Re: [Valgrind-developers] ir_opt.c tweaks

[Florian K. <fl...@ei...>]

On 20.06.25 22:31, Florian Krohm wrote:
> 
> There is no bug that I could find and I don't have a list either. But it would 
> be worthwhile to fill in the gaps. Constant folding is a cheap and easy thing to 
> do and effective. Algebraic identities are a different story as they need to 
> recurse the subtrees (sameIRexprs).
> 
> Anyhow, I'll bite and make a pass and see what is missing for operations of 
> integer type except Ity_I128. 

I opened https://bugs.kde.org/show_bug.cgi?id=506211 to track this issue.