This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Patch: Add raw address to java stack trace if line number unavailable.


>>>>> "David" == David Daney <ddaney@avtrex.com> writes:

David> This patch adds a new system property
David> 'gnu.gcj.runtime.NameFinder.show_raw' that when set true will
David> print the IP in the stack trace if the line number is
David> unavailable.

This looks reasonable to me.

David> Back in 3.3 and 3.4 the addresses used to print out, but that
David> capability was lost with the new stack trace infrastructure.  So in
David> someways you could look at this as fixing a regression.

Cute.  I don't agree with that theory but...

David> Currently regression testing on i686-pc-linux-gnu.
David> OK to commit if no regressions?

Yes, please.

David> +          t->append (Long::toHexString (offset + 1));

Other places in the code explicitly cast offset to a jlong.
I suspect you need a cast here as well.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]