Quantcast

RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Chris Plummer
Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8176797
http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk

After fixing 8175342 (see the other RFR I just posted), this test
started to fail instead of assert. The problem is 256000 is too small of
a stack size on the platform in question. It needs to be at least 256k.

The test has a "stackSize" local variable that is the String "256". It
runs the jvm with -Xss of this size, appending "k" to it on one run and
"000" in the next. The -Xss256k version is fine. The -Xss256000 version
is not. I'm not sure why the test chose to specify the stackSize as a
String instead of an int. As part of this fix I changed it to an int,
and just multiply it by 1024 in the case where -Xss is specified with a
byte size rather the an k size. The test passes now on the platform
where it was failing.

thanks,

Chris

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

David Holmes
Hi Chris,

Looks good to me.

Thanks,
David

On 16/03/2017 2:59 PM, Chris Plummer wrote:

> Hello,
>
> Please review the following:
>
> https://bugs.openjdk.java.net/browse/JDK-8176797
> http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk
>
> After fixing 8175342 (see the other RFR I just posted), this test
> started to fail instead of assert. The problem is 256000 is too small of
> a stack size on the platform in question. It needs to be at least 256k.
>
> The test has a "stackSize" local variable that is the String "256". It
> runs the jvm with -Xss of this size, appending "k" to it on one run and
> "000" in the next. The -Xss256k version is fine. The -Xss256000 version
> is not. I'm not sure why the test chose to specify the stackSize as a
> String instead of an int. As part of this fix I changed it to an int,
> and just multiply it by 1024 in the case where -Xss is specified with a
> byte size rather the an k size. The test passes now on the platform
> where it was failing.
>
> thanks,
>
> Chris
>
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Christian Tornqvist
In reply to this post by Chris Plummer
Hi Chris,

This looks good, thanks for fixing this.

Thanks,
Christian

-----Original Message-----
From: hotspot-runtime-dev [mailto:[hidden email]] On Behalf Of Chris Plummer
Sent: Thursday, March 16, 2017 12:59 AM
To: hotspot-runtime-dev <[hidden email]>; [hidden email]
Subject: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Hello,

Please review the following:

https://bugs.openjdk.java.net/browse/JDK-8176797
http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk

After fixing 8175342 (see the other RFR I just posted), this test started to fail instead of assert. The problem is 256000 is too small of a stack size on the platform in question. It needs to be at least 256k.

The test has a "stackSize" local variable that is the String "256". It runs the jvm with -Xss of this size, appending "k" to it on one run and "000" in the next. The -Xss256k version is fine. The -Xss256000 version is not. I'm not sure why the test chose to specify the stackSize as a String instead of an int. As part of this fix I changed it to an int, and just multiply it by 1024 in the case where -Xss is specified with a byte size rather the an k size. The test passes now on the platform where it was failing.

thanks,

Chris


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Stanislav Smirnov
In reply to this post by Chris Plummer
Hi Chris,

changes look good, the only note, seems like formatting moved a bit
-                "-Xss" + stackSize + "000", "-XshowSettings", "-jar", testJar.getAbsolutePath());
+                    "-Xss" + (stackSize * 1024), "-XshowSettings", "-jar", testJar.getAbsolutePath());


Best regards,
Stanislav Smirnov





> On 16 Mar 2017, at 07:59, Chris Plummer <[hidden email]> wrote:
>
> Hello,
>
> Please review the following:
>
> https://bugs.openjdk.java.net/browse/JDK-8176797
> http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk
>
> After fixing 8175342 (see the other RFR I just posted), this test started to fail instead of assert. The problem is 256000 is too small of a stack size on the platform in question. It needs to be at least 256k.
>
> The test has a "stackSize" local variable that is the String "256". It runs the jvm with -Xss of this size, appending "k" to it on one run and "000" in the next. The -Xss256k version is fine. The -Xss256000 version is not. I'm not sure why the test chose to specify the stackSize as a String instead of an int. As part of this fix I changed it to an int, and just multiply it by 1024 in the case where -Xss is specified with a byte size rather the an k size. The test passes now on the platform where it was failing.
>
> thanks,
>
> Chris
>

