Hello Sergey,
the change looks fine to me.
Thanks,
Andrew
On 8/13/2014 4:31 AM, Sergey Bylokhov wrote:
Hi Jim.
Yes, you are right, I missed it even after attentive viewing.
Typo was fixed:
http://cr.openjdk.java.net/~serb/8042199/webrev.03
Hi Sergey,
I understand that the type was changed
On 12.08.2014 1:34, Jim Graham wrote:
Hi Sergey,
Is the -g:none the result of #2 below?
This was changed to align with javac debug=flase... in build
xml(typo was fixed as well).
Also, if I read the email trail correctly then source/target=1.6 is
only there because JDK 9 compiler doesn't
The new Readme explanation looks good. Thanks for updating the new code
for pre-1.5.
I notice that one of the changes (in CMMTests) is to a line with a typo
(Platfrom instead of Platform both in the code and in the comment on the
same line), but fixing the typo might affect a lot of other
Hi, Jim.
Actually a Boolean was changed to a boolean, to skip autoboxing.
http://cr.openjdk.java.net/~serb/8042199/webrev.02/src/share/demo/java2d/J2DBench/src/j2dbench/tests/cmm/CMMTests.java.sdiff.html
The new Readme explanation looks good. Thanks for updating the new code
for pre-1.5.
I
Hi Sergey,
I understand that the type was changed for a reason, but the variable is
spelled Platfrom - which is not a word - and the same text appears in
the comment there.
The word intended there is, I believe, Platform...
...jim
On 8/12/14 4:20 PM, Sergey Bylokhov
Hi Jim.
Yes, you are right, I missed it even after attentive viewing.
Typo was fixed:
http://cr.openjdk.java.net/~serb/8042199/webrev.03
Hi Sergey,
I understand that the type was changed for a reason, but the variable is
spelled Platfrom - which is not a word - and the same text appears in
Thanks Sergey, looks good - approved.
It's interesting to note that you fixed the text that controls the
name of the command line option and option listed in results files, but
since the option is commented out anyway then I'm guessing that there
are no benchmark scripts that could have been
Hi Sergey,
Is the -g:none the result of #2 below?
Also, if I read the email trail correctly then source/target=1.6 is only
there because JDK 9 compiler doesn't let you request anything earlier.
The Readme doesn't mention this and it should.
Also, I'm not sure why it says that it requires at
Hello, Phil.
jdk 9 now supports -target 1.6+/-source 1.6+ options only. Looks like
we should use this minimum version too.
If you have no objections I'll prepare the new version of the fix
On 14.05.2014 21:09, Phil Race wrote:
Hmm .. the thing here is that I know that Jim was very keen that
Perhaps we have to make that the default but add a comment that since this
is bundled with JDK 9 it must use at least a 1.6 target but the intention is
that it be able to be compiled with and targeted to, earlier JDKs
BTW I guess dest-dist is OK but I imagine Jim really did mean dest
-# java
The only intention was that we be able to compare against older runtimes
when needed. We could ask whether we care about how we currently
compare against 1.2 any more...? If we're really that curious, we can
either change the target and compile with an older compiler, or find an
older
Hmm .. the thing here is that I know that Jim was very keen that this
benchmark always be runnable on JDK 1.2
Deleting the comment to that effect and setting source level to 1.5 goes
against this.
What is the rationale, and why can't it be reverted to be able to build
on 1.4 and run
on 1.2 ? If
12 matches
Mail list logo
