That's actually a nice question!
In your example, the compiler is not able to compute the change of ap, and here's why "x" is alternating between 1 and 2

You can see [a similar question](https://www.newton.so/view/63372c85f20834f8447ac533) there it is explained a bit more clearly explained what is a revoked reference.

## About revoked references
The scope of some types of references in Cairo (e.g., return values of functions and temporary variables) is somewhat restricted.
For example, a call to another function may revoke those references.

When you write "let" you do not create a variable, instead you create a reference to the memory slot.
I may even cause an error, for example:

```cairo
 let (x) = foo(1);
 let (y) = foo(2);
 serialize_word(x + y);
```
will cause an error:

```cairo
Error: code:43:20: Reference 'x' was revoked.
    serialize_word(x + y);
                   ^
Reference was defined here:
code:41:10
    let (x) = foo(1);
         ^
```


I am currently running this code

```
func main() {

    let x = [ap];
    [ap] = 1, ap++;

    [ap] = 2, ap++;

    [ap] = x, ap++;
    jmp rel -1;  // Jump to the previous instruction.
}

And what is happening is that "x" is alternating between 1 and 2 and I couldn't figure out why.

Also, it wasn't very clear to me what a "Revoked Reference" means. Is it when you cannot access a certain [ap] when running a specific function?

This is the link to the exercise:
https://www.cairo-lang.org/docs/how_cairo_works/consts.html#revoked-references

Revoked References exercise from Cairo-lang

Algorithms are just different and you check unit overflow on different stages of the check.

In solidity, you have a mistake `pow(1, 3, *) == 0`:


```solidity
let zx := mul(z, x) // x is zero; zx is also zero
...
z := div(zxRound, base) //zxRound = 0 + half; base = half // 2, means z = 0

```

in the second example straight after the first iteration

```cairo
let (n_LteOne) = uint256_le(n, Uint256(low=1, high=0));
        
        if(n_LteOne == TRUE) {
            return (res=z);
        }

```

returns `res=1`

I'm trying to make a FixedPoint cairo lib, but I'm having some trouble making a pow function. 

It's inspired by this function

```solidity

    function pow(
        uint256 x,
        uint256 y,
        uint256 base
    ) internal pure returns (uint256 z) {
        assembly {
            switch x
            case 0 {
                switch y
                case 0 {
                    // 0 ** 0 = 1
                    z := base
                }
                default {
                    // 0 ** y = 0
                    z := 0
                }
            }
            default {
                switch mod(y, 2)
                case 0 {
                    // If y is even, store base in z for now.
                    z := base
                }
                default {
                    // If y is odd, store x in z for now.
                    z := x
                }

                // Shifting right by 1 is like dividing by 2.
                let half := shr(1, base)

                for {
                    // Shift y right by 1 before looping to halve it.
                    y := shr(1, y)
                } y {
                    // Shift y right by 1 each iteration to halve it.
                    y := shr(1, y)
                } {
                    // Revert immediately if x ** 2 would overflow.
                    // Equivalent to iszero(eq(div(xSquared, x), x)) here.
                    if shr(128, x) {
                        revert(0, 0)
                    }

                    // Store x squared.
                    let xSquared := mul(x, x)

                    // Round to the nearest number.
                    let xSquaredRound := add(xSquared, half)

                    // Revert if xSquared + half overflowed.
                    if lt(xSquaredRound, xSquared) {
                        revert(0, 0)
                    }

                    // Set x to scaled xSquaredRound.
                    x := div(xSquaredRound, base)

                    // If n is even:
                    if mod(y, 2) {
                        // Compute z * x.
                        let zx := mul(z, x)

                        // If z * x overflowed:
                        if iszero(eq(div(zx, x), z)) {
                            // Revert if x is non-zero.
                            if iszero(iszero(x)) {
                                revert(0, 0)
                            }
                        }

                        // Round to the nearest number.
                        let zxRound := add(zx, half)

                        // Revert if zx + half overflowed.
                        if lt(zxRound, zx) {
                            revert(0, 0)
                        }

