Thomas Schwinge wrote: > Ha, I, myself, am the GDB guru here ;-)! I had a look at the log again, > experimented some more, and finally got it going with the following > patch. However, I have absolutely no idea whether that is correct in all > cases, etc. Should perhaps target_wait (a.k.a. gnu-nat.c's gnu_wait) be > doing that?
Adding switch_to_thread at this location seems correct to me (even though it should be a no-op on most targets). Could you test your patch on a non-Hurd target to make sure, anyway? > + /* TODO. How to keep this synchronized with gnu-nat.c's own counting? */ Hmm. It would appear that "set exec-wrapper" is currently broken with the gnu-nat.c target, right? Thanks, Ulrich -- Dr. Ulrich Weigand GNU Toolchain for Linux on System z and Cell BE [EMAIL PROTECTED]