panic with 32 bit signed int in 64b bitfield

[jerrinot]

Thanks for this cool macro!

This code is panicking with value out of bounds:

#[test]
fn negative_signed() {
    #[bitfield(u64)]
    struct MyBitfield {
        negative: i32,
        #[bits(32)]
        __: u32,
    }

    let _ = MyBitfield::new().with_negative(-1);
}

When I change it to

#[test]
fn negative_signed() {
    #[bitfield(u32)]
    struct MyBitfield {
        negative: i32,
    }

    let f = MyBitfield::new().with_negative(-1);
    assert_eq!(-1, f.negative())
}

then it's working fine.

[jerrinot]

I've been looking at this for a bit. It seems the root cause is in signed casting which extends the signed bit.

This demonstrates the idea: https://github.com/wrenger/bitfield-struct-rs/compare/main...jerrinot:bitfield-struct-rs:sign_shifting?expand=1#diff-b1a35a68f14e696205874893c07fd24fdb88882b47c23cc0e0c80a30c7d53759R733-R761

I am fairly sure there is a better fix, I have no idea what am I doing :)

[wrenger]

Oh yes, thanks for catching this. This happens when the fields have the same bit size as their signed type. I've redesigned the bounds check accordingly in 8de2078, which will land in 5.5.

[jerrinot]

many thanks for a super quick fix and release! I just included the latest version in my project and it works like a charm!