[Bug 394] Y2038 time overflow/pre Epoch bug : java.lang.ExceptionInInitializerError on systems with uninitialized clock

[bugzilla-daemon at icedtea.classpath.org]

http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=394


gnu_andrew at member.fsf.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Platform|all                         |32-bit




------- Comment #10 from gnu_andrew at member.fsf.org  2010-06-24 18:58 -------
The Y2038 bug is an issue with 32-bit systems only as time_t = pointer size. 
So it affects x86 and arm, but not x86_64 where time_t will be 64-bit.

src/os/linux/vm/os_linux.cpp, src/os/solaris/vm/os_solaris.cpp,
src/os/linux/vm/hpi_linux.hpp and src/os/linux/vm/hpi_linux.hpp all call
gettimeofday which will return a 32-bit seconds since the epoch value on 32-bit
systems.

See /usr/include/bits/typesizes.h:

#define __TIME_T_TYPE           __SLONGWORD_TYPE

from /usr/include/bits/types.h:

#define __SLONGWORD_TYPE        long int

A C long is 32-bit on 32-bit systems, and 64-bit on 64-bit systems.

Do we know if the glibc hackers have any ideas about this yet?  That seems a
better place for a fix.


-- 
Configure bugmail: http://icedtea.classpath.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.