Thread View: comp.os.linux.hardware
2 messages
2 total messages
Started by Danial Rubin
Tue, 15 Aug 1995 17:49
Problems talking to SCSI disk at boot (Linux/NCR53c810)
Author: Danial Rubin
Date: Tue, 15 Aug 1995 17:49
Date: Tue, 15 Aug 1995 17:49
56 lines
3034 bytes
3034 bytes
Hello,
I recently put a NCR53c810 PCI SCSI card in my ACER montherboard with
the NCR bios. After finally getting the SCSI card correctly
configured, as far as I can tell, I am having troubles getting it to
talk to my SCSI IBM 0663-E15 1.0 gig hard disk. During the booting of
Linux everything seems ok until the SCSI config phase gets to that disk.
I also have an Archive Viper 150 QIC tape in the SCSI chain and it
configures that ok. Then it hangs on the SCSI disk (next target). When
I unplug the power to the disk and reboot all is well. I successfully
loaded a cpio file off of the tape drive and it works great. Here is
an output of the kernel during the SCSI config phase at boot time with both
the hard disk and the tape drive connected and powered up:
scsi-ncr53c7,8xx : at PCI bus 0, device 3, function 0
scsi-ncr53c7,8xx : warning : revision of 2 is greater than 1.
scsi-ncr53c7,8xx : NCR53c810 at memory 0x2000000, io 0xd000, irq 15
scsi0 : using io mapped access
scsi0 : using initiator ID 0
scsi0 : using edge triggered interrupts.
scsi0 ; burst length 2
scsi0 : using 40MHz SCSI clock
scsi0 : m_to_n = 0x10, n_to_m = 0x20, n_to_n = 0x30
scsi0 : NCR code relocated to 0x20d74c
scsi0 : testing
scsi0 : test 1 started
scsi0 : tests complete.
scsi0 : NCR53c{7,8}xx (rel 3)
scsi : 1 host.
Vendor: ARCHIVE Model: VIPER 150 21247 Rev: -005
Type: Sequential-Access ANSI SCSI revision: 01
Detected scsi tape st0 at scsi0, id 1, lun 0
scsi0: target 2 requesting synchronous transfer transfer period 100ns, offset 12
scsi0: setting target 2 to asynchronous SCSI
The boot process just hangs after the "scsi0: setting target 2 to
asynchronous SCSI" line. I have to power cycle the system. I did notice the
mismatch between synchronous and asynchronous transfers in the last two lines
but I don't know enough about what was going on at the low level to
understand what it means. I had the drive working in a Sun workstation just
fine prior to attempting to put it in my PC.
In the literature (2 pages...) that came with the hard disk there is the
following paragraph which may be relevant:
Comments:
The 0663-E15 drive listed above has modified microcode which does not return
the 'Unit Attention' condition in response to the first SCSI command sent
following a power up or SCSI bus reset. Additionally the microcode on the
drive supports the 'Blind Write' feature commonly used by Apple Platforms.
Both of these features are needed to allow attachment to the Apple environment.
Any ideas to what is going on are appreciated....
Thanks
- Dan
Re: Problems talking to SCSI disk at boot (Linux/NCR53c810)
Author: Etienne Pralong
Date: Wed, 23 Aug 1995 08:05
Date: Wed, 23 Aug 1995 08:05
81 lines
3529 bytes
3529 bytes
In article <40qmn5$31g@rigel.infinet.com>, rubin@infinet.com says...
>
>Hello,
>
>I recently put a NCR53c810 PCI SCSI card in my ACER montherboard with
>the NCR bios. After finally getting the SCSI card correctly
>configured, as far as I can tell, I am having troubles getting it to
>talk to my SCSI IBM 0663-E15 1.0 gig hard disk. During the booting of
>Linux everything seems ok until the SCSI config phase gets to that disk.
>I also have an Archive Viper 150 QIC tape in the SCSI chain and it
>configures that ok. Then it hangs on the SCSI disk (next target). When
>I unplug the power to the disk and reboot all is well. I successfully
>loaded a cpio file off of the tape drive and it works great. Here is
>an output of the kernel during the SCSI config phase at boot time with both
>the hard disk and the tape drive connected and powered up:
>
I hope it will help this a copy of Warp Pharmacy from
http://p16.sma.ch:8080/~jos/NCR53c810.html
Symptoms
NCRADD: One or more of your SCSI devices is malfunctioning.
Hardware
NCR 53c810 SCSI Adapter
Problem
The adapter has not been configured to use level triggering.
Procedure
Change the adapter so that it uses LEVEL triggering as per the
Notes
I have had problems with my NCR 53c810 since februar 1995 with OS/2 Warp ver.
3.0. I got the very same error message as
you:
NCRADD: One or more of your SCSI devices is malfunctioning.
I have found that it is not the NCR 53c810 nor the OS2CAM.ADD which has guilt
in making this trouble. The problem is the
MANUAL for the NCR 53c810 which recommends to use EDGE triggering.
I trusted the manual to be correct in this statement. But after sending and
receiving some email to NCR.
I found that it is the LEVEL triggering which works with OS2CAM.ADD.
After I changed to LEVEL triggering on the NCR 53c810 (with jumper settings)
all my SCSI devices works in
OS2CAM.ADD. I have a Pioneer DRM 604x SCSI CD-ROM player, and now it works very
well.
I am also able to install LINUX on my computer when using LEVEL triggering. I
have tried to install LINUX on my computer
since Autumn 1994 with no success. I could not access any SCSI devices earlier.
But suddenly everything worked with LEVEL
triggering.
But LINUX is even more specific to configure than OS/2 WARP. I had to put my
motherboard to INT C (with jumpers) for
the PCI slot 1, and my NCR 53c810 to INTR C# LEVEL triggering (with jumpers).
In the CMOS BIOS Setup I had to put
my NCR 53c810 to irq=11, and PCI slot 1 to irq=11. It was important to put the
irq=11 two places in the CMOS BIOS
Setup for LINUX installation. The MANUAL for my motherboard GA 586 IS told me
that INT C requires irq=11. The
motherboard MANUAL said:
INT A means irq=9
INT B means irq=10
INT C means irq=11
INT D means irq=14
I hope this makes it clearer to other users of the NCR 53c810.
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads