[MLton-devel] mlton problem (fwd)

Stephen Weeks MLton@mlton.org
Tue, 13 Aug 2002 22:30:16 -0700


> We got on one of our codes the following with mlton 20020728 + the
> patches explained in the last email:
> 
>           useless starting
>             useless raised in 2.90 + 1.07 (27% GC)
>          closureConvertSimplify raised in 15.77 + 2.34 (13% GC)
>       pre codegen raised in 55.75 + 19.73 (26% GC)
>    Compile SML raised in 55.75 + 19.73 (26% GC)
> MLton raised in 55.75 + 19.73 (26% GC)
> mlton: x_0 has no analyze var value property                

This bug probably indicates that some pass has produced code that
violates the SSA dominance condition.  Once I know which pass, I will
take a closer look.

> It is in fact triggered by the -exn-history true flag.

I just ran all the regressions and a self-compile with 20020714 and
-exn-history true, and everything worked fine.  So I don't yet have a
test program that I can use to debug with.

To help isolate the bug, please compile the program -type-check true.
That will show which pass produced the invalid IL program.  And of
course, if you can create any program fragment that causes the problem
and that you are willing to send (either to MLton, or to me in
private), then that would help immensely.



-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel