[MLton-devel] experimental release of source-level profiling

Matthew Fluet fluet@CS.Cornell.EDU
Mon, 20 Jan 2003 16:49:10 -0500 (EST)


> > GCStateVolatile was supposed to ensure that it was written to it's memory
> > location every time it was modified.  Do you have a small example that
> > exhibits the bug?  Or any outstanding changes that affect it?
>
> I checked in the change to make gcState.currentSource volatile.
> Running a MLton with that change, compiling any example with
> 	-profile time -profile-stack true -profile-basis true
> should show the problem.  I'm testing it on wc-scanStream.  Here is
> the problematic Machine code and the assembly produced by the native
> codegen.  Note that the assembly stores $0x13E into %edi, but does not
> write it to memory.  So, the call to GC_profileEnter sees the wrong
> currentSource.

O.k.  I don't have time to look at it today, but I see the problem and
will think about a solution.



-------------------------------------------------------
This SF.NET email is sponsored by: FREE  SSL Guide from Thawte
are you planning your Web Server Security? Click here to get a FREE
Thawte SSL guide and find the answers to all your  SSL security issues.
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel