<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hello Anthony,<br>
<br>
Could you please review the third version of the fix containing
modifications discussed with you in the previous letter.<br>
<br>
Webrev: <a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~alitvinov/8005607/webrev.02">http://cr.openjdk.java.net/~alitvinov/8005607/webrev.02</a><br>
<br>
This version of the fix differs from the previous in the following
places:<br>
<ol>
<li>A comment about the place of invocation of the method
"XErrorHandlerUtil.init" was added to a documentation block of
the method.</li>
<li>A code related to XShmAttach function common to the files
"src/solaris/native/sun/awt/awt_GraphicsEnv.c" and
"src/solaris/native/sun/java2d/x11/X11SurfaceData.c" was
extracted into a separate function "TryXShmAttach" declared in
"src/solaris/native/sun/awt/awt_GraphicsEnv.h" file.</li>
<li>All JNI code related to X error handling was implemented as
corresponding macros defined in
"src/solaris/native/sun/awt/awt_util.h" file.</li>
</ol>
Thank you,<br>
Anton<br>
<br>
<div class="moz-cite-prefix">On 1/31/2013 7:42 PM, Anton Litvinov
wrote:<br>
</div>
<blockquote cite="mid:510A90DF.2060501@oracle.com" type="cite">Hello
Anthony,
<br>
<br>
Thank you for the review and these remarks. Surely, the comment
will be added. I think that all JNI code related to XShmAttach can
be definitely transferred into a separate dedicated function,
which will be declared in
"src/solaris/native/sun/awt/awt_GraphicsEnv.h" file. I will try to
wrap all JNU calls connected with XErrorHandler into the
particular "WITH_XERROR_HANDLER", "RESTORE_XERROR_HANDLER"
functions or macros.
<br>
<br>
Thank you,
<br>
Anton
<br>
<br>
On 1/31/2013 4:57 PM, Anthony Petrov wrote:
<br>
<blockquote type="cite">Hi Anton,
<br>
<br>
A couple comments:
<br>
<br>
1. src/solaris/classes/sun/awt/X11/XErrorHandlerUtil.java
<br>
<blockquote type="cite"> 80 private static void init(long
display) {
<br>
</blockquote>
<br>
This method is private and isn't called from anywhere in this
class itself. This looks confusing. Please add a comment stating
that this method is invoked from native code, and from where
exactly.
<br>
<br>
<br>
2. Interesting that we use this machinery to call the
XShmAttach() from native code twice, and the code looks quite
similar in each case. Would it be possible to extract the common
code in a separate function (a-la BOOL TryXShmAttach(...)) to
avoid code replication? There are other usages as well, so we
could also introduce a macro (such as the old
EXEC_WITH_XERROR_HANDLER but now with other arguments) that
would minimize all the JNU_ calls required to use this
machinery.
<br>
<br>
<br>
Otherwise the fix looks great.
<br>
<br>
-- <br>
best regards,
<br>
Anthony
<br>
<br>
On 1/30/2013 20:14, Anton Litvinov wrote:
<br>
<blockquote type="cite"> Hello Anthony,
<br>
<br>
Could you, please, review a second version of the fix, which
is based on an idea of reusing the existing AWT native global
error handler from Java 2D native code.
<br>
<br>
Webrev:
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~alitvinov/8005607/webrev.01">http://cr.openjdk.java.net/~alitvinov/8005607/webrev.01</a>
<br>
<br>
The fix consists of the following parts:
<br>
<br>
1. Migration of all X error handling code from XToolkit to
a new
<br>
XErrorHandlerUtil class for resolution of
interdependency between
<br>
a static initialization block of XToolkit and a block
initializing
<br>
java.awt.GraphicsEnvironment singleton. Such dependency
is created
<br>
by new calls to XToolkit static methods from
<br>
"src/solaris/native/sun/awt/awt_GraphicsEnv.c",
<br>
"src/solaris/native/sun/java2d/x11/X11SurfaceData.c"
files.
<br>
2. Substitution of XToolkit.WITH_XERROR_HANDLER,
<br>
XToolkit.RESTORE_XERROR_HANDLER ... for corresponding
methods,
<br>
fields of XErrorHandlerUtil class in all places of JDK
source
<br>
code, where they were used.
<br>
3. Substitution of all found native X error handlers which
are set in
<br>
native code (awt_GraphicsEnv.c, X11SurfaceData.c,
<br>
GLXSurfaceData.c) for new synthetic Java error handlers.
<br>
4. Removal of X error handling code used by the native
error handlers
<br>
from "solaris/native/sun/awt/awt_util.c"
<br>
"solaris/native/sun/awt/awt_util.h" files.
<br>
<br>
Thank you,
<br>
Anton
<br>
<br>
On 1/11/2013 3:45 PM, Anthony Petrov wrote:
<br>
<blockquote type="cite">I'm not Jim, but as I indicated
earlier my opinion is that the easiest way to fix this is to
install the existing J2DXErrHandler() only once. That is, it
is the second option listed by you. Of course, the
J2DXErrHandler needs to be updated as well to detect whether
2D code wants to use it at the moment or it must simply
delegate to the previous handler (i.e. where the code
currently installs/uninstalls the handler, it must instead
set a global boolean flag or something.)
