[Gc] Strange Signal Bug on Linux

Lothar Scholz scholz at scriptolutions.com
Sun Jul 30 10:35:44 PDT 2006


Hello Hans,

Sunday, July 30, 2006, 7:12:47 PM, you wrote:

BH> The collector uses SIGPWR and SIGXCPU on Linux in order to stop and
BH> restart threads.  (The original reason for that choice was that they
BH> seemed to be the least frequently used signals that existed everywhere.
BH> At this point, the main reason for it is probably backward
BH> compatibility.)

Thanks, the eiffel compiler indeed generated code that set a handler
for SIGXCPU. Removing it everything works fine.

I really whish that things like this could be documented in one common
place.

BH> I assume you meant 7.0 CVS?  (This behavior has actually been stable for
BH> a very long time.)

No i'm still using a version from april this year, this was 6.7


-- 
Best regards,
 Lothar Scholz                mailto:scholz at scriptolutions.com



More information about the Gc mailing list