Ruby testsuite failures because of pointer mangling on 32-bit ARM?

[Tom Tromey]

>>>>> "Carlos" == Carlos O'Donell <carlos@redhat.com> writes:

Carlos> I've had one report that the pointer mangling support on 32-bit ARM
Carlos> that was added in October might be causing the Ruby [1] testsuite
Carlos> to fail.

Note that if a port does pointer mangling in jmpbuf then it is handy for
gdb if the port also has the appropriate sdt.h probe in its longjmp
implementation.  That way gdb can find the longjmp target without
needing to copy glibc's pointer demangling.  There's a test case in the
gdb test suite that covers this.

Tom