[m-rev.] for review: port Mercury to MacOS X 10.6

Julien Fischer juliensf at csse.unimelb.edu.au
Thu Oct 8 18:21:08 AEDT 2009




On Thu, 8 Oct 2009, Ian MacLarty wrote:

> On Wed, Oct 7, 2009 at 10:54 PM, Julien Fischer
> <juliensf at csse.unimelb.edu.au> wrote:
>>
>> On Wed, 7 Oct 2009, Ian MacLarty wrote:
>>
>>> On Wed, Oct 7, 2009 at 10:48 PM, Julien Fischer
>>> <juliensf at csse.unimelb.edu.au> wrote:
>>>>
>>>> On Wed, 7 Oct 2009, Ian MacLarty wrote:
>>>>
>>>>> On Wed, Oct 7, 2009 at 9:22 PM, Julien Fischer
>>>>> <juliensf at csse.unimelb.edu.au> wrote:
>>>>>>
>>>>>> On Wed, 7 Oct 2009, Ian MacLarty wrote:
>>>>>>
>>>>>>>>> -    # Suppress a warning from the linker on Darwin 9.  `-s' is
>>>>>>>>> deprecated
>>>>>>>>> on it
>>>>>>>>> -    # and has no effect other than causing the linker to emit a
>>>>>>>>> warning.
>>>>>>>>> +    # Suppress a warning from the linker on Darwin 9+.  `-s' is
>>>>>>>>> deprecated on
>>>>>>>>> +    # it and has no effect other than causing the linker to emit a
>>>>>>>>> warning.
>>>>>>>>>    #
>>>>>>>>>    case "$host" in
>>>>>>>>> -        *apple*darwin*9*)
>>>>>>>>> +        i686*apple*darwin*)
>>>>>>>>>            LD_STRIP_FLAG=""
>>>>>>>>
>>>>>>>> Why is this change here?  Won't it break the Mac OS X PPC port? i.e.
>>>>>>>> don't you want to match against *apple*darwin*9*?  (Is the warning
>>>>>>>> present with Darwin 10?)
>>>>>>>>
>>>>>>>
>>>>>>> Yes the warning is present in Darwin 10.  I wanted to turn it off for
>>>>>>> all future versions of Intel Darwin too. As far as I know Mac OS X
>>>>>>> 10.6 is Intel only.  I think 10.5 is Intel only also.
>>>>>>
>>>>>> No, 10.5 is also PPC.
>>>>>>
>>>>>
>>>>> Okay, I'll change it to check for 9 and 10 then.
>>>>
>>>> That wasn't my point - you need to remove the i686 part, because it will
>>>> also need to match "powerpc".
>>>>
>>>
>>> What about 10.3 where -s wasn't a problem?  I suppose it doesn't hurt
>>> to remove it for all version.
>>
>> I suggest leaving any 10.3 related stuff alone until we are in a
>> position to test it again.
>>
>
> I'm confused about what you want me to do.  Do you want me to remove
> the i686 part or not?

I want this:

 	case "$host" in
 		*apple*darwin*9*|*apple*darwin*10)
 			LD_STRIP_FLAG=""
 		;;
 	esac

i.e. the existing behaviour is unchanged for 10.3 and 10.4 and we don't
pass -s to the linker on 10.5 with PPC.  (The i686 part should not be
there for 10.5 and is irrelevant for 10.6 because that is Intel-only
anyway.)

Julien.


More information about the reviews mailing list