[Bug gdb/28471] Internal error on Assertion `pid != 0' on AIX

["kadler at us dot ibm.com" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=28471

--- Comment #2 from Kevin Adler <kadler at us dot ibm.com> ---
Thanks, I did make some changes last night that were along the lines you talked
about.

Firstly, functions take in a ptid_t parameter now and don't rely on
inferior_ptid.

- get_signaled_thread now returns a thread_info* and the iterate_over_threads
call from pd_update was moved there. This was only needed when pd_activate(1)
was called in pd_enable, which now calls get_signaled_thread. The
switch_to_thread call that was formerly in pd_update is now there too, since
that was the only place it was used.

- pd_update now ends after sync_threadlists.

- pd_activate no longer calls pd_update and takes no args. Callers now call
pd_update() themselves.


This seems to have fixed the issue for me. No clue whether the whole
sync_threadlists is still needed or the larger question around user-space
thread support in general.

I'll look at the other changes you mentioned: removing the scope_restorer, the
code that sets inferior_ptid, etc.

-- 
You are receiving this mail because:
You are on the CC list for the bug.