[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sisuite-users
Subject:    [Sisuite-users] Sun V6x Progress Report and Open Issues
From:       Stan Heller <Stanley.Heller () Sun ! COM>
Date:       2003-11-19 18:18:55
[Download RAW message or body]


Hello all,

Yes. We're back. Thanks to feedback last time, we were
able to successfully get a client to load on the V65.
There are still some issues getting the load to happen
without manual intervention. Enclosed is my partner
Dave's summary notes of the issues we are having. If
needed, I will repost my previous message which gives
all of our environmental config information if requested.
This is again for version 3.1.5, which I know is unstable.
Maybe this will help point the way to making it more stable.


Here are Dave's notes:
==========================

1) ASM CODE ERROR
make clean
make all

gcc -E -D__KERNEL__ -I/tmp/systemimager-3.1.5/src/linux-2.4.20/include  -traditional \
-DCHIP=710 fake7.c | grep -v '^#' | perl -s script_asm.pl -ncr710 script_asm.pl : \
Illegal combination of registers in line 72 :   MOVE CTEST7 & 0xef TO CTEST7  Either \
source and destination registers must be the same,  or either source or destination \
register must be SFBR. make[3]: *** [sim710_d.h] Error 255

It appears that ether the Perl script script_asm.pl has some problem handling the asm \
line: MOVE CTEST7 & 0xef TO CTEST7

or, there is a syntax error with this line. I switched it to the following:
MOVE CTEST7 TO SFBR
MOVE SFBR & 0xef TO CTEST7

2) INITRD.RUL PROBLEM 
The file: /systemimager-3.1.5/initrd_source/initrd.rul has the following line:
find $(INITRD_BUILD_DIR) -depth -exec touch -t 196912311900 {} \;

Redhat 8.0 did not like the date but it was okay with:
find $(INITRD_BUILD_DIR) -depth -exec touch -t 197012311900 {} \;

3) LOAD_MY_MODULES PROBLEM
The script: systemimager-3.1.5/initrd_source/skel/etc/init.d/rcS
has a call to the function load_my_modules right after the call to ifconfig_loopback. \
This is apparently too early in the sequence of events as our calls to insmod in the \
INSMOD_COMMANDS script were all failing. Moving the load_my_modules call further down \
to just before the autodetect_hardware_and_load_modules function call seems to \
resolve this.

4) Even though the INSMOD_COMMANDS script is now functioning and our SCSI devices are \
being loaded, it appears that the SCSI modules (scsi_mod.o, sd_mod.o, aic79xx.o) are \
not getting loaded using our versions that we put in the my_modules directory but \
instead are getting them from the /lib/modules/2.4.20-boel/kernel/drivers/scsi/ \
directory. This would be a problem were we to try and load some other driver that is \
not supported by SI.

5) While troubleshooting the rcS, functions and INSMOD_COMMANDS scripts, it was not \
real clear what the correct make target we should use to be certain that our changes \
would be included. Sometimes we would see the previous scripts on the target.

==========================

Again, any insight the community can give us would be appreicated. We are SO CLOSE.

					=stan

-- 
Stan Heller
Systems Engineer
Sun Microsystems Professional Services
Customer Ready Systems   (415) 254-5667


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
_______________________________________________
Sisuite-users mailing list
Sisuite-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sisuite-users


[prev in list] [next in list] [prev in thread] [next in thread]