Making changed to bytecodeInterpreter.cpp
Kumar Saurav
kumars at ncsu.edu
Mon Nov 5 11:52:28 PST 2012
The crash log is attached. I checked the memory using the command "top"
and it shows "1174016k" as free.
The "make hotspot" works if I just want to trap branches (i.e. changes
made to bytecodeInterpreter.cpp), however, I also want to trap messages
being printed to screen. I found that system.out.println("xyz")
ultimately ends up in a function writeBytes, in io_util.c (path
./openjdk-ecj/jdk/src/share/native/java/io/io_util.c). So, I want to
include a printf in this function to confirm that I can trap
system.out.println. Once I include this, I will have to issue make (I
could not find any other rule which would translate changes in io_util.c
to other dependant files).
Best Regards, Sincères Salutations, Mit Freundlichen Grüßen,
*
*Kumar Saurav*
*
On 11/05/2012 12:54 PM, Krystal Mok wrote:
> Hi Kumar,
>
> Could you post the crash log
> (/home/icedtea6-1.11.4/openjdk/langtools/make/hs_err_pid5811.log)
> somewhere to let people get more details?
>
> From the log in your mail it doesn't look like a JIT compiler issue.
> This list (hotspot-compiler-dev) is for discussion on HotSpot VM's JIT
> compilers. The problem you're hitting looks more like a zlib problem,
> or you might be really running low on memory.
> It'll take more information on the problem to be able to redirect you
> to a more appropriate list.
>
> BTW, when you run "make" in the main directory, it'll build the whole
> JDK, where as running "make hotspot" will only build the VM. If the
> latter is sufficient for you to carry on your experiment, you might
> just want to build the VM alone and save the trouble of dealing with
> this crash.
>
> - Kris
>
> On Tue, Nov 6, 2012 at 1:11 AM, Kumar Saurav <kumars at ncsu.edu
> <mailto:kumars at ncsu.edu>> wrote:
>
> Hi,
>
> I have installed icedtea6-1.11.4 and am trying to modify
> bytecodeInterpreter.cpp to trap branch conditions. Once I make
> changes to trap branch conditions and try to do "make hotspot" in
> the icedtea6-1.11.4 main directory, it works. But when I try using
> just "make" the build stops with the following error:
>
> -def-build-bootstrap-tool:
>
> build-bootstrap-javac:
> mmap failed for CEN and END part of zip file
> [jar] error while reading original manifest in file:
> /home/icedtea6-1.11.4/openjdk.build/langtools/build/bootstrap/lib/javac.jarnull
> #
> # There is insufficient memory for the Java Runtime Environment to
> continue.
> # Native memory allocation (malloc) failed to allocate 32756 bytes
> for ChunkPool::allocate
> # An error report file with more information is saved as:
> # /home/icedtea6-1.11.4/openjdk/langtools/make/hs_err_pid5811.log
> make[3]: *** [build] Error 1
> make[3]: Leaving directory
> `/home/icedtea6-1.11.4/openjdk/langtools/make'
> make[2]: *** [langtools-build] Error 2
> make[2]: Leaving directory `/home/icedtea6-1.11.4/openjdk'
> make[1]: *** [build_product_image] Error 2
> make[1]: Leaving directory `/home/icedtea6-1.11.4/openjdk'
> make: *** [stamps/icedtea.stamp] Error 2
>
> I read the log file and it mentioned that one of the things I
> should try is to reduce the Java heap size. So, I used the flag
> -Xmx256m in /acinclude.m4 but this didn't help.
>
> Can you suggest the build process to follow if we make changes to
> files like bytecodeInterpreter.cpp and bytecodeInterpreter.h?
> --
> Best Regards, Sincères Salutations, Mit Freundlichen Grüßen,
> *
>
> *Kumar Saurav*
>
> *
> *
>
> **
>
> **
>
> Graduate Student-Computer Engineering,
> ECE Department
>
>
> *
> **Mobile +1 919 931 7626
> North Carolina State University,
> Raleigh, NC, USA.
>
> /Privileged/Confidential Information may be contained in this
> message. If you are not an intended recipient, you are hereby
> notified that any disclosure, reproduction, distribution or other
> use of this communication and any attachments is strictly
> prohibited. If you have received this communication in error,
> please notify the sender by reply transmission and delete the
> message without copying and disclosing it./
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20121105/5036126c/attachment-0001.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hs_err_pid5811.log
Type: text/x-log
Size: 102540 bytes
Desc: not available
Url : http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/attachments/20121105/5036126c/hs_err_pid5811-0001.log
More information about the hotspot-compiler-dev
mailing list