sym53c8xx_2 make scsitap slow

sym53c8xx_2 make scsitap slow

[frage schnell]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 386 bytes --]

I used a HP ulrium storage on a tekteam DC-390U2B with kernel 2.6.5 and
modul sym53c8xx 
  tar writs with 11MB/s to the tape
with new kernel 2.6.8 or 2.6.10
   tar writs with 3MB/s to the tape

wath have I to to to get the same speed with the new kennel 

-- 
+++ Sparen Sie mit GMX DSL +++ http://www.gmx.net/de/go/dsl
AKTION für Wechsler: DSL-Tarife ab 3,99 EUR/Monat + Startguthaben

Re: sym53c8xx_2 make scsitap slow

[Matthias Andree]

"frage schnell" <frage.schnell@gmx.de> writes:

> I used a HP ulrium storage on a tekteam DC-390U2B with kernel 2.6.5 and
> modul sym53c8xx 
>   tar writs with 11MB/s to the tape
> with new kernel 2.6.8 or 2.6.10
>    tar writs with 3MB/s to the tape
>
> wath have I to to to get the same speed with the new kennel 

Please provide dmesg logs to shed some light on the detection of the
adaptor and tape drive.

-- 
Matthias Andree

Re: sym53c8xx_2 make scsitap slow

[frage schnell]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 6302 bytes --]

"frage schnell" <frage.schnell@gmx.de> writes:  
  
>> I used a HP ulrium storage on a tekteam DC-390U2B with kernel 2.6.5 and  
>> modul sym53c8xx   
>>   tar writs with 11MB/s to the tape  
>> with new kernel 2.6.8 or 2.6.10  
>>    tar writs with 3MB/s to the tape  
>>  
>> wath have I to to to get the same speed with the new kennel   
  
>Please provide dmesg logs to shed some light on the detection of the  
>adaptor and tape drive.  
  
dmesg for kernel 2.6.5 scsi related parts   
  
Linux version 2.6.5-7.111-smp (geeko@buildhost) (gcc version 3.3.3 (SuSE  
Linux))SCSI subsystem initialized  
Red Hat/Adaptec aacraid driver (1.1.2-lk1 Oct 14 2004)  
AAC0: kernel 4.0.4 build 5770  
AAC0: monitor 4.0.4 build 5770  
AAC0: bios 4.0.0 build 5770  
AAC0: serial ba2b1bfafaf001  
aacraid:        NMI ISR: NMI_PRIMARY_ATU_ERROR  
scsi0 : aacraid  
  Vendor: ADAPTEC   Model: Adaptec RAID5     Rev: V1.0  
  Type:   Direct-Access                      ANSI SCSI revision: 02  
SCSI device sda: 287105024 512-byte hdwr sectors (146998 MB)  
sda: Write Protect is off  
sda: Mode Sense: 03 00 00 00  
SCSI device sda: drive cache: write through  
 sda: sda1 sda2  
Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0  
  Vendor: ADAPTEC   Model: Adaptec Volume    Rev: V1.0  
  Type:   Direct-Access                      ANSI SCSI revision: 02  
SCSI device sdb: 143552512 512-byte hdwr sectors (73499 MB)  
sdb: Write Protect is off  
sdb: Mode Sense: 03 00 00 00  
SCSI device sdb: drive cache: write through  
 sdb: sdb1 sdb2  
Attached scsi removable disk sdb at scsi0, channel 0, id 1, lun 0  
Capability LSM initialized  
ReiserFS: sda1: found reiserfs format "3.6" with standard journal  
ReiserFS: sda1: using ordered data mode  
reiserfs: using flush barriers  
ReiserFS: sda1: journal params: device sda1, size 8192, journal first  
block 18,  
max trans len 1024, max batch 900, max commit age 30, max trans age 30  
ReiserFS: sda1: checking transaction log (sda1)  
reiserfs: disabling flush barriers on sda1  
ReiserFS: sda1: Using r5 hash to sort names  
VFS: Mounted root (reiserfs filesystem) readonly.  
Trying to move old root to /initrd ... failed  
Unmounting old root  
Trying to free ramdisk memory ... failed  
Freeing unused kernel memory: 244k freed  
Adding 1052216k swap on /dev/sdb1.  Priority:42 extents:1  
sk98lin: Network Device Driver v6.23  
(C)Copyright 1999-2004 Marvell(R).  
eth0: 3Com Gigabit LOM (3C940)  
      PrefPort:A  RlmtMode:Check Link State  
sym0: <895> rev 0x1 at pci 0000:02:0c.0 irq 20  
sym0: Tekram NVRAM, ID 15, Fast-40, LVD, parity checking  
sym0: SCSI BUS has been reset.  
scsi1 : sym-2.1.18i  
  
