Strange (out of memory?) crash about esp-hal

[t-moe]

Ok. Thanks.

Then I will close this, since enabling flip-link yields the behavior I expected and I have some other pressing things to take care of first before going down the rabbit hole here.

[bjoernQ]

I can only reproduce the crash in debug mode - in release it's fine. (and usually, release should be the preferred build profile)

Seeing the Guru Meditation usually happens if something crashes before esp-hal gets a chance to set the interrupt-vector. That is not the case here - probably the code just manages to even overwrite the interrupt vector

flip-link is not a default - with flip-link enabled I get

Stack overflow detected at 0x42035fc6 called by 0x42033e54
0x42035fc6 - core::intrinsics::is_aligned_and_not_null
    at C:\Users\Bjoern\.rustup\toolchains\nightly-x86_64-pc-windows-msvc\lib\rustlib\src\rust\library\core\src\intrinsics.rs:2635
0x42033e54 - core::slice::raw::from_raw_parts::runtime
    at C:\Users\Bjoern\.rustup\toolchains\nightly-x86_64-pc-windows-msvc\lib\rustlib\src\rust\library\core\src\slice\raw.rs:99

[t-moe]

Thanks for the fast reply @bjoernQ and for trying this out.

Do you think it is worth to additionally investigate why it still crashes? Or should we close this with "its up to the compiler to allocate structures as he wants" (and two structures of the same size, dont necessarily need to be allocated the same way) ?

[bjoernQ]

I guess analyzing it is a lot of effort (especially the generated assembly in debug mode is hard to read) and I'm not sure it will yield any useable insights.

If you still want to look into what is really happening and share your results - that's certainly welcome