LGTM.
--Karen
On 04/26/12 12:55, Drew Fisher wrote:
Good afternoon!
My push for the fix of 7164145 actually broke SPARC installs to whole
disk on VTOC. I need to get a code review for the following CRs:
7120436 <http://monaco.us.oracle.com/detail.jsf?cr=7120436> TI throw
exception if try to delete extended partition and install in solaris2 partition
7164336 <http://monaco.us.oracle.com/detail.jsf?cr=7164336> new pep8 error in
usr/src/cmd/text-install/partition_edit_screen.py after changeset 1655:4e0a5494f84
7164689 <http://monaco.us.oracle.com/detail.jsf?cr=7164689> The fix for7164145
<http://monaco.us.oracle.com/detail.jsf?cr=7164145> causes issues for SPARC
incapable of installing to GPT
https://cr.opensolaris.org/action/browse/caiman/drewfish/ti_bugs_2/webrev/
For 7164689, we need to pass an extra attribute on the desired Disk
object to tell text-install to skip the partition edit screen. Since
we track if the user selects "Use the Whole Disk" or "Use a Partition
on the Disk" in the UI, I chose to use that variable as the attribute
to the Disk object. We can't rely on Disk.whole_disk because, for
VTOC, it's toggled to False due to the need to manually lay out slices
to hand to ZFS.
I've tested this on both x86 (where Disk.whole_disk = True due to
grub2 being able to handle GPT) and on a SPARC machine incapable of
installing to GPT. I no longer get the traceback as shown in 7164689
on either system.
For 7120436, Karen reviewed this change yesterday but additional eyes
are always welcome.
-Drew
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