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Kumar Srinivasan
In reply to this post by Chris Plummer
Hello,

I can't recall why  String vs int.
Formatting seems off,

   81         tr = doExec(javaCmd, "-Xms64m", "-Xmx512m",
   82                 "-Xss" + stackSize + "k", "-XshowSettings", "-jar", testJar.getAbsolutePath());
   83         containsAllOptions(tr);
   84         if (!tr.isOK()) {
   85             System.out.println(tr);
   86             throw new RuntimeException("test fails");
   87         }
   88         tr = doExec(javaCmd, "-Xms65536k", "-Xmx712m",
   89                     "-Xss" + (stackSize * 1024), "-XshowSettings", "-jar", testJar.getAbsolutePath());

Otherwise looks good to me.

Kumar


> Hello,
>
> Please review the following:
>
> https://bugs.openjdk.java.net/browse/JDK-8176797
> http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk
>
> After fixing 8175342 (see the other RFR I just posted), this test
> started to fail instead of assert. The problem is 256000 is too small
> of a stack size on the platform in question. It needs to be at least
> 256k.
>
> The test has a "stackSize" local variable that is the String "256". It
> runs the jvm with -Xss of this size, appending "k" to it on one run
> and "000" in the next. The -Xss256k version is fine. The -Xss256000
> version is not. I'm not sure why the test chose to specify the
> stackSize as a String instead of an int. As part of this fix I changed
> it to an int, and just multiply it by 1024 in the case where -Xss is
> specified with a byte size rather the an k size. The test passes now
> on the platform where it was failing.
>
> thanks,
>
> Chris
>

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: RRF(XS)(10): 8176797: [TESTBUG] tools/launcher/Settings.java -Xss size is too small

Chris Plummer
Hi Kumar,

I explained the formatting changein a previous email, but I see now the
formatting change I did on line 82 makes it inconsistent with similar
source on line 82, so I'll change it back.

thanks,

Chris

On 3/16/17 9:56 AM, Kumar Srinivasan wrote:

> Hello,
>
> I can't recall why  String vs int.
> Formatting seems off,
>
>   81         tr = doExec(javaCmd, "-Xms64m", "-Xmx512m",
>   82                 "-Xss" + stackSize + "k", "-XshowSettings",
> "-jar", testJar.getAbsolutePath());
>   83         containsAllOptions(tr);
>   84         if (!tr.isOK()) {
>   85             System.out.println(tr);
>   86             throw new RuntimeException("test fails");
>   87         }
>   88         tr = doExec(javaCmd, "-Xms65536k", "-Xmx712m",
>   89                     "-Xss" + (stackSize * 1024),
> "-XshowSettings", "-jar", testJar.getAbsolutePath());
>
> Otherwise looks good to me.
>
> Kumar
>
>
>> Hello,
>>
>> Please review the following:
>>
>> https://bugs.openjdk.java.net/browse/JDK-8176797
>> http://cr.openjdk.java.net/~cjplummer/8176797/webrev.00/webrev.jdk
>>
>> After fixing 8175342 (see the other RFR I just posted), this test
>> started to fail instead of assert. The problem is 256000 is too small
>> of a stack size on the platform in question. It needs to be at least
>> 256k.
>>
>> The test has a "stackSize" local variable that is the String "256".
>> It runs the jvm with -Xss of this size, appending "k" to it on one
>> run and "000" in the next. The -Xss256k version is fine. The
>> -Xss256000 version is not. I'm not sure why the test chose to specify
>> the stackSize as a String instead of an int. As part of this fix I
>> changed it to an int, and just multiply it by 1024 in the case where
>> -Xss is specified with a byte size rather the an k size. The test
>> passes now on the platform where it was failing.
>>
>> thanks,
>>
>> Chris
>>
>

Loading...