Reviewer needed: backport of 6796786 and 7042070 fixes into IcedTea6 HEAD

[Pavel Tisnovsky]

Andrew John Hughes wrote:
> On Tue, Jun 28, 2011 at 03:17:28PM +0200, Jiri Vanek wrote:
>> On 06/24/2011 09:49 AM, Pavel Tisnovsky wrote:
>>> Hi all,
>>>
>>> it is possible to backport 6796786 and 7042070 fixes into IcedTea6 HEAD
>>> please?
>>>
>>> hg diff generated against recent IcedTea6 HEAD is included in an
>>> attachment. Please note, that these patches should be applied with the
>>> following order (6796786 ->  7042070) because the second patch simply
>>> fixes issue in a regression test added through the first patch.
>>>
>>> ChangeLog entry:
>>>
>>> 2011-06-24  Pavel Tisnovsky<ptisnovs at redhat.com>
>>>
>>>         * Makefile.am: added new patches
>>>         * NEWS: updated with backports
>>>         * patches/openjdk/6796786-invalid_FP_identity_transform.patch:
>>>         * patches/openjdk/7042070-Typo_in_Test6796786.patch:
>>>         Backport of 6796786 and 7042070 fixes.
>>>
>>> Can anybody please review this change?
>>>
>>> Thank you in advance,
>>> Pavel
>>
>> I have walked through  patch(es).  With my poor knowledge of C I can not judge 100% (but when it is backport it can be enough) but java side looks  OK.
>> Build and test run on F13 upon HEAD is ok for both backports. Please slow down!-D
>>
> 
> Yes, please slow down a bit.  Not only is a little annoying having to
> continually merge when working on stuff for HEAD, as it invalidates my
> testing, but you're not really allowing sufficient time for people to
> look at these (the opposite of before where it was too slow...).

ah I see. Do you think that one simple patch per day is acceptable? I'm
going to synchronize mainly AWT and Java2D stuff and these changes are
usually quite simple.

> 
> In particular, this is a HotSpot patch.  Have you tested it on both
> supported versions?  I see no indication of that in your submission.

Yes, this patch is (now ;-) tested against hs19 and hs20. Thank you for
pointing me out that it could be a problem patching older hs (it's
obvious, but I forget to check that)

> 
>> Enjoy your commit:)
>>
>> J.
>