                        // Return properly scaled zxRound.
                        z := div(zxRound, base)
                    }
                }
            }
        }
    }
```

and I wrote this one in Cairo

```cairo
    func pow{syscall_ptr: felt*, pedersen_ptr: HashBuiltin*, range_check_ptr: felt}(
        x: Uint256,
        n: Uint256,
        baseUnit: Uint256
    ) -> (res: Uint256) {
        alloc_locals;

        let (xIsZero) = uint256_eq(x, Uint256(low=0, high=0));

        if(xIsZero == TRUE) {
            let (nIsZero) = uint256_eq(n, Uint256(low=0, high=0));
            if(nIsZero == TRUE) {
                // 0**0 = 1
                return (res=Uint256(low=1, high=0));
            } 
            // 0**n = 0
            return (res=Uint256(low=0, high=0));            
        }

        let (nHalf, nHalfRem) = uint256_unsigned_div_rem(n, Uint256(low=2, high=0));
        let (nIsEven) = uint256_eq(nHalfRem, Uint256(low=0, high=0));
        let (z) = _assignZIf(nIsEven, baseUnit, x);

        let (baseUnitHalf, baseUnitHalfRem) = uint256_unsigned_div_rem(baseUnit, Uint256(low=2, high=0));

        let (res) = pow_loop(x, nHalf, z, baseUnit, baseUnitHalf);

        return (res=res);
    }

    func pow_loop{syscall_ptr: felt*, pedersen_ptr: HashBuiltin*, range_check_ptr: felt}(
        x: Uint256, 
        n: Uint256, 
        z: Uint256,
        baseUnit: Uint256,
        baseUnitHalf: Uint256
    ) -> (res: Uint256) {
        alloc_locals;
    
        let (n_LteOne) = uint256_le(n, Uint256(low=1, high=0));
        
        if(n_LteOne == TRUE) {
            return (res=z);
        }

        let (xx, xxOverflow) = uint256_mul(x, x);
        assert xxOverflow.low = 0;
        assert xxOverflow.high = 0;

        let (xxRound, xSquaredRoundCarry) = uint256_add(xx, baseUnitHalf);
        let (roundingOverflowed) = uint256_lt(xxRound, xx);
        assert roundingOverflowed = FALSE;

        let (temp_x, temp_x_rem) = uint256_unsigned_div_rem(xxRound, baseUnit);

        let (nHalf, nHalfRem) = uint256_unsigned_div_rem(n, Uint256(low=2, high=0));
        let (nHalfIsEven) = uint256_eq(nHalfRem, Uint256(low=0, high=0));
        if(nHalfIsEven ==  TRUE) {
            let (zx, zxOverflow) = uint256_mul(z, x);
            assert zxOverflow.low = 0;
            assert zxOverflow.high = 0;

            let (shouldBeZ, shouldBeZRem) = uint256_unsigned_div_rem(zx, x);
            let (zxNotOverflowed) = uint256_eq(shouldBeZ, z);
            assert zxNotOverflowed = TRUE;

            let (zxRound, zxRoundCarry) = uint256_add(zx, baseUnitHalf);
            let (zxRoundOverflowed) = uint256_lt(zxRound, zx);
            assert zxRoundOverflowed = FALSE;

            let (temp_z, temp_zRem) = uint256_unsigned_div_rem(zxRound, baseUnit);
            return pow_loop(x=temp_x, n=nHalf, z=temp_z, baseUnit=baseUnit, baseUnitHalf=baseUnitHalf);
        } else {
            return pow_loop(x=temp_x, n=nHalf, z=z, baseUnit=baseUnit, baseUnitHalf=baseUnitHalf);
        }
    }

