Hi Nicolas,
The fix looks good to me, too. I've published a webrev at:
http://cr.openjdk.java.net/~anthony/8-30-doLog-7163898.0/
and in fact I've just pushed it:
http://hg.openjdk.java.net/jdk8/awt/jdk/rev/9ed029a0326d
Thank you very much for the contribution!
--
best regards,
Anthony
On 4/
Hi,
I have reviewed this patch and it looks good to me.
Thanks,
/Staffan
On 23 apr 2012, at 08:06, Nicolas Carranza wrote:
> hi Anthony,
>
> The little patch against the current
> http://hg.openjdk.java.net/jdk8/jdk8/jdk repo is attached.
> Running:
> time -f '%P' java PerformanceRegressionTe
hi Anthony,
The little patch against the current
http://hg.openjdk.java.net/jdk8/jdk8/jdk repo is attached.
Running:
time -f '%P' java PerformanceRegressionTest2 useLogging
3 times using the (JRE 1.7.0_03-b04 + escapeDoLog.patch) gives the
following CPU usages: 14%, 13%, 13%.
Please let me kno
On 04/18/2012 05:08 AM, Anthony Petrov wrote:
> Hi Nicolas,
>
> Frederico's patch is needed - with that patch we avoid string
> concatenation operations in XAWT code if logging has been disabled.
> These operations caused heavy memory usage, and eliminating them
> resolved the issue reported by Fre
Here are the results of running (3 times):
time -f '%P' java PerformanceRegressionTest2 useLogging
on my laptop using different JREs:
JRE6: 13%, 13%, 12%
JRE7: 20%, 22%, 22%
JRE7*: 12%, 11%, 12%
JRE7_u4: 12%, 13%, 13%
JRE8: 13%, 14%, 14%
where:
JRE6:
java version "1.6.0_30"
Java(TM) SE Runtime
Hi Nicolas,
Frederico's patch is needed - with that patch we avoid string
concatenation operations in XAWT code if logging has been disabled.
These operations caused heavy memory usage, and eliminating them
resolved the issue reported by Frederico. We don't want to revert that
patch. In fact,
On 18/04/2012 02:01, Nicolas Carranza wrote:
Federico's patch solves the performance regression (see:
http://mail.openjdk.java.net/pipermail/jdk7u-dev/2012-April/002694.html
). But instead of this patch I find better to change the PlatformLogger
class: insert "if(!isLoggable(level)) return;" at
Hi Nicolas,
Federico's patch has also been applied to 7u4.
FYI bug number: 7104625
David
-
On 18/04/2012 11:01 AM, Nicolas Carranza wrote:
Federico's patch solves the performance regression (see:
http://mail.openjdk.java.net/pipermail/jdk7u-dev/2012-April/002694.html
). But instead of th
Thank you, here are my comments:
http://mail.openjdk.java.net/pipermail/jdk7u-dev/2012-April/002702.html .
Cheers,
Nicolas
On 04/17/2012 05:28 AM, Anthony Petrov wrote:
> Here's the relevant thread:
>
> http://mail.openjdk.java.net/pipermail/awt-dev/2011-October/001952.html
>
> It was concerning
Federico's patch solves the performance regression (see:
http://mail.openjdk.java.net/pipermail/jdk7u-dev/2012-April/002694.html
). But instead of this patch I find better to change the PlatformLogger
class: insert "if(!isLoggable(level)) return;" at the beginning of the
method sun.util.logging.Pl
On 17/04/2012 8:28 PM, Anthony Petrov wrote:
Here's the relevant thread:
http://mail.openjdk.java.net/pipermail/awt-dev/2011-October/001952.html
It was concerning building strings that are supposed to be logged while
the logging itself wasn't actually enabled.
The problem reported by Nicolas s
Here's the relevant thread:
http://mail.openjdk.java.net/pipermail/awt-dev/2011-October/001952.html
It was concerning building strings that are supposed to be logged while
the logging itself wasn't actually enabled.
The problem reported by Nicolas seems to be about logger initialization,
so
java.util.logging.LogManager isn't loaded when running the
PerformanceRegressionTest2 without using logging
(http://www.java.net/forum/topic/performance/general-performance-discussion/jre7s-javaawt-performance-degradation-caused-javautillogging-linux):
> java -verbose:class PerformanceRegressionTes
Nicolas,
Thanks for reporting the problem. I am curious if you have collected
any profiling data that shows
sun.util.logging.PlatformLogger.redirectPlatformLoggers() is the cause
of the performance degradation. This redirection is done as part of the
logging initialization before JFrame is
This sounds like it may be related to an issue that was reported a while
ago. I don't recall the exact details but there was some massive array
being allocated if I recall correctly. The fix was to add a guard to the
logging calls in the AWT code. But in this case if logging is being used
then
java.util.logging is currently handled by the Serviceability Team.
Adding that alias to this e-mail thread...
Dan
On 4/16/12 12:17 AM, Nicolas Carranza wrote:
Moving the mouse pointer over a java.awt.Window using JRE7 consumes
aprox. twice the CPU time as with JRE6 when logging is used (when a
Moving the mouse pointer over a java.awt.Window using JRE7 consumes
aprox. twice the CPU time as with JRE6 when logging is used (when a call
to java.util.logging.getLogger(String)---even without using the returned
Logger--- has been done). Keyboard key event processing suffers the same
performance
17 matches
Mail list logo