md: Autodetecting RAID arrays.  
md: autorun ...  
md: ... autorun DONE.  
device-mapper: 4.1.0-ioctl (2003-12-10) initialised: dm@uk.sistina.com  
8139too Fast Ethernet driver 0.9.27  
  Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D  
  Type:   Sequential-Access                  ANSI SCSI revision: 03  
sym0:5: FAST-40 WIDE SCSI 80.0 MB/s ST (25.0 ns, offset 15)  
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0  
Attached scsi generic sg1 at scsi0, channel 0, id 1, lun 0,  type 0  
Attached scsi generic sg2 at scsi1, channel 0, id 5, lun 0,  type 1  
hw_random hardware driver 1.0.0 loaded  
  
st: Version 20040318, fixed bufsize 32768, s/g segs 256  
Attached scsi tape st0 at scsi1, channel 0, id 5, lun 0  
st0: try direct i/o: yes (alignment 512 B), max page reachable by HBA  
1048575  
subfs 0.9  
  
  
st0: Block limits 1 - 16777215 bytes.  
 
dmesg for kernel 2.6.5 scsi related parts   
 
 
 Linux version 2.6.8-24.10-smp (geeko@buildhost) (gcc version 3.3.4 (pre 
3.3.5 20040809)) #1 SMP Wed Dec 22 11:54:27 UTC 2004 
 
SCSI subsystem initialized 
aacraid: falsely claims to have parameter commit 
Red Hat/Adaptec aacraid driver (1.1.2-lk2 Dec 22 2004) 
ACPI: PCI interrupt 0000:02:0a.0[A] -> GSI 22 (level, low) -> IRQ 209 
AAC0: kernel 4.0.4 build 5770 
AAC0: monitor 4.0.4 build 5770 
AAC0: bios 4.0.0 build 5770 
AAC0: serial ba2b1bfafaf001 
AAC0: Non-DASD support enabled. 
aacraid:        NMI ISR: NMI_PRIMARY_ATU_ERROR 
scsi0 : aacraid 
  Vendor: ADAPTEC   Model: Adaptec RAID5     Rev: V1.0 
  Type:   Direct-Access                      ANSI SCSI revision: 02 
SCSI device sda: 287105024 512-byte hdwr sectors (146998 MB) 
sda: Write Protect is off 
sda: Mode Sense: 03 00 00 00 
SCSI device sda: drive cache: write through 
 sda: sda1 sda2 
Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0 
  Vendor: ADAPTEC   Model: Adaptec Volume    Rev: V1.0 
  Type:   Direct-Access                      ANSI SCSI revision: 02 
SCSI device sdb: 143552512 512-byte hdwr sectors (73499 MB) 
sdb: Write Protect is off 
sdb: Mode Sense: 03 00 00 00 
SCSI device sdb: drive cache: write through 
 sdb: sdb1 sdb2 
Attached scsi removable disk sdb at scsi0, channel 0, id 1, lun 0 
Capability LSM initialized 
 
