fix(tvix/eval): instantiate *new* closures from blueprints each time r/4659

The previous closure refactoring introduced a bug in which the same
closure object would get mutated constantly for each instance of a
closure, which is incorrect behaviour.

This commit instead introduces an explicit new Value variant for the
internal "blueprint" that the compiler generates (essentially just the
lambda) and uses this variant to construct the closure at runtime.

If the blueprint ever leaks out to a user somehow that is a critical
bug and tvix-eval will panic.

As a ~treat~ test for this, the fibonacci function is being used as it
is a self-recursive closure (i.e. different instantiations of the same
"blueprint") getting called with different values and it's good to
have it around.

Change-Id: I485de675e9bb0c599ed7d5dc0f001eb34ab4c15f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/6323
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>

1 files changed, 2 insertions, 2 deletions

diff --git a/tvix/eval/src/value/function.rs b/tvix/eval/src/value/function.rs
index d0209cc50725..e5db43d58ace 100644
--- a/tvix/eval/src/value/function.rs
+++ b/tvix/eval/src/value/function.rs
@@ -29,7 +29,7 @@ impl Lambda {
 
 #[derive(Clone, Debug)]
 pub struct InnerClosure {
-    pub lambda: Lambda,
+    pub lambda: Rc<Lambda>,
     pub upvalues: Vec<Value>,
 }
 
@@ -38,7 +38,7 @@ pub struct InnerClosure {
 pub struct Closure(Rc<RefCell<InnerClosure>>);
 
 impl Closure {
-    pub fn new(lambda: Lambda) -> Self {
+    pub fn new(lambda: Rc<Lambda>) -> Self {
         Closure(Rc::new(RefCell::new(InnerClosure {
             upvalues: Vec::with_capacity(lambda.upvalue_count),
             lambda,