Please review correction of regression test Test6929067

Dr Andrew John Hughes gnu_andrew at
Tue Nov 16 14:55:28 PST 2010

On 16 November 2010 11:28, Pavel Tisnovsky <ptisnovs at> wrote:
> David Holmes wrote:
>> Hi Pavel,
>> Pavel Tisnovsky said the following on 11/15/10 23:36:
>>> can anybody please review correction of regression test
>>> runtime/6929067/Test6929067?
>>> I changed this test in a way that it now correctly works on x86_64
>>> platform, for example. Tested on RHEL 5 x86_64 but it should works on
>>> another systems too.
>> This is a fix for 64-bit systems but the invoke program is still being
>> compiled as 32-bit. I'm a little surprised it can then load the 64-bit VM.
> Hi David,
> it seems that "invoke" is compiled as 64-bit executable (on RHEL 5
> x86_64 at least):
> invoke: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), for
> GNU/Linux 2.6.9, dynamically linked (uses shared libs), for GNU/Linux
> 2.6.9, not stripped
>> David
>>> Here is webrev created against jdk7/hotspot-rt/hotspot:
>>> (this webrev is the same as webrev created against jdk7/tl/
>>> Thank you in advance
>>> Pavel

I think David may be referring to the need for -m32 or -m64, where gcc
is configured.  On GNU/Linux platforms, gcc tends to produce a 64-bit
binary on x86_64 by default but on Solaris, it produces a 32-bit

andrew at rohan:~$ gcc -g -o test test.c
andrew at rohan:~$ file test
test:		ELF 32-bit LSB executable 80386 Version 1 [FPU], dynamically
linked, not stripped
andrew at rohan:~$ gcc -g -o test -m64 test.c
andrew at rohan:~$ file test
test:		ELF 64-bit LSB executable AMD64 Version 1, dynamically linked,
not stripped

My presumption is the test is supposed to work on both.
Andrew :-)

Free Java Software Engineer
Red Hat, Inc. (

Support Free Java!
Contribute to GNU Classpath and the OpenJDK

PGP Key: 94EFD9D8 (
Fingerprint: F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8

More information about the hotspot-runtime-dev mailing list