pthread_call_every and RISC OS 3.7

Peter Naulls peter at chocky.org
Fri Jul 29 04:14:52 PDT 2005



I'm seeing some odd behaviour with pthread_call_every under RISC OS 3.7.
In fact, it's in Firefox, and you can reproduce this by running the
binary through DDT under 3.7.

The exact error varies, but for example, I see an abort fetching the
upcall handler address.  a4 and all other registers contain perfectly
reasonable values.   Of course, this works just fine under RISC OS 4 and
above as evidenced by Firefox working on such systems.  The problem
occurs once the program becomes a wimp task.

-- 
Peter Naulls - peter at chocky.org        | http://www.chocky.org/
----------------------------------------------------------------------------
RISC OS C Programming                  | http://www.riscos.info/c/



More information about the gcc mailing list