[prev in list] [next in list] [prev in thread] [next in thread]
List: openjdk-serviceability-dev
Subject: Re: [15] RFR(XS): 8247730: 2 JNI exception pending defect groups in DwarfParser.cpp
From: "serguei.spitsyn () oracle ! com" <serguei ! spitsyn () oracle ! com>
Date: 2020-06-23 23:16:43
Message-ID: eeb08b95-b418-7645-21c1-b2644f1bfb47 () oracle ! com
[Download RAW message or body]
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">On 6/23/20 16:00, Chris Plummer wrote:<br>
</div>
<blockquote type="cite"
cite="mid:f2ecda3a-3ddb-7bc6-d79e-3ec53b5bc821@oracle.com">
<div class="moz-cite-prefix">On 6/23/20 3:01 PM, <a
class="moz-txt-link-abbreviated"
href="mailto:serguei.spitsyn@oracle.com"
moz-do-not-send="true">serguei.spitsyn@oracle.com</a> wrote:<br>
</div>
<blockquote type="cite"
cite="mid:bbf2347b-f382-5690-624c-74e63748ef45@oracle.com">
<div class="moz-cite-prefix">On 6/23/20 14:57, <a
class="moz-txt-link-abbreviated"
href="mailto:serguei.spitsyn@oracle.com"
moz-do-not-send="true">serguei.spitsyn@oracle.com</a> wrote:<br>
</div>
<blockquote type="cite"
cite="mid:35066aff-525e-09cd-fb91-37d37f3ac909@oracle.com">
<div class="moz-cite-prefix">Hi Chris,<br>
<br>
<pre><span class="new">+ if (!env->ExceptionOccurred()) {</span>
env->ThrowNew(ex_cls, "DWARF not found");
<span class="new">+ }
</span></pre>
Should the <span class="new">!env->ExceptionOccurred()
be replaced with </span><span class="new">env->ExceptionOccurred()</span>
<span class="new">?<br>
</span></div>
</blockquote>
<br>
Sorry. It is correct as it is is.<br>
But then the question is what happen in the case of <span
class="new">env->ExceptionOccurred() ?<br>
</span></blockquote>
We end up throwing whatever exception FindClass left pending
rather than throwing the DebuggerException. It's preferable that
DebuggerException be thrown, but in reality the FindClass
shouldn't ever fail. If it does, then something else is very
broken or we are out of memory.<br>
</blockquote>
<br>
I agree, it is unlikely the FindClass will ever fail in this
context. <br>
Otherwise, something is something very broken and can fail in many
different ways. <br>
<br>
<blockquote type="cite"
cite="mid:f2ecda3a-3ddb-7bc6-d79e-3ec53b5bc821@oracle.com"> <br>
The way around this issue is to do the FindClass in
Java_sun_jvm_hotspot_debugger_linux_amd64_DwarfParser_init0 and
save it static variable. I could take that approach if you wish.<br>
</blockquote>
<br>
As I understand, this is fix is to please the Parfait.<br>
I'm okay with the fix as it is.<br>
<br>
Thanks,<br>
Serguei<br>
<br>
<br>
<blockquote type="cite"
cite="mid:f2ecda3a-3ddb-7bc6-d79e-3ec53b5bc821@oracle.com"> <br>
thanks,<br>
<br>
Chris<br>
<blockquote type="cite"
cite="mid:bbf2347b-f382-5690-624c-74e63748ef45@oracle.com"><span
class="new"> <br>
Thanks,<br>
Serguei</span><br>
<span class="new"></span><br>
<span class="new"></span>
<blockquote type="cite"
cite="mid:35066aff-525e-09cd-fb91-37d37f3ac909@oracle.com">
<div class="moz-cite-prefix"><span class="new"> Thanks,<br>
Serguei<br>
</span> <br>
<br>
On 6/23/20 11:38, Chris Plummer wrote:<br>
</div>
<blockquote type="cite"
cite="mid:6be65951-87d0-c85a-4ba6-9154c7188be3@oracle.com">Hello,
<br>
<br>
Please review the following: <br>
<br>
<a class="moz-txt-link-freetext"
href="https://bugs.openjdk.java.net/browse/JDK-8247730"
moz-do-not-send="true">https://bugs.openjdk.java.net/browse/JDK-8247730</a>
<br>
<a class="moz-txt-link-freetext"
href="http://cr.openjdk.java.net/~cjplummer/8247730/webrev.00/"
moz-do-not-send="true">http://cr.openjdk.java.net/~cjplummer/8247730/webrev.00/</a>
<br>
<br>
There are two locations were we make a JNI call with the
possibility of a pending exception. This is new code in JDK
15, so it is being addressed there. <br>
<br>
thanks, <br>
<br>
Chris <br>
</blockquote>
<br>
</blockquote>
<br>
</blockquote>
<br>
</blockquote>
<br>
</body>
</html>
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic