RFC: backport S7103224 to icedtea6

[Omair Majid]

On 10/26/2011 06:53 PM, Dr Andrew John Hughes wrote:
> On 17:41 Wed 26 Oct     , Omair Majid wrote:
>> Hi,
>>
>> I would like to backport the fix for 7103224 to icedtea6. This bug
>> allows icedtea6 to be built with (very) new glibc versions. Please see
>> the original email for details:
>> http://mail.openjdk.java.net/pipermail/hotspot-dev/2011-October/004589.html
>>
>> I have attached the patches for IcedTea6 HEAD and 1.10.
>>
>> ChangeLog:
>> 2011-10-25  Omair Majid<omajid at redhat.com>
>>
>>      * patches/openjdk/7103224-glibc_name_collision.patch: New file.
>>      * Makefile.am (ICEDTEA_PATCHES): Add the above.
>>
>> Does anyone have any thoughts or concerns?
>>
>
> This patch seems to do more than the summary says.  Why
> is MAXFLOAT being removed for example?
>

Someone from Apple pointed out another name clash between MAXFLOAT and 
the Apple compilers/libc [1] very soon after I posted my patch. I guess 
it's too much of a hassle to create a separate bug for a one-line fix 
removing MAXFLOAT and so Tom decided to combine the changes [2]. As 
mentioned in the email, MAXFLOAT is not used anywhere so removing it 
should be fairly safe.

> You can commit to HEAD without approval.  I suggest you do
> this and let it soak for some time before backporting to
> 1.10.
>

Sure. I think we will be applying this patch in Fedora locally then; it 
is causing icedtea6-1.10.4 (the security update!) builds to fail for F16 :(

> What about 7 HEAD and 7 2.0?
>

Unfortunately, my machine cant handle too many concurrent icedtea builds 
:) I will post patches after I have built and tested icedtea7.

Thanks,
Omair

[1] 
http://mail.openjdk.java.net/pipermail/hotspot-dev/2011-October/004591.html
[2] 
http://mail.openjdk.java.net/pipermail/hotspot-dev/2011-October/004594.html