|Dataflow and function calls firstname.lastname@example.org (Yuri Gribov) (2009-07-23)|
|Re: Dataflow and function calls email@example.com (gopi) (2009-07-28)|
|Re: Dataflow and function calls firstname.lastname@example.org (Yuri Gribov) (2009-07-29)|
|Re: Dataflow and function calls email@example.com (kamal) (2009-08-01)|
|From:||Yuri Gribov <firstname.lastname@example.org>|
|Date:||Wed, 29 Jul 2009 18:17:45 +0400|
|Posted-Date:||29 Jul 2009 22:59:35 EDT|
Thank you for your answer but please note that I was not asking about
My problem is not finding out what variables may be changed by call
(or by some ambigous command) but how to incorporate this information
into dataflow framework. We have 2 main opportunities: consider that
call kills reaching definitions or that it does not. Both lead to
problems during optimization (see my first message).
Up to now I have found only one online course which mentions this
problem - Rajiv Gupta's Compiler Construction (see slide 7 in
http://www.cs.ucr.edu/~gupta/teaching/201-09/My3.pdf ). Unfortunately
it does not give any details.
Return to the
Search the comp.compilers archives again.