Some interesting results in the aic7xxx slowdown problem

DISCUSSION Suparna Bhattacharya 2 participants 6 messages Aug 19, 2005 – Aug 30, 2005 3 messages not in the archive SCSI

lore

Share & tools

lore

A message that is not in the archive (sent off-list or not yet received): 1123119347.5019.78.camel@mulgrave

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 1123166928.5026.8.camel@mulgrave

Suparna Bhattacharya

author

2005-08-19 09:45 UTC #3 lore raw

James,

I finally got around to trying out 2.6.13-rc6 which (I think) includes the AIC driver modifications to use block layer tagging. But it looks like I'm out of luck -- the system doesn't even come up now. This is what I see.

...
Initializing IPsec netlink socket
NET: Registered protocol family 1
NET: Registered protocol family 17
NET: Registered protocol family 15
Starting balanced_irq
Using IPI Shortcut mode
VFS: Cannot open root device "sda6" or unknown-block(8,6) Please append a correct "root=" boot option

Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(8,6)

Is this a known problem ? I do not get it if I configure the OLD aic driver instead.

Some snippets from the scsi bootup messages :

...
scsi0:A:11:0: Tagged Queuing enabled. Depth 32
target0:0:11: Beginning Domain Validation target0:0:11: wide asynchronous.
target0:0:11: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:11: Ending Domain Validation
target0:0:12: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access

...

ACPI: PCI Interrupt 0000:00:01.1[A] -> GSI 17 (level, low) -> IRQ 17 scsi1 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7896/97 Ultra2 SCSI adapter> aic7896/97: Ultra2 Wide Channel B, SCSI Id=7, 32/253 SCBs

target1:0:0: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03

....

SCSI device sda: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sda: drive cache: write through
SCSI device sda: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sda: drive cache: write through
sda:
Attached scsi disk sda at scsi0, channel 0, id 9, lun 0 SCSI device sdb: 35548320 512-byte hdwr sectors (18201 MB)
....

Attached scsi generic sg0 at scsi0, channel 0, id 9, lun 0, type 0 Attached scsi generic sg1 at scsi0, channel 0, id 10, lun 0, type 0

...

Regards
Suparna

Signature · 3 lines

Suparna Bhattacharya ([email protected])
Linux Technology Center
IBM Software Lab, India

James Bottomley

maintainer

2005-08-19 13:22 UTC #4 lore raw

Quoted 1 line · Suparna Bhattacharya

sda:

Well, this line is identifying the root cause. It says that the partition code came up with no partitions for this device. Do any of your other devices come up with paritions? (as in could this be a simple device node transposition issue?)

James

Suparna Bhattacharya

author

2005-08-25 11:42 UTC #5 lore raw

Quoted 7 lines · James Bottomley

Quoted 1 line · Suparna Bhattacharya
sda:
Well, this line is identifying the root cause. It says that the partition code came up with no partitions for this device. Do any of your other devices come up with paritions? (as in could this be a simple device node transposition issue?)

Yes, it looks like that. The ordering now seems to be id 9, 10, ... 14, 0, 1, 2, instead of the default of id 0, 1, 2 ... Did you expect a change in probe ordering ?

When I switch back to the old AIC driver the ordering goes back to default.

Regards
Suparna

Quoted 4 lines

James

Signature · 3 lines

Suparna Bhattacharya ([email protected])
Linux Technology Center
IBM Software Lab, India

James Bottomley

maintainer

2005-08-29 15:21 UTC #6 lore raw

Quoted 9 lines · Suparna Bhattacharya

Quoted 7 lines · James Bottomley
Quoted 1 line · Suparna Bhattacharya
sda:
Well, this line is identifying the root cause. It says that the partition code came up with no partitions for this device. Do any of your other devices come up with paritions? (as in could this be a simple device node transposition issue?)

Quoted 6 lines

Yes, it looks like that. The ordering now seems to be id 9, 10, ... 14, 0, 1, 2, instead of the default of id 0, 1, 2 ... Did you expect a change in probe ordering ?
When I switch back to the old AIC driver the ordering goes back to default.

Er, boggle!

I know of no change to the new driver that could affect the target scanning order (this is target, not lun, right?). Target scan should be handled sequentially in the mid-layer (scsi_scan.c) in a for loop that runs from 0 to 15 (well shost->max_id). There is a reverse_order flag, could that be set somehow?