 
ACPI: PCI interrupt 0000:02:0c.0[A] -> GSI 20 (level, low) -> IRQ 217 
sym0: <895> rev 0x1 at pci 0000:02:0c.0 irq 217 
md: Autodetecting RAID arrays. 
sym0: Tekram NVRAM, ID 15, Fast-40, LVD, parity checking 
sym0: SCSI BUS has been reset. 
scsi1 : sym-2.1.18j 
md: autorun ... 
md: ... autorun DONE. 
device-mapper: 4.1.0-ioctl (2003-12-10) initialised: dm@uk.sistina.com 
evdev_connect: evdev cde12780 handle cde1279c name event0 
evdev_connect: evdev cde12580 handle cde1259c name event1 
evdev_connect: evdev cde6aa80 handle cde6aa9c name event2 
  Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D 
  Type:   Sequential-Access                  ANSI SCSI revision: 03 
scsi(1:0:5:0): Beginning Domain Validation 
scsi(1:0:5:0): Ending Domain Validation 
hw_random hardware driver 1.0.0 loaded 
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0 
Attached scsi generic sg1 at scsi0, channel 0, id 1, lun 0,  type 0 
Attached scsi generic sg2 at scsi1, channel 0, id 5, lun 0,  type 1 
 
 
  
st: Version 20040403, fixed bufsize 32768, s/g segs 256 
Attached scsi tape st0 at scsi1, channel 0, id 5, lun 0 
st0: try direct i/o: yes (alignment 512 B), max page reachable by HBA 
1048575 
subfs 0.9 
 
  
 
 

-- 
+++ Sparen Sie mit GMX DSL +++ http://www.gmx.net/de/go/dsl
AKTION für Wechsler: DSL-Tarife ab 3,99 EUR/Monat + Startguthaben

Re: sym53c8xx_2 make scsitap slow

[Matthias Andree]

"frage schnell" <frage.schnell1@gmx.de> writes:

older SuSE kernel:

>   Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D  
>   Type:   Sequential-Access                  ANSI SCSI revision: 03  
> sym0:5: FAST-40 WIDE SCSI 80.0 MB/s ST (25.0 ns, offset 15)  

newer SuSE Linux 9.2 kernel:

>   Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D 
>   Type:   Sequential-Access                  ANSI SCSI revision: 03 
> scsi(1:0:5:0): Beginning Domain Validation
> scsi(1:0:5:0): Ending Domain Validation

No information on speed here after domain validation? Suspect.

Is your bus properly terminated (terminator at the end of your cable)?
Proper U2W LVD cable in use?

Or is write validation perhaps stepping on the brake to avoid accidents
because the cabling or termination is hosed?


Example from my machine (an older Tekram adaptor as you can see,
a DC-390F, which can do 16 bit and 20 MXfer/s, only single-ended):

Linux version 2.6.8-24.10-default (geeko@buildhost) (gcc version 3.3.4 (pre 3.3.
5 20040809)) #1 Wed Dec 22 11:54:27 UTC 2004
...
ACPI: PCI interrupt 0000:00:0d.0[A] -> GSI 11 (level, low) -> IRQ 11
sym0: <875> rev 0x26 at pci 0000:00:0d.0 irq 11
sym0: Tekram NVRAM, ID 7, Fast-20, SE, parity checking
sym0: SCSI BUS has been reset.
scsi0 : sym-2.1.18j
...
  Vendor: TANDBERG  Model:  TDC 4222         Rev: =07:
  Type:   Sequential-Access                  ANSI SCSI revision: 02
scsi(0:0:6:0): Beginning Domain Validation
sym0:6: FAST-5 SCSI 4.8 MB/s ST (208.0 ns, offset 8)
scsi(0:0:6:0): Domain Validation skipping write tests
scsi(0:0:6:0): Ending Domain Validation
  Vendor: TANDBERG  Model: SLR6              Rev: 0404
  Type:   Sequential-Access                  ANSI SCSI revision: 02
scsi(0:0:8:0): Beginning Domain Validation
sym0:8: wide asynchronous.
sym0:8: FAST-20 WIDE SCSI 40.0 MB/s ST (50.0 ns, offset 15)
scsi(0:0:8:0): Domain Validation skipping write tests
scsi(0:0:8:0): Ending Domain Validation
...




Please do not forward or Cc: posts to the linux-scsi
mailing list to my mailbox, I'm on the list.

-- 
Matthias Andree

Re: sym53c8xx_2 make scsitap slow

[frage schnell]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 3593 bytes --]

 
 
  
>> older SuSE kernel: 
 
>>   Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D   
>>   Type:   Sequential-Access                  ANSI SCSI revision: 03   
>> sym0:5: FAST-40 WIDE SCSI 80.0 MB/s ST (25.0 ns, offset 15)   
 
>>newer SuSE Linux 9.2 kernel: 
 
>>   Vendor: HP        Model: Ultrium 1-SCSI    Rev: N27D  
>>   Type:   Sequential-Access                  ANSI SCSI revision: 03  
>> scsi(1:0:5:0): Beginning Domain Validation 
>> scsi(1:0:5:0): Ending Domain Validation 
 
>No information on speed here after domain validation? Suspect. 
 
>Is your bus properly terminated (terminator at the end of your cable)? 
>Proper U2W LVD cable in use? 
 
>Or is write validation perhaps stepping on the brake to avoid accidents 
>because the cabling or termination is hosed? 
  
The dmesg outputs with booth kenels ar from today  
from  the same maschine withe te same linux  
 
 
I put a new disk to the same scsi bus  
  the same strange for the tape  
  and the corect one for the disk 
 
 
 
Jan 13 14:54:24 server5 kernel: ACPI: PCI interrupt 0000:02:0c.0[A] -> GSI 
20 (l 
evel, low) -> IRQ 217 
Jan 13 14:54:25 server5 kernel: sym0: <895> rev 0x1 at pci 0000:02:0c.0 
irq 217 
Jan 13 14:54:25 server5 kernel: sym0: Tekram NVRAM, ID 15, Fast-40, LVD, 
parity 
checking 
Jan 13 14:54:25 server5 kernel: sym0: SCSI BUS has been reset. 
Jan 13 14:54:25 server5 kernel: scsi2 : sym-2.1.18j 
Jan 13 14:54:29 server5 kernel:   Vendor: HP        Model: Ultrium 1-SCSI   

Rev 
: N27D 
Jan 13 14:54:29 server5 kernel:   Type:   Sequential-Access                 

ANS 
I SCSI revision: 03 
Jan 13 14:54:29 server5 kernel: scsi(2:0:5:0): Beginning Domain Validation 
Jan 13 14:54:30 server5 kernel: scsi(2:0:5:0): Ending Domain Validation 
Jan 13 14:54:30 server5 kernel: Attached scsi tape st0 at scsi2, channel 
0, id 5 
, lun 0 
Jan 13 14:54:30 server5 kernel: st0: try direct i/o: yes (alignment 512 
B), max 
page reachable by HBA 1048575 
 
 
 
 
Jan 13 14:54:30 server5 kernel: Attached scsi generic sg2 at scsi2, 
channel 0, i 
d 5, lun 0,  type 1 
Jan 13 14:54:30 server5 kernel:   Vendor: IBM       Model: DDYS-T36950N     

Rev: SA2A 
Jan 13 14:54:30 server5 kernel:   Type:   Direct-Access                     

ANSI SCSI revision: 03 
Jan 13 14:54:30 server5 kernel: sym0:6:0: tagged command queuing enabled, 
command queue depth 16. 
 
 
 
Jan 13 14:54:30 server5 kernel: scsi(2:0:6:0): Beginning Domain Validation 
Jan 13 14:54:30 server5 kernel: sym0:6: wide asynchronous. 
Jan 13 14:54:30 server5 kernel: sym0:6: FAST-40 WIDE SCSI 80.0 MB/s ST 
(25.0 ns, offset 31) 
Jan 13 14:54:30 server5 kernel: scsi(2:0:6:0): Ending Domain Validation 
Jan 13 14:54:30 server5 kernel: SCSI device sdc: 71687340 512-byte hdwr 
sectors(36704 MB) 
Jan 13 14:54:30 server5 kernel: SCSI device sdc: drive cache: write back 
Jan 13 14:54:31 server5 kernel:  sdc: unknown partition table 
Jan 13 14:54:31 server5 kernel: Attached scsi disk sdc at scsi2, channel 
0, id 6, lun 0 
Jan 13 14:54:31 server5 kernel: Attached scsi generic sg3 at scsi2, 
channel 0, id 6, lun 0,  type 0 
Jan 13 14:54:41 server5 /etc/dev.d/block/50-hwscan.dev[6547]: new block 
device /block/sdc 
Jan 13 14:54:41 server5 /etc/dev.d/block/51-subfs.dev[6588]: mount block 
device/block/sdc 
Jan 13 14:55:51 server5 kernel: Synchronizing SCSI cache for disk sdc: 
Jan 13 14:55:51 server5 kernel: sym0: detaching ... 
Jan 13 14:55:51 server5 kernel: sym0: resetting chip 
 
 
 
  
 

-- 
+++ Sparen Sie mit GMX DSL +++ http://www.gmx.net/de/go/dsl
AKTION für Wechsler: DSL-Tarife ab 3,99 EUR/Monat + Startguthaben

Re: sym53c8xx_2 make scsitap slow

[frage schnell]

I thik i have fond the bug into the driver (see diff file bellow) 
 
 
 
--- linux-2.6.10/drivers/scsi/sym53c8xx_2/sym_hipd.c    2004-12-24 
22:33:48.000000000 +0100 
+++ linux-2.6.10.p/drivers/scsi/sym53c8xx_2/sym_hipd.c  2005-01-20 
17:31:19.364112800 +0100 
@@ -1538,7 +1538,7 @@ 
        /* 
         *  negotiate using PPR ? 
         */ 
-       if (scsi_device_dt(sdev)) { 
+        if (tp->tinfo.goal.options & PPR_OPT_MASK) { 
                nego = NS_PPR; 
        } else { 
                /* 
 
 
now the tape will be scand correct and the speed is ok 
 
Jan 20 18:16:34 server5 kernel: ACPI: PCI interrupt 0000:02:0c.0[A] -> GSI 
20 (level, low) -> IRQ 20 
Jan 20 18:16:34 server5 kernel: sym0: <895> rev 0x1 at pci 0000:02:0c.0 
irq 20 
Jan 20 18:16:34 server5 kernel: sym0: Tekram NVRAM, ID 15, Fast-40, LVD, 
paritychecking 
Jan 20 18:16:34 server5 kernel: sym0: SCSI BUS has been reset. 
Jan 20 18:16:34 server5 kernel: scsi7 : sym-2.1.18m 
Jan 20 18:16:38 server5 kernel:   Vendor: HP        Model: Ultrium 1-SCSI   

Rev: N27D 
Jan 20 18:16:38 server5 kernel:   Type:   Sequential-Access                 

ANSI SCSI revision: 03 
Jan 20 18:16:38 server5 kernel:  target7:0:5: Beginning Domain Validation 
Jan 20 18:16:39 server5 kernel: sym0:5: wide asynchronous. 
Jan 20 18:16:39 server5 kernel: sym0:5: FAST-40 WIDE SCSI 80.0 MB/s ST 
(25.0 ns, offset 15) 
Jan 20 18:16:39 server5 kernel:  target7:0:5: Ending Domain Validation 
Jan 20 18:16:40 server5 kernel: Attached scsi tape st0 at scsi7, channel 
0, id 5, lun 0 
Jan 20 18:16:40 server5 kernel: st0: try direct i/o: yes (alignment 512 
B), maxpage reachable by HBA 1048575 
Jan 20 18:16:40 server5 kernel: Attached scsi generic sg2 at scsi7, 
channel 0, id 5, lun 0,  type 1 
 
 

-- 
GMX im TV ... Die Gedanken sind frei ... Schon gesehen?
Jetzt Spot online ansehen: http://www.gmx.net/de/go/tv-spot