Daria Mehra wrote:
> Mike Sullivan wrote:
>> Daria Mehra wrote:
>>
>>> Please advise soon whether I must make this change, since the
>>> turnaround time for a nightly build is around 10 hours these days,
>>> and I am targeting build 99 for integration.
>>
>> Making a few changes then using a full nightly build to see if your
>> changes work is usually vastly overkill.
>
> Clearly so. I follow the steps you listed (except for 'checkproto' which
> I didn't know about), but still I'd like to do a full nightly build on
> both sparc and amd before turning in the c-team checklist.
and I'd like you to as well, since I don't want to be yelling at you
after you putback and the build has issues. And you don't want that
either :)
>
> Question... in testing my package, I noticed that if I try installing
> the sparc version of SUNWlibnet on an amd machine, it "works"; the
> package installation process creates sparcv9 directory and reports
> success. I thought installation would fail after running some common
> check for architecture, but that doesn't seem to be the case. Is this
> normal?
yes. It's not necessarily intuitive for live use but that's so you could
be, say, setting up a diskless sparc client on your amd64 server.
Mike
