Removed faulty diagnostic on inlining over generated functions.

[orizi]

Stack:

• Updated sierra ap-change finding to be based on final lowering. #6258
• Removed faulty diagnostic on inlining over generated functions. #6257 ⬅

⚠️ Part of a stack created by spr. Do not merge manually using the UI - doing so may have unexpected results.

---

This change is 

[ilyalesokhin-starkware]

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

    let FunctionWithBodyLongId::Semantic(semantic_function_id) = function_id.lookup_intern(db)
    else {
        // Diagnostics are not relevant for generated functions.

why not?

Code quote:

    // Diagnostics are not relevant for generated functions.

[ilyalesokhin-starkware]

Reviewed all commit messages.
Reviewable status: 0 of 2 files reviewed, 2 unresolved discussions (waiting on @orizi)

---

crates/cairo-lang-sierra-generator/src/ap_change_test_data/tests line 85 at r1 (raw file):

fn cycle_if_some(x: Option<felt252>) {
    match x {
        Option::Some(x) => while x != 0 {},

why are you changing the test?

Code quote:

 Option::Some(x) => while x != 0 {},

[orizi]

Reviewable status: 0 of 2 files reviewed, 2 unresolved discussions (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

why not?

generated functions are not marked with inline(always) ever - as this is an attribute on an item.

---

crates/cairo-lang-sierra-generator/src/ap_change_test_data/tests line 85 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

why are you changing the test?

that was the actual way i wanted to write the test.

[ilyalesokhin-starkware]

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

Previously, orizi wrote…

generated functions are not marked with inline(always) ever - as this is an attribute on an item.

What is the "faulty" diagnostic?

Also, what happens if we add more inline diagnostic in the future and they are relevant for generated code?

[orizi]

Reviewable status: 0 of 2 files reviewed, 2 unresolved discussions (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

What is the "faulty" diagnostic?

Also, what happens if we add more inline diagnostic in the future and they are relevant for generated code?

You cannot write loops within a always inline func.

If we ever mark closures or loops with inline always we would require a fix.
But I don't really think we'd ever be doing that.

[ilyalesokhin-starkware]

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

Previously, orizi wrote…

You cannot write loops within a always inline func.

If we ever mark closures or loops with inline always we would require a fix.
But I don't really think we'd ever be doing that.

can you change it to

Code snippet:

let inline_config = match function_id.lookup_intern(db) {
     semantic =>  db.function_declaration_inline_config(semantic_function_id)?
     generated => InlineConfiguration::None 
}

[orizi]

Reviewable status: 0 of 2 files reviewed, 2 unresolved discussions (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-lowering/src/inline/mod.rs line 40 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

can you change it to

Done.

[ilyalesokhin-starkware]

Reviewed 1 of 2 files at r1, 1 of 1 files at r2.
Reviewable status: all files reviewed (commit messages unreviewed), all discussions resolved (waiting on @orizi)