Add CAIRO_DEBUG_SIERRA_GEN env var..

[ilyalesokhin-starkware]

No description provided.

[reviewable-StarkWare]

This change is 

[orizi]

Reviewable status: 0 of 1 files reviewed, 2 unresolved discussions

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

                if std::env::var("CAIRO_DEBUG_SIERRA_GEN").is_ok() {
                    // If we are debugging sierra generation, we wan't to finish the compilation
                    // rather than panic.

the ap-overflow case isn't an issue for deciding on inlining?

Code quote:

                    // If we are debugging sierra generation, we wan't to finish the compilation
                    // rather than panic.

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

                if std::env::var("CAIRO_DEBUG_SIERRA_GEN").is_ok() {
                    // If we are debugging sierra generation, we wan't to finish the compilation
                    // rather than panic.

Suggestion:

                    // If we are debugging sierra generation, we want to finish the compilation
                    // rather than panic.

[ilyalesokhin-starkware]

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, orizi wrote…

the ap-overflow case isn't an issue for deciding on inlining?

I think that if it happens here, it will happen in the final program. I just that that it was common enough that we should provide users the ability to debug it and get the code location from cairo-execute

[ilyalesokhin-starkware]

Reviewable status: 0 of 1 files reviewed, 2 unresolved discussions (waiting on @orizi)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

                if std::env::var("CAIRO_DEBUG_SIERRA_GEN").is_ok() {
                    // If we are debugging sierra generation, we wan't to finish the compilation
                    // rather than panic.

Done.

[orizi]

Reviewable status: 0 of 1 files reviewed, 2 unresolved discussions (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

I think that if it happens here, it will happen in the final program. I just that that it was common enough that we should provide users the ability to debug it and get the code location from cairo-execute

i'm actually not sure that is true for size estimation of const-folding:

if const_evaluated_as_true() {
   short()
} else {
   long()
}

[ilyalesokhin-starkware]

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, orizi wrote…

i'm actually not sure that is true for size estimation of const-folding:

if const_evaluated_as_true() {
   short()
} else {
   long()
}

I don't follow.

[ilyalesokhin-starkware]

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

I don't follow.

You mean that the compilation of long would fail, but the compiled program will only include short?

[orizi]

Reviewed all commit messages.
Reviewable status: 0 of 1 files reviewed, 1 unresolved discussion (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

You mean that the compilation of long would fail, but the compiled program will only include short?

yes

[ilyalesokhin-starkware]

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, orizi wrote…

yes

I can restore it, but I guess this is true for any misscompilation that doesn't get into the final code.

Do we want people to report such cases or not?

[orizi]

Reviewable status: 0 of 1 files reviewed, 1 unresolved discussion (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

I can restore it, but I guess this is true for any misscompilation that doesn't get into the final code.

Do we want people to report such cases or not?

ap-change specifically does sound like a special case for me.

[ilyalesokhin-starkware]

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, orizi wrote…

ap-change specifically does sound like a special case for me.

Why?

is the case of huge code that becomes small because of const-folding an interesting one?

[orizi]

Reviewable status: 0 of 1 files reviewed, 1 unresolved discussion (waiting on @ilyalesokhin-starkware)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, ilyalesokhin-starkware wrote…

Why?

is the case of huge code that becomes small because of const-folding an interesting one?

i don't know - but it is a case that may happen now, and would start crashing after this change.

[ilyalesokhin-starkware]

Reviewable status: 0 of 1 files reviewed, 1 unresolved discussion (waiting on @orizi)

---

crates/cairo-lang-compiler/src/db.rs line 45 at r1 (raw file):

Previously, orizi wrote…

i don't know - but it is a case that may happen now, and would start crashing after this change.

Done.

[orizi]

Reviewed 1 of 1 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @ilyalesokhin-starkware)