Re: RFR(XS) 8148785: Update class file version to 53 for JDK-9

[harold seigel]

Hi Alan,

Thanks for looking at the change.   I'll drop the comment before 
checking it in.


Harold

On 2/3/2016 9:32 AM, Alan Bateman wrote:



On 03/02/2016 14:12, harold seigel wrote:

Hi Aleksey,

Thanks for the review.  The plan is to first change the runtime to 
accept version 53 files and then change the tools to generate them. 
Hopefully, this will reduce incompatibility problems.  See 
JDK-8148651  for 
details.
Right, there are several things that need to happen and I think we 
need hotspot accepting 53.0 before javac starts to generate them.


The webrevs look okay to me, except it might be better to drop "New 
Module attribute" from the comment in classfileParser.cpp until JSR 
376 is further along and we line up the trucks to bring the module 
system into JDK 9.


-Alan

Re: RFR(XS) 8148785: Update class file version to 53 for JDK-9

[harold seigel]

Hi Aleksey,

Thanks for the review.  The plan is to first change the runtime to 
accept version 53 files and then change the tools to generate them. 
Hopefully, this will reduce incompatibility problems.  See JDK-8148651 
 for details.


Thanks, Harold

On 2/3/2016 8:36 AM, Aleksey Shipilev wrote:

On 02/03/2016 04:16 PM, harold seigel wrote:

Open webrevs:
   http://cr.openjdk.java.net/~hseigel/bug_8148785.jdk/
   http://cr.openjdk.java.net/~hseigel/bug_8148785.hs/

+1


JBS Bug: https://bugs.openjdk.java.net/browse/JDK-8148785

Description seems to imply we change the compiled class file versions to
53.0. But this change only makes VM to accept 53.0, it does not actually
produce 53.0 files yet. Are we sure the patches do what 8148785 intended?

Anyhow, there seems to be little risk with this particular change.

-Aleksey

Re: RFR(XS) 8148785: Update class file version to 53 for JDK-9

[Alan Bateman]

On 03/02/2016 14:12, harold seigel wrote:

Hi Aleksey,

Thanks for the review.  The plan is to first change the runtime to 
accept version 53 files and then change the tools to generate them. 
Hopefully, this will reduce incompatibility problems.  See JDK-8148651 
 for details.
Right, there are several things that need to happen and I think we need 
hotspot accepting 53.0 before javac starts to generate them.


The webrevs look okay to me, except it might be better to drop "New 
Module attribute" from the comment in classfileParser.cpp until JSR 376 
is further along and we line up the trucks to bring the module system 
into JDK 9.


-Alan

Re: RFR(XS) 8148785: Update class file version to 53 for JDK-9

[Aleksey Shipilev]

On 02/03/2016 04:16 PM, harold seigel wrote:
> Open webrevs:
>   http://cr.openjdk.java.net/~hseigel/bug_8148785.jdk/
>   http://cr.openjdk.java.net/~hseigel/bug_8148785.hs/

+1

> JBS Bug: https://bugs.openjdk.java.net/browse/JDK-8148785

Description seems to imply we change the compiled class file versions to
53.0. But this change only makes VM to accept 53.0, it does not actually
produce 53.0 files yet. Are we sure the patches do what 8148785 intended?

Anyhow, there seems to be little risk with this particular change.

-Aleksey