Hello Chuck,
The problem you report is known since a few time. It is a bug from me and
only occurs with SYM53C896 revision 1 and PROFILE_OPTION enabled for the
sym53c8xx driver. I have a 896 rev. 1 and a 896 rev. 3 but donnot use the
rev. 1 since a long time.
The below patch should fix:
--- sym53c8xx.c.0123 Sun Jan 23 15:52:27 2000
+++ sym53c8xx.c Sun Jan 23 15:52:42 2000
@@ -5481,7 +5481,7 @@
#ifndef SCSI_NCR_PROFILE_SUPPORT
#define XXX 0
#else
-#define XXX 3
+#define XXX 2
#endif
np->script0->dataphase[XXX] = cpu_to_scr(SCR_JUMP);
np->script0->dataphase[XXX+1] =
------------ Cut Here -------------
Btw, the 'profile option' is not that useful. I moved it from ncr to sym
for people who might have liked it so much that they would have been
extremally disappointed not to have it for the sym53c8xx driver. :-)
If you donnot set this driver option in the kernel config, the problem
will also vanish.
To complete your revisit of the drivers:), I invite you to have a look at:
ftp://ftp.lsil.com/HostAdapterDrivers/linux/
Pamela Delaney from LSILOGIC has implemented Ultra-3 support for the
53C1010 in the sym53c8xx driver and SCSI domain validation.
She did an excellent work. Latest driver versions are sym-3.31 and
ncr-1.61. For now, only patches against stable kernels are available.
These driver versions have been heavily tested by LSILOGIC and will be
candidate soon for both stable and latest Linux kernels.
G�rard.
On Fri, 21 Jan 2000, Chuck Lever wrote:
> possible bug here, or maybe i'm misconfiguring.
>
> the system is a four-way intel SC450NX with a sym53c810 on one PCI bus,
> and two sym53c896a's on a second PCI bus. kernel is 2.3.39.
>
> if i build the kernel with the ncr53c8xx driver, no SCSI problems.
>
> if i build the kernel with the sym53c8xx driver, the system loops during
> the driver's detection phase with these messages on the console:
>
> sym53c896-1: restart (scsi reset).
> sym53c896-1: handling phase mismatch from SCRIPTS.
> sym53c896-1: Downloading SCSI SCRIPTS.
> sym53c896-1: 0: ERROR (81:0) (3-a7-0) (0/7) @ (script 200:fe0013e0)
> sym53c896-1: script cmd = 800c0000
> sym53c896-1: regdump da 10 80 07 47 00 00 02 02 03 00 a7 80 01 07 09 00 c0
> 9a 01 08 00 00 00
>
> -- repeat ad nauseum --
>
> (copied by hand; serial console isn't working at the moment)
>
> i've also seen the looping behavior with a 2.2.13 kernel built with
> sym53c8xx. this kernel had booted previously, now suddenly it won't.
>
> the "assume boards are SYMBIOS compatible" option has not been enabled.
> i've also tried disabling the master parity bit, as was suggested by a
> previous poster on this list, to no avail.
>
> i'm not much of a SCSI expert, and could use a pointer. thanks!
>
> - Chuck Lever
> --
> corporate: <[EMAIL PROTECTED]>
> personal: <[EMAIL PROTECTED]> or <[EMAIL PROTECTED]>
>
> The Linux Scalability project:
> http://www.citi.umich.edu/projects/linux-scalability/
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