    func _assignZIf{syscall_ptr: felt*, pedersen_ptr: HashBuiltin*, range_check_ptr: felt}(
        nIsEven: felt,
        baseUnit: Uint256,
        x: Uint256
    ) -> (res: Uint256) {
        if(nIsEven == TRUE) {
            return (res=baseUnit);
        }
        return (res=x);
    }
```

The thing is that I don't get the same results.

Here are some example of my results compared to the solidity function

- pow(1, 3, 10^18): Cairo = 1 / Solidity = 0

- pow(10^18, 5, 10^18): Cairo = 4000000000000000000 (4x10^18) / Solidity = 32000000000000000000 (32x10^18)

- pow(10^19, 5, 10^18): Cairo = 100000000000000000000 (10^20) / Solidity = 100000000000000000000000 (10^23)

I can't see what I'm doing wrong in my code

Fixed Point pow operation error

answering to this message [https://www.newton.so/view/636d020159c30b8efc8d1d86#answer-636e24358470e1a247695014](https://www.newton.so/view/636d020159c30b8efc8d1d86#answer-636e24358470e1a247695014)

starkgate is not supporting Goerli-2 yet, well it supports Goerli-1, but they are not same

1. Can anyone explain how to get Goerli ETH tokens?

2. Also want to know how to bridge tokens from the Goerli ETH testnet and move them to Goerli2 (StarkNet)?


---

> thank's for your answers, summing up how to transfer funds and how to use Faucet on Ethereum:


1. For receiving ETH on Goerli Test Net, I used [Goerli PoW Faucet](https://goerli-faucet.pk910.de/).
As an alternative, [https://goerlifaucet.com/](https://goerlifaucet.com/) can be used, but it is required to authorize.

2. I installed ArgentX using the links below and transferred funds using the [HashStack service for StarkNet](https://goerli2-bridge.hashstack.finance)

![https://www.bbntimes.com/images/Ethereum_Faucet.jpeg](https://www.bbntimes.com/images/Ethereum_Faucet.jpeg)

StarkNet Faucet: How can I get a Goerli-2 eth faucet tokens (StarkNet)?

Hi! I like writing my Cairo code in vim. However, I don't like that Cairo lang code has no auto-format and syntax highlighting (see screenshot):

![https://sun9-86.userapi.com/impg/oNPtr4oks9yNwg3MDoX8c5b-B4IBgK1DARpTbg/RJgXBHR3mEY.jpg?size=1242x682&quality=96&sign=3390ee79c30971ba0dc245122af04905&type=album](https://sun9-86.userapi.com/impg/oNPtr4oks9yNwg3MDoX8c5b-B4IBgK1DARpTbg/RJgXBHR3mEY.jpg?size=1242x682&quality=96&sign=3390ee79c30971ba0dc245122af04905&type=album)

I'm looking for a plugin to write Cairo contracts on Starknet in vim with all pretty formatting. Can you recommend any and how to install it as well?



Does anyone know a vim plugin for StarkNet smart contracts?

Hey, I’m learning Cairo and I want to know more about Gas Optimization. In Solidity compiler, there is a difference between writing uint128 and uint256. Similarly to C and other languages, for example:

```solidity
contract ThisIsNotAnOptimizedContract{
	uint128 Zero;
	uint256 One;
	uint128 Two;
}

contract ThisIsAGasOptimizedContract{
	uint128 Zero;
	uint128 Two;
	uint256 One;
}

```

`uint128` variables each occupy a separate `uint256`just for themselves, wasting 128 bits for each slot (Like the first contract). The Solidity compiler puts 2 `uint128` in the one 256-bit memory slot when they are declared next to each other. 

I’m curious if there is any similar issue in Cairo. `Uint256` uses two felt objects for low/high bits of the value.

```python
struct Uint256 {
    // The low 128 bits of the value.
    low: felt,
    // The high 128 bits of the value.
    high: felt,
}
```

Is there any similar practice in Cairo to optimize `felt` and Uint256 objects to save some compiler steps?

How do you optimize gas in Cairo with Uint256/felt?

Can someone explain the difference `between unsigned_div_rem` and `uint256_unsigned_div_rem`?

Why unit version process so many extra steps, but overall number of steps for computing are different for these methods?



Can someone explain the difference between unsigned_div_rem and uint256_unsigned_div_rem?

Hey, I'm curious if uint256 variables are safe to use? They are using 2x128 bits, while a native felt uses only 252 bits. 

Can anyone explain?

Popular tags

Profile

vargastartup

Wallet address: 0x67608C8c5557aD66e2742ed2e8E14125FB83B47e

Recent questions

Recent answers