[M3devel] Problems with setjmp/longjmp exception-handling and optimization

[Tony Hosking]

Just to summarise the status of optimization and setjmp/longjmp  
exception handling.  Currently, optimizing breaks for setjmp/longjmp  
exceptions because register locals that are live across the setjmp  
call will revert to their value at the time of the setjmp instead of  
having the value at the time of the longjmp.  The fix is to force the  
compiler front-end to allocate variables assigned inside TRY blocks to  
memory, so that those updates are not lost.  This is only necessary  
for the setjmp/longjmp scheme.  I am working on this right now.

Antony Hosking | Associate Professor | Computer Science | Purdue  
University
305 N. University Street | West Lafayette | IN 47907 | USA
Office +1 765 494 6001 | Mobile +1 765 427 5484



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m3lists.elegosoft.com/pipermail/m3devel/attachments/20090512/67822c07/attachment-0001.html>