James

Suparna Bhattacharya

author

2005-08-30 04:41 UTC #7 lore raw

Quoted 21 lines · James Bottomley

ailer: Evolution 2.0.4 (2.0.4-6)
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: amavisd-new at linux.ibm.com
Status: RO
Content-Length: 1065
Lines: 28
Quoted 13 lines · Suparna Bhattacharya
Quoted 7 lines · James Bottomley
Quoted 1 line · Suparna Bhattacharya
sda:
Well, this line is identifying the root cause. It says that the partition code came up with no partitions for this device. Do any of your other devices come up with paritions? (as in could this be a simple device node transposition issue?)
Yes, it looks like that. The ordering now seems to be id 9, 10, ... 14, 0, 1, 2, instead of the default of id 0, 1, 2 ... Did you expect a change in probe ordering ?

Quoted 8 lines

Quoted 2 lines
When I switch back to the old AIC driver the ordering goes back to default.
Er, boggle!
I know of no change to the new driver that could affect the target scanning order (this is target, not lun, right?). Target scan should be handled sequentially in

Here is what the log looks like:

ACPI: PCI Interrupt 0000:00:01.0[A] -> GSI 17 (level, low) -> IRQ 17 scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7896/97 Ultra2 SCSI adapter> aic7896/97: Ultra2 Wide Channel A, SCSI Id=7, 32/253 SCBs

target0:0:9: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:9:0: Tagged Queuing enabled. Depth 32
target0:0:9: Beginning Domain Validation
target0:0:9: wide asynchronous.
target0:0:9: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:9: Ending Domain Validation
target0:0:10: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:10:0: Tagged Queuing enabled. Depth 32
target0:0:10: Beginning Domain Validation
target0:0:10: wide asynchronous.
target0:0:10: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:10: Ending Domain Validation
target0:0:11: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:11:0: Tagged Queuing enabled. Depth 32
target0:0:11: Beginning Domain Validation
target0:0:11: wide asynchronous.
target0:0:11: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:11: Ending Domain Validation
target0:0:12: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:12:0: Tagged Queuing enabled. Depth 32
target0:0:12: Beginning Domain Validation
target0:0:12: wide asynchronous.
target0:0:12: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:12: Ending Domain Validation
target0:0:13: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:13:0: Tagged Queuing enabled. Depth 32
target0:0:13: Beginning Domain Validation
target0:0:13: wide asynchronous.
target0:0:13: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:13: Ending Domain Validation
target0:0:14: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi0:A:14:0: Tagged Queuing enabled. Depth 32
target0:0:14: Beginning Domain Validation
target0:0:14: wide asynchronous.
target0:0:14: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target0:0:14: Ending Domain Validation
Vendor: IBM Model: EXP300 S160 Rev: D014
Type: Processor ANSI SCSI revision: 03
target0:0:15: asynchronous.
target0:0:15: Beginning Domain Validation
target0:0:15: Domain Validation skipping write tests
(scsi0:A:15:0): refuses synchronous negotiation. Using asynchronous transfers
target0:0:15: Ending Domain Validation
ACPI: PCI Interrupt 0000:00:01.1[A] -> GSI 17 (level, low) -> IRQ 17 scsi1 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7896/97 Ultra2 SCSI adapter>
aic7896/97: Ultra2 Wide Channel B, SCSI Id=7, 32/253 SCBs

