ -- Linus Torvalds, FSF T-shirt
Linux werewolf 2.4.19-pre10-jam3, Mandrake Linux 8.3 (Cooker) for i586
gcc (GCC) 3.1.1 (Mandrake Linux 8.3 3.1.1-0.4mdk)
On Wed, Jun 19, 2002 at 11:49:25PM +0200, J.A. Magallon wrote:
> You are hangin things on the 50pin connector, which is SE, so you could never
> get 160Mb/s (LVD, transmit on up and down), you could at most go at 80Mb/s.
80MB/s is also LVD. The fastest you can go on a SE narrow chain is in
fact 20MB/s.
> >
> >Channel A Target 0 Negotiation Settings
> > User: 40.000MB/s transfers (40.000MHz DT, offset 255)
> > Goal: 20.000MB/s transfers (20.000MHz, offset 31)
> > Curr: 20.000MB/s transfers (20.000MHz, offset 31)
>
> Hardware thinks the disk just can do Goal=20Mb/s. Why ?
The 29160 card detects which cable a device is hooked to and the driver
then sets the maximum allowed speed according to that. When you plug a
drive into a SE cable, the driver knows it and sets the maximum speed to
SE speeds.
> It says nothing about 16 bits... Are you forcing it to narrow
> with a jumper ??
16bit is off in the BIOS settings, so it wouldn't get used regardless.
> >Channel A Target 2 Negotiation Settings
> > User: 160.000MB/s transfers (80.000MHz DT, offset 255, 16bit)
> > Goal: 20.000MB/s transfers (20.000MHz, offset 15)
> > Curr: 20.000MB/s transfers (20.000MHz, offset 15)
>
> It recognizes the drive can do wide, but as the first device in the
> chain forced narrow, goal is narrow.
No, the order of devices has no impact on this. No device has any impact
on any other device's speed settings unless they are both connected to the
same physical cable and one of the devices is forcing a cable that *could*
be in LVD mode into SE mode instead. Short of that, all device
negotiations are 100% independant of each other.
--
Doug Ledford <[email protected]> 919-754-3700 x44233
Red Hat, Inc.
1801 Varsity Dr.
Raleigh, NC 27606
On 2002.06.20 Doug Ledford wrote:
>
>No, the order of devices has no impact on this. No device has any impact
>on any other device's speed settings unless they are both connected to the
>same physical cable and one of the devices is forcing a cable that *could*
>be in LVD mode into SE mode instead. Short of that, all device
>negotiations are 100% independant of each other.
>
As I understood, that is what he said, he had everything connected to the
50pin...
--
J.A. Magallon \ Software is like sex: It's better when it's free
mailto:[email protected] \ -- Linus Torvalds, FSF T-shirt
Linux werewolf 2.4.19-pre10-jam3a, Mandrake Linux 8.3 (Cooker) for i586
gcc (GCC) 3.1.1 (Mandrake Linux 8.3 3.1.1-0.4mdk)