Michael Holzheu wrote:
>
> Hi Matt,
>
> Don't be afraid! We tested our changes carefully also on i386/Linux.
> Everything should work fine. Hopefully even better than before :-)
Great. My only issue was how it related to the dump.h changes over
from vmdump.h, and whether you had any soak time with that. But
since your method invokes off your own crash dump, it shouldn't be
a big deal.
> Regarding the question about an s390/s390x rpm, it would be good,
> if you could send me the source rpm, when you have it ready.
> Then I will ensure that it compiles on s390 and s390x.
Got it. I'll send you an E-mail and a download location when it
is ready.
> At the moment we are not allowed to ship binary rpms due to
> legal reasons.
>
> But IBM has a started a program called "Linux Community
> Development System" which provides access to a Linux/390.
>
> If you are interested in a Linxu/390 System, check the following link:
>
> http://www.ibm.com/servers/eserver/zseries/os/linux/lcds
>
> So for now, it would be sufficient to have a source rpm on your
> web site which can be compiled on Linux/390(x) without problems.
I'll read through the S/390 stuff in the morning ... :)
Thanks, Michael!
--Matt
> Best Regards
>
> Michael
>
> ------------------------------------------------------------------------
> Linux/390 Development
> Phone: +49-7031-16-2360, Bld 71032-06-109
> Email: holzheu@xxxxxxxxxx
>
> "Matt D. Robinson" <yakker@xxxxxxxxxxxxxx>@oss.sgi.com on 09/13/2001
> 03:00:47 AM
>
> Please respond to "Matt D. Robinson" <yakker@xxxxxxxxxxxxxx>
>
> Sent by: owner-lkcd@xxxxxxxxxxx
>
> To: Andreas Herrmann/Germany/IBM@IBMDE
> cc: lkcd@xxxxxxxxxxx, Luc Chouinard <lucc@xxxxxxx>, Tom Morano
> <tjm@xxxxxxx>, Michael Holzheu/Germany/IBM@IBMDE,
> yakker@xxxxxxxxxxxxxx
> Subject: Re: lkcd checkin, s390x support
>
> Andreas Herrmann wrote:
> > Hi,
> >
> > today I have checked in several files.
> > Most important is, we added lcrash support for a new platform.
> > It is s390x, which is 64 bit.
> >
> > >From our side it is ok now to create the new lkcd release.
>
> Wow, when all of you make a check-in, you don't mess around. :)
> Okay, got all the changes. Are you going to roll an RPM around
> the same time we do, or are you going to do your own? I'd need
> an S390 to roll the RPM against, which I just don't have hanging
> around the house. :)
>
> Thanks for all the bug fixes, additions, and clean-ups along
> the way. I'll go through these changes, although it will take
> me a while. Needless to say, I won't hold up 4.0 just to do
> the review -- it'll still go out Friday.
>
> --Matt
>
> > Besides that we did some further changes and bug fixes.
> > Following an overview about what we have done:
> >
> > Consolidated header files.
> >
> > libklib/include/klib.h
> >
> > libklib/include/asm-alpha/kl_error.h (removed)
> > libklib/include/asm-ia64/kl_error.h (removed)
> > libklib/include/asm-i386/kl_error.h (removed)
> > libklib/include/asm-s390/kl_error.h (removed)
> > libklib/include/kl_error.h (added)
> >
> > libklib/include/asm-alpha/kl_mem.h
> > libklib/include/asm-ia64/kl_mem.h
> > libklib/include/asm-i386/kl_mem.h
> > libklib/include/asm-s390/kl_mem.h
> > libklib/include/kl_mem.h (added for common stuff)
> >
> > libklib/include/asm-alpha/kl_stabs.h (removed)
> > libklib/include/asm-ia64/kl_stabs.h (removed)
> > libklib/include/asm-i386/kl_stabs.h (removed)
> > libklib/include/asm-s390/kl_stabs.h (removed)
> > libklib/include/kl_stabs.h (added)
> >
> > Included support for type information of so called register variables
> > (N_RSYM in stabs format).
> >
> > libklib/kl_stabs.c
> >
> > Included support for intels PSE (Page Size Extension) in memory mapping:
> >
> > libklib/arch/alpha/kl_page.c
> > libklib/arch/ia64/kl_page.c
> > libklib/arch/i386/kl_page.c
> > libklib/arch/s390/kl_page.c
> > libklib/kl_mem.c
> > libklib/kl_memory.c
> >
> > Changed vtop behaviour: switched on mem mapping for all virtual addresses
> >
> > lcrash/cmds/cmd_vtop.c
> >
> > s390 specific changes:
> >
> > lcrash/arch/s390/lib/s390-report.c
> > lcrash/arch/s390/lib/s390-util.c
> > lcrash/arch/s390/lib/trace.c
> > lcrash/include/arch-s390/trace.h
> > libklib/arch/s390/kl_s390_util.c
> >
> > Removed unnecessary include directives to kernel header files. Especially
> > the
> > handling of struct utsname was changed.
> >
> > lcrash/arch/s390/cmds/cmd_s390dbf.c
> > lcrash/cmds/cmd_stat.c
> > lcrash/include/lcrash.h
> > lcrash/util.c
> > libklib/include/dump.h
> >
> > BUGFIXES:
> >
> > - Inserted setjmp before first call to longjump.
> >
> > lcrash/main.c
> >
> > - "dis -F <funcname> -w file", rediretion of output did not work
> correctly.
> >
> > lcrash/arch/alpha/lib/dis.c
> > lcrash/arch/ia64/lib/dis.c
> > lcrash/arch/s390/lib/dis.c
> >
> > - Setup type info immediately after reading namelist.
> >
> > libklib/kl_nmlist.c
> >
> > - Avoid printing of control characters.
> >
> > libklib/kl_print.c
> >
> > - Kernel without module support can be analyzed again.
> >
> > lcrash/cmds/cmd_symtab.c
> > lcrash/cmds/cmd_module.c
> > libklib/kl_util.c
> > libklib/klib.c
> >
> > - Fixed the computation of memory size in livedump.
> >
> > lcrash/vmdump.c
> >
> > - s390x port:
> >
> > (a whole bunch of files)
> >
> > Regards,
> >
> > Andreas
> >
> > --
> > Linux for eServer Development
> > Tel : +49-7031-16-4640
> > Notes mail : Andreas Herrmann/GERMANY/IBM@IBMDE
> > email : aherrman@xxxxxxxxxx
|