target1:0:0: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi1:A:0:0: Tagged Queuing enabled. Depth 32
target1:0:0: Beginning Domain Validation
target1:0:0: wide asynchronous.
target1:0:0: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target1:0:0: Ending Domain Validation
target1:0:1: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi1:A:1:0: Tagged Queuing enabled. Depth 32
target1:0:1: Beginning Domain Validation
target1:0:1: wide asynchronous.
target1:0:1: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target1:0:1: Ending Domain Validation
target1:0:2: asynchronous.
Vendor: IBM-ESXS Model: ST318305LC !# Rev: B245
Type: Direct-Access ANSI SCSI revision: 03
scsi1:A:2:0: Tagged Queuing enabled. Depth 32
target1:0:2: Beginning Domain Validation
target1:0:2: wide asynchronous.
target1:0:2: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 63) target1:0:2: Ending Domain Validation
Vendor: IBM-ESXS Model: DTN018C1UCDY10F Rev: S25J
Type: Direct-Access ANSI SCSI revision: 03
target1:0:3: asynchronous.
scsi1:A:3:0: Tagged Queuing enabled. Depth 32
target1:0:3: Beginning Domain Validation
target1:0:3: wide asynchronous.
target1:0:3: FAST-40 WIDE SCSI 80.0 MB/s ST (25 ns, offset 127) target1:0:3: Ending Domain Validation
Vendor: IBM Model: AuSaV1S2 Rev: 0
Type: Processor ANSI SCSI revision: 02
target1:0:8: asynchronous.
target1:0:8: Beginning Domain Validation
target1:0:8: Ending Domain Validation
SCSI device sda: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sda: drive cache: write through
SCSI device sda: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sda: drive cache: write through
sda:
Attached scsi disk sda at scsi0, channel 0, id 9, lun 0 SCSI device sdb: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdb: drive cache: write through
SCSI device sdb: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdb: drive cache: write through
sdb: unknown partition table
Attached scsi disk sdb at scsi0, channel 0, id 10, lun 0 SCSI device sdc: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdc: drive cache: write through
SCSI device sdc: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdc: drive cache: write through
sdc: sdc1
Attached scsi disk sdc at scsi0, channel 0, id 11, lun 0 SCSI device sdd: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdd: drive cache: write through
SCSI device sdd: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdd: drive cache: write through
sdd: sdd1
Attached scsi disk sdd at scsi0, channel 0, id 12, lun 0 SCSI device sde: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sde: drive cache: write through
SCSI device sde: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sde: drive cache: write through
sde: sde1 sde2 sde3 sde4
Attached scsi disk sde at scsi0, channel 0, id 13, lun 0 SCSI device sdf: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdf: drive cache: write through
SCSI device sdf: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdf: drive cache: write through
sdf: sdf1
Attached scsi disk sdf at scsi0, channel 0, id 14, lun 0 SCSI device sdg: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdg: drive cache: write through
SCSI device sdg: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdg: drive cache: write through
sdg: sdg1 sdg2 sdg3 sdg4 < sdg5 sdg6 >
Attached scsi disk sdg at scsi1, channel 0, id 0, lun 0 SCSI device sdh: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdh: drive cache: write through
SCSI device sdh: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdh: drive cache: write through
sdh: sdh1 sdh2 sdh3 sdh4 < sdh5 >
Attached scsi disk sdh at scsi1, channel 0, id 1, lun 0 SCSI device sdi: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdi: drive cache: write through
SCSI device sdi: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdi: drive cache: write through
sdi: sdi1
Attached scsi disk sdi at scsi1, channel 0, id 2, lun 0 SCSI device sdj: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdj: drive cache: write through
SCSI device sdj: 35548320 512-byte hdwr sectors (18201 MB) SCSI device sdj: drive cache: write through
sdj: sdj1
Attached scsi disk sdj at scsi1, channel 0, id 3, lun 0 Attached scsi generic sg0 at scsi0, channel 0, id 9, lun 0, type 0

Attached scsi generic sg1 at scsi0, channel 0, id 10, lun 0,  type 0
Attached scsi generic sg2 at scsi0, channel 0, id 11, lun 0,  type 0
Attached scsi generic sg3 at scsi0, channel 0, id 12, lun 0,  type 0
Attached scsi generic sg4 at scsi0, channel 0, id 13, lun 0,  type 0
Attached scsi generic sg5 at scsi0, channel 0, id 14, lun 0,  type 0
Attached scsi generic sg6 at scsi0, channel 0, id 15, lun 0,  type 3
Attached scsi generic sg7 at scsi1, channel 0, id 0, lun 0,  type 0
Attached scsi generic sg8 at scsi1, channel 0, id 1, lun 0,  type 0
Attached scsi generic sg9 at scsi1, channel 0, id 2, lun 0,  type 0

Attached scsi generic sg10 at scsi1, channel 0, id 3, lun 0, type 0 Attached scsi generic sg11 at scsi1, channel 0, id 8, lun 0, type 3
ieee1394: raw1394: /dev/raw1394 device initialized

Regards
Suparna

Signature · 3 lines

Suparna Bhattacharya ([email protected])
Linux Technology Center
IBM Software Lab, India

James Bottomley

maintainer

2005-08-30 13:52 UTC #8 lore raw

Quoted 3 lines · Suparna Bhattacharya

scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7896/97 Ultra2 SCSI adapter> aic7896/97: Ultra2 Wide Channel A, SCSI Id=7, 32/253 SCBs

[...]

Quoted 3 lines

scsi1 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7896/97 Ultra2 SCSI adapter> aic7896/97: Ultra2 Wide Channel B, SCSI Id=7, 32/253 SCBs

Aha, what I think has happened is that scsi0 and scsi1 have reversed their detection order. This is a consequence of moving aic7xxx to the new PCI model. We no-longer control the detection order, we get probed in the order that the PCI subsystem decides. In your case, it must have decided to detect the external card first.

The PCI subsystem pulls a few tricks to try to prevent this, but most of them rely on ACPI I think ... I assume this isn't an ACPI system?

James