Hi Alok,

Alok Aggarwal wrote:
> Hi Jan,
>
> On Wed, 11 Nov 2009, Jan Damborsky wrote:
>
>> Hi Jean, Alok,
>>
>> could I please ask you to review fix for following bugs ?
>>
>> <http://defect.opensolaris.org/bz/show_bug.cgi?id=7966>7966 setting 
>> 'livemode=text' is no-op for AI, should be removed from installadm(1M)
>> 8140  Fix for 6440 applied to x86 AI images would reduce x86 AI 
>> memory consumption
>>
>>
>> Thank you very much,
>> Jan
>>
>> webrev:
>> http://cr.opensolaris.org/~dambi/bug-8140/
>
> installadm-common.sh: lines 463-464: Sorry if I'm dense but
> I don't see why this check is needed. If x86.microroot exists,
> it also means that split 32/64 bit archives don't exist; so
> we should pick up the boot archive from /${BootLofs}/x86.microroot.
>
> Otherwise, pick it up from the appropriate platform directory.
> Which cases would this miss?

You are right. Too much 'ifs' as Jean also pointed out :-)
I have removed one of them and added comment.

The webrev has been updated accordingly.

>
> installadm-common.sh: line 464: Is the "\" before $ISADIR really
> needed?

Yes, since otherwise $ISADIR would be evaluated and we would end up with

module$ /${BootLofs}/platform/i86pc//boot_archive

in GRUB instead of desired

module$ /${BootLofs}/platform/i86pc/$ISADIR/boot_archive


>
> ai_plat_setup.py: I wonder why we didn't catch this while testing
> 8347.

Creating that symlink was intentional - it was stop gap solution before 
fix for 8140
is integrated in order not to break 64bit XEN PV, as it scans following 
locations for
64 bit boot archive:

/boot/x86.microroot
/boot/amd64/x86.microroot
/platform/i86pc/amd64/boot_archive


More details about XEN dependencies on install can be seen at:

http://hub.opensolaris.org/bin/view/Community+Group+xen/install-depedencies

Thank you very much for review !
Jan


Reply via email to