Re: [Valgrind-developers] valgrind: r5266 - in branches/FNWRAP: coregrind coregrind/m_debuginfo coregrind/m_scheduler include

[Josef W. <Jos...@gm...>]

On Friday 02 December 2005 15:28, sv...@va... wrote:
> Initial function wrapping hacks.  Note, a work in progress - expect
> borkage.  Works well enough to intercept pthread_mutex_{lock,unlock}.

Just to understand the idea: You have a per-thread flag to decide about
non-redirection. Ie. only one wrapper can be active at one point in time.
Is this enough for pthread? If one pthread function is calling another
one in its implementation, no wrapper will be called.

Is there any way to support longjmps in wrapped functions?

Josef