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: [libmudflap] switching to splay trees


Hi -

> >By the way, someone naughtily broke mudflap support for C++ in the last
> >few days (mudflap_finish gets a NULL dereference during its building its
> >synthetic constructor function).
>
> That might have been me.  (I was not testing on a system where mudflap 
> wanted to build, but I can do that.)  [...]

The problem shows up even without a functional libmudflap -
cc1plus -fmudflap <anything.ii> would likely show the problem.

In gcc/cp/cp-mudflap.c's mflang_flush_calls(), there is no longer
a call to start_function().  This appears to make the subsequent four
references to current_function_decl SEGV with a null pointer dereference.
Commenting them out makes the whole test suite successful, but I'd like
to be sure that this was intentional.

- FChE

Attachment: pgp00000.pgp
Description: PGP signature


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