On Friday, October 28, 2011 00:39:41 you wrote:
> However, in rare cases, there is a problem with arrays and their
> modifiers. The option requires CIL to generate temporary variables
> with more complex types than previously, and the resultant code is
> not always compilable. Consider the follow
On Sunday, October 30, 2011 16:52:04 Mihai T. Lazarescu wrote:
> I use --dosimplify in an application that traces memory
> accesses. For this application, as for most if not all CIL
> usage, it is important to have an error-free transformation
> for the broadest class of sources.
>
> I would sug
On Sun, Oct 30, 2011 at 07:14:26PM +0400, Pavel Shved wrote:
> > > 2) could this patch be added to upstream nevertheless? I think
> > > that the benefits to the verification tools, for which the
> > > options is useful, are much greater than the very rarely
> > > encountered issue with "very vola
Thank you for the comments, I'll take them into account and form a new
patch.
On Friday, October 28, 2011 16:55:56 you wrote:
> Why does --no-convert-field-offsets imply --no-split-structs?
> Shouldn't the user be to enable them separately?
>
> > @@ -89,6 +89,12 @@ let simplAddrOf = ref true
>
