Hardware

Somebody has reported that making the 0680 chip based PCI-to-IDE card worked on Blade1000.
http://forums.sun.com/thread.jspa?threadID=5067148&start=15.

Now I want to expand the storage for Blade1000 by the Adaptec-ASH-1233 which is powered by SiI0680A chip.
But it does NOT work.

Any body can help me? Thanks in advance!!!


The following is the detailed infomation:

My system env:
Blade 1000 with solaris 10 U7(5/09).

I have tried the following (they all dose NOT work):

1. The card with original bios v3.2.10 (prtconf -pv report class-code: 00018000)

2. update bios to base bios (none raid, b3308.bin) v3.3.08 (class-code: 00010185)

3. erase the bios flash (class-code: 00010185)
( I think the bios is programed by x86 instruction set , it will not work in sparc.
by the way, any Solaris(sparc) device driver expert can clear that if the dirver will communication with the card bios?
If it dose , I think it is imporssible to make it work in sparc system)

4. Follwing a suggestion from sun HCL and repeat 2 ,3 ( class-code: 00018000)
add the following nvramrc script in the obp (nvedit)
0: probe-all install-console banner
1: cd /pci@8,700000/ide@1
2: 00018000 " class-code" integer-property
3: 00000680 " subsystem-id" integer-property
4: " pci1095,680" " name" string-property
5: device-end
Adjust the path on line 1 to reflect your system store with "nvstore", set the "use-nvramrc?" variable to "true" and issue a "reset".
After a boot the controller will look just as a non-RAID controller.




The following is the system info:
################################################################################
with a maxtor 250G disk connected OR a Western Digital 300G disk connected


<1> prtdiag :

pci 33 +s/system-board ide-pci1095,680
okay /pci@8,700000/ide@1



<2> prtconf -pv :


Node 0xf00a9874
assigned-addresses: 81000810.00000000.00000500.00000000.00000008.81
000814.00000000.00000508.00000000.00000004.81000818.00000000.00000510.00000000.0
0000008.8100081c.00000000.0000050c.00000000.00000004.81000820.00000000.00000520.
00000000.00000010.82000824.00000000.0012c000.00000000.00002000.82000830.00000000
.00180000.00000000.00080000
reg: 00000800.00000000.00000000.00000000.00000000.01000810.00000000
.00000000.00000000.00000008.01000814.00000000.00000000.00000000.00000004.0100081
8.00000000.00000000.00000000.00000008.0100081c.00000000.00000000.00000000.000000
04.01000820.00000000.00000000.00000000.00000010.02000824.00000000.00000000.00000
000.00002000.02000830.00000000.00000000.00000000.00080000
compatible: 'pci1095,680.1095.680.8502' 'pci1095,680.1095.680' '
pci1095,680' 'pci1095,680.8502' 'pci1095,680' 'pciclass,010185' + 'pciclas
s,0101'
name: 'ide' (is raid in my try 1)
fast-back-to-back:
devsel-speed: 00000001
class-code: 00010185 (is 00018000 in my try 1 and 4)
interrupts: 00000001
latency-timer: 00000040
cache-line-size: 00000010
max-latency: 00000000
min-grant: 00000000
subsystem-id: 00000680
subsystem-vendor-id: 00001095
revision-id: 00000002
device-id: 00000680
vendor-id: 00001095



<3> prtconf -D :

ide, instance #1 (driver name: uata)
dad, instance #0 (driver name: dad)



<4> /var/adm/messages:

Jun 6 17:39:51 bird pcisch: [ID 370704 kern.info] PCI-device: ide@1, uata1
Jun 6 17:39:51 bird genunix: [ID 936769 kern.info] uata1 is /pci@8,700000/ide@1
Jun 6 17:39:51 bird ebus: [ID 521012 kern.info] gpio_873170 at ebus0: offset 1,
300600
Jun 6 17:39:51 bird genunix: [ID 936769 kern.info] gpio_873170 is /pci@8,700000
/ebus@5/gpio@1,300600
Jun 6 17:39:51 bird uata: [ID 114370 kern.info] dad0 at uata1
Jun 6 17:39:51 bird uata: [ID 347839 kern.info] target 0 lun 0
Jun 6 17:39:51 bird genunix: [ID 936769 kern.info] dad0 is /pci@8,700000/ide@1/
dad@0,0
Jun 6 17:39:51 bird pseudo: [ID 129642 kern.info] pseudo-device: trapstat0
Jun 6 17:39:51 bird genunix: [ID 936769 kern.info] trapstat0 is /pseudo/trapsta
t@0
Jun 6 17:39:51 bird cmlb: [ID 107833 kern.warning] WARNING: /pci@8,700000/ide@1/dad@0,0 (dad0):
Jun 6 17:39:51 bird Corrupt label; wrong magic number



<5> run any of the following commands will cause the system to reboot:


# devfsadm
# ls /devices/pci@8,700000/ide@1/
# devinfo -i /devices/pci@8,700000/ide@1/dad@0,0

and give the following messages:

SUNW-MSG-ID: SUNOS-8000-0G, TYPE: Error, VER: 1, SEVERITY: Major
EVENT-TIME: 0x4a502cda.0x1dcd9638 (0x30075ddaae0)
PLATFORM: SUNW,Sun-Blade-1000, CSN: -, HOSTNAME: bird
SOURCE: SunOS, REV: 5.10 Generic_139555-08
DESC: Errors have been detected that require a reboot to ensure system
integrity. See http://www.sun.com/msg/SUNOS-8000-0G for more information.
AUTO-RESPONSE: Solaris will attempt to save and diagnose the error telemetry
IMPACT: The system will sync files, save a crash dump if needed, and reboot
REC-ACTION: Save the error summary below in case telemetry cannot be saved

###########################################################################################
if a old Quantum Fireball 20G connected , the following is different with 250G/300G disk connected

<1> prtconf -D :

ide, instance #1 (driver name: uata)
<2> the system NOT report "Corrupt label; wrong magic number"
<3> And the above three commans in <5> will NOT reboot the system