hg: hsx/hsx24/hotspot: 8001192: allow duplicate bug ids in hs24
David Holmes
david.holmes at oracle.com
Mon Oct 22 18:16:54 PDT 2012
On 22/10/2012 11:00 PM, Volker Simonis wrote:
> On Mon, Oct 22, 2012 at 1:42 PM, David Holmes<david.holmes at oracle.com> wrote:
>> On 22/10/2012 5:42 PM, Volker Simonis wrote:
>>>
>>> Hi,
>>>
>>> could you please explain why we need duplicate bug ids in hsx24?
>>
>>
>> The same reason as needed in other hsx version: changesets covering the same
>> bugids can come in from different paths. The fixes are the same bit they are
>> distinct changesets.
>>
>
> Yes, but this is something I don't understand. Why don't you use
> different bug IDs for different changesets?
> Anyway you create new bug IDs for backports and other codelines (those
> starting with 2xxxxxx). So why not using them when checking in such
> 'distinct changesets' into other codelines?
In the past we have only had the primary bug id to use - even for
backports coming in through different paths. With Jira we do now have
much more obvious distinct issue numbers for the backports, but the
process has been re-affirmed to continue to use the primary bug id for
changesets.
I can't address the "why" aspect of this.
David
-----
> Additionally, it is not always true that the fixes are "really" the
> same. Sometimes they have to be adapted (e.g. when they are backported
> from the hsx main-line to an older hsxYY stabilization code line).
> Having different bug IDs for such changes would make thinks a lot
> clearer from my point of view.
>
> Regards,
> Volker
>
>>> Apparently this was not necessary for hsx23.
>>
>>
>> I think you will find it is necessary in most of the hsx families.
>>
>> David
>>
>>
>>>
>>> Regards,
>>> Volker
>>>
>>> On Sat, Oct 20, 2012 at 1:39 AM,<alejandro.murillo at oracle.com> wrote:
>>>>
>>>> Changeset: 81d17d9ab7c9
>>>> Author: amurillo
>>>> Date: 2012-10-19 16:37 -0700
>>>> URL: http://hg.openjdk.java.net/hsx/hsx24/hotspot/rev/81d17d9ab7c9
>>>>
>>>> 8001192: allow duplicate bug ids in hs24
>>>> Reviewed-by: jcoomes
>>>>
>>>> ! .jcheck/conf
>>>>
>>
More information about the hotspot-dev
mailing list