FWD » Core Development » Runtime Infrastructure

12/08/2022

04:23 PM Bug #5699: broken logfile name on Windows appserver clients (and probably other cases on Windows)

Is @string@ not a pointer to @WCHAR@? If the contents of @string@ are not wide characters, then that explains the re... Greg Shah

11:42 AM Bug #5699 (WIP): broken logfile name on Windows appserver clients (and probably other cases on Wi...

I was able to reproduce it. The problem is in @replace_str@ function. I've added to the code a few comments to show v... Galya B

01:55 PM Feature #6939: improve proxy performance by hard-coding method invocation

In the lambda approach, the target instance is already captured, so changing the array also changes the target. Greg Shah

01:43 PM Feature #6939: improve proxy performance by hard-coding method invocation

Greg Shah wrote:
> If we need the ability to temporarily replace the target methods for an implementation,
I don't ... Constantin Asofiei

01:29 PM Feature #6939: improve proxy performance by hard-coding method invocation

If we need the ability to temporarily replace the target methods for an implementation, we should consider a kind of ... Greg Shah

01:10 PM Feature #6939: improve proxy performance by hard-coding method invocation

Constantin Asofiei wrote:
> ... directly from the bytecode of the i.e. DMO Implementation class setter/getter.
wh... Constantin Asofiei

01:07 PM Feature #6939: improve proxy performance by hard-coding method invocation

Eric, the main issue of the interceptor flavor of the @InvocationHandler@ is that in some cases the target instance o... Constantin Asofiei

12/07/2022

07:07 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

From Galya, via email:
> I haven't looked at the Java classes instantiated by the spawner, so let me ask if there ... Greg Shah

12/05/2022

09:29 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

Greg Shah wrote:
> The approach seems reasonable.
>
> Eugenie: Do you have any thoughts, questions or concerns?
... Eugenie Lyzenko

09:11 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

The approach seems reasonable.
Eugenie: Do you have any thoughts, questions or concerns? Greg Shah

06:18 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

The Windows spawner is a lot more simple than the UNIX one in the process creation part (more complicated in the secu... Galya B

04:41 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

Greg Shah wrote:
> The test code doesn't help me understand the process tree/pid list/redirected output log name for... Galya B

12/02/2022

11:17 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

The test code doesn't help me understand the process tree/pid list/redirected output log name for the FWD client. Ca... Greg Shah

12/01/2022

10:49 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

I created a simple test to verify @execvp@ behaviour and uploaded it as zip. The test function in @C@ prints the pid ... Galya B

09:12 AM Bug #5701 (WIP): logfile names for -O outputToFile use the spawner pid instead of the Java (FWD c...

Constantin Asofiei wrote:
> The spawner will use @NativeSecureConnection.command@ to determine the launch command fo... Galya B

07:08 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

Galya Bogdanova wrote:
> ... running in background where no client driver seems to be instantiated in the Java code,... Constantin Asofiei

06:51 AM Bug #5701: logfile names for -O outputToFile use the spawner pid instead of the Java (FWD client ...

This is how I read the logic for the spawner (I'll be going with the Unix example):
@ClientBuilder#localStart()@ b... Galya B

11/17/2022

07:41 AM Feature #6939: improve proxy performance by hard-coding method invocation

Eric Faulhaber wrote:
> But, recently we discovered that @ReflectASM@ actually can perform worse than reflection on ... Constantin Asofiei

11/16/2022

03:28 PM Feature #6939: improve proxy performance by hard-coding method invocation

When I originally implemented the @com.goldencode.proxy@ package, the pressing goal at the time was to improve memory... Eric Faulhaber

02:48 PM Feature #6939 (New): improve proxy performance by hard-coding method invocation

Eric Faulhaber

11/09/2022

06:58 AM Feature #6692: move FWD to Java 17

I've tried to move a bit and have done initial migration to gradle 7.5.1 and tried to build FWD with Java 17.
Just ... Tomasz Domin