<br>
<br>
While the first option (reusing the existing AWT machinery)
is an interesting idea in general, I think it is complex and
would require too much additional testing and bring an
unjustified risk to the solution for such a basic problem.
<br>
<br>
-- <br>
best regards,
<br>
Anthony
<br>
<br>
On 1/11/2013 14:44, Anton Litvinov wrote:
<br>
<blockquote type="cite">Hello Jim,
<br>
<br>
Thank you very much for the review and provision of a new
idea of a solution. Elimination of the logic, which
sets/unsets J2DXErrHandler() for each call
"XShmAttach(awt_display, &shminfo))" should
effectively resolve the issue, but only in case if all
other native error handlers, which were set by the system
function "XSetErrorHandler()" in JDK or in any external
library, observe the rule of relaying of all events, which
are not relative to them, to the previously saved error
handlers. Otherwise an error generated during "XShmAttach"
function call will not be handled by J2DXErrHandler().
<br>
<br>
Could you answer the following question. By setting
J2DXErrHandler() only once and forever do you mean usage
of AWT global event handler "static int
ToolkitErrorHandler(Display * dpy, XErrorEvent * event)"
from "src/solaris/native/sun/xawt/XlibWrapper.c" with Java
synthetic handlers or creation of another global native
error handler with J2DXErrHandler as native synthetic
handler?
<br>
<br>
Thank you,
<br>
Anton
<br>
<br>
On 1/10/2013 5:44 AM, Jim Graham wrote:
<br>
<blockquote type="cite">I think I'd rather see some way to
prevent double-adding the handler in the first place as
well. Since it is only ever used on errors I also think
it is OK to set it once and leave it there forever...
<br>
<br>
...jim
<br>
<br>
On 1/9/13 8:08 AM, Anthony Petrov wrote:
<br>
<blockquote type="cite">Hi Anton et al.,
<br>
<br>
If I read the description of the bug correctly,
specifically this part:
<br>
<br>
<blockquote type="cite">The problem occurs, if another
thread (for example, GTK thread) is
<br>
doing the same sort of thing concurrently. This can
lead to the
<br>
following situation.
<br>
JVM thread: Sets J2DXErrHandler(), saves
ANY_PREVIOUS_HANDLER as
<br>
previous GTK thread: Sets some GTK_HANDLER,
saves
<br>
J2DXErrHandler() as previous JVM thread: Restores
<br>
ANY_PREVIOUS_HANDLER GTK thread: Restores
J2DXErrHandler() JVM
<br>
thread: Sets J2DXErrHandler(), saves
J2DXErrHandler() as previous
<br>
</blockquote>
<br>
It is obvious that at this final step 2D is in an
inconsistent state. We
<br>
don't expect to replace our own error handler (and it
shouldn't have
<br>
been there in the first place).
<br>
<br>
I realize that the fix you propose works around this
problem. But this
<br>
doesn't look like an ideal solution to me.
<br>
<br>
BTW, IIRC, in JDK7 (and 6?) we decided to set the
actual X11 error
<br>
handler only once and never replace it. All the rest
of the
<br>
push_handler/pop_handler logic is now located in Java
code (see
<br>
XToolkit.SAVED_ERROR_HANDLER() and the surrounding
logic). I believe
<br>
that we should somehow share this machinery with the
2D code to avoid
<br>
this sort of problems. Though I'm not sure if this
will eliminate this
<br>
exact issue.
<br>
<br>
<br>
2D/AWT folks: any other thoughts?
<br>
<br>
-- <br>
best regards,
<br>
Anthony
<br>
<br>
On 12/29/2012 17:44, Anton Litvinov wrote:
<br>
<blockquote type="cite">Hello,
<br>
<br>
Please review the following fix for a bug.
<br>
<br>
Bug:
<a class="moz-txt-link-freetext" href="http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=8005607">http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=8005607</a>
<br>
<a class="moz-txt-link-freetext" href="https://jbs.oracle.com/bugs/browse/JDK-8005607">https://jbs.oracle.com/bugs/browse/JDK-8005607</a>
<br>
Webrev:
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~alitvinov/8005607/webrev.00">http://cr.openjdk.java.net/~alitvinov/8005607/webrev.00</a>
<br>
<br>
The bug consists in a crash which is caused by a
stack overflow for
<br>
the reason of an infinite recursion in AWT native
function
<br>
J2DXErrHandler() under certain circumstances on
32-bit Linux OS. The
<br>
fix is based on introduction of the logic, which
detects indirect
<br>
recursive calls to J2DXErrHandler() by means of a
simple counter, to
<br>
J2DXErrHandler() native function. Such a solution
requires minimum
<br>
code changes, does not alter the handler's code
significantly and
<br>
eliminates this bug.
<br>
<br>
Adding <a class="moz-txt-link-abbreviated" href="mailto:2d-dev@openjdk.java.net">2d-dev@openjdk.java.net</a> e-mail alias to the
list of recipients
<br>
of this letter, because the edited function's name
is related to Java
<br>
2D area of JDK, despite of the fact that the edited
file is located in
<br>
AWT directory.
<br>
<br>
Thank you,
<br>
Anton
<br>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<br>
</body>
</html>