Re: [CVS ci] more PIR codegen improvements

2004-03-22 Thread Dan Sugalski
At 7:11 PM +0100 3/22/04, Leopold Toetsch wrote: Sterling Hughes <[EMAIL PROTECTED]> wrote: On Mar 22, 2004, at 11:42 AM, Leopold Toetsch wrote: * end is now inserted in @MAIN if missing perhaps this should be a compiler warning instead? If both C<@MAIN> and C are missing, we could compile a

Re: [CVS ci] more PIR codegen improvements

2004-03-22 Thread Leopold Toetsch
Sterling Hughes <[EMAIL PROTECTED]> wrote: > On Mar 22, 2004, at 11:42 AM, Leopold Toetsch wrote: >> * end is now inserted in @MAIN if missing > perhaps this should be a compiler warning instead? If both C<@MAIN> and C are missing, we could compile a module. So sometimes you'll get a warning an

Re: [CVS ci] more PIR codegen improvements

2004-03-22 Thread Sterling Hughes
On Mar 22, 2004, at 11:42 AM, Leopold Toetsch wrote: * end is now inserted in @MAIN if missing perhaps this should be a compiler warning instead? -sterling

[CVS ci] more PIR codegen improvements

2004-03-22 Thread Leopold Toetsch
* end is now inserted in @MAIN if missing * in @MAIN P1 isn't moved around * -Oc: replace savetop with up to 3 pushtopX ops It's a good idea to declare the main function: .sub _main @MAIN ... .end leo