1.92 regression in PyO3: unable to determine layout for ... because ... cannot be normalized

[theemathas]

[INFO] [stdout] error[E0080]: unable to determine layout for `<<T as PyClassImpl>::BaseType as PyClassBaseType>::LayoutAsBase` because `<<T as PyClassImpl>::BaseType as PyClassBaseType>::LayoutAsBase` cannot be normalized
[INFO] [stdout]  --> /rustc/f3f12444a017add0468f683f3a10656d29438a05/library/core/src/mem/mod.rs:1247:26
[INFO] [stdout]   |
[INFO] [stdout]   = note: evaluation of `<pycell::impl_::PyClassObject<T> as std::mem::SizedTypeProperties>::ALIGN` failed here
[INFO] [stdout] 
[INFO] [stdout] 
[INFO] [stdout] For more information about this error, try `rustc --explain E0080`.

• https://crater-reports.s3.amazonaws.com/beta-1.92-4/beta-2025-11-10/gh/umar052001.blitzparse/log.txt
• https://crater-reports.s3.amazonaws.com/beta-1.92-4/beta-2025-11-10/gh/Jayson-Fong.python-libbruteforce/log.txt

[madsmtm]

This only shows up with cargo build (cargo check doesn't seem to be enough).

I bisected https://github.com/umar052001/blitzparse:

searched nightlies: from nightly-2025-10-23 to nightly-2025-10-24
regressed nightly: nightly-2025-10-24
searched commit range: 1d23d06...6501e64
regressed commit: 4b3ba58

bisected with cargo-bisect-rustc v0.6.11

Host triple: aarch64-apple-darwin
Reproduce with:

cargo bisect-rustc --start 2025-09-01 --end 2025-11-19 --prompt

Which points to #147793

CC @cjgillot @scottmcm @oli-obk

[oli-obk]

Left a comment on the PR that regressed things, should be a trivial fix, but I don't know when I'd get to it, so if someone else can push it over the finish line that would be great

[madsmtm]

FYI, reducing a bit, it can be hit with just:

[package]
name = "foo"
version = "0.1.0"
edition = "2021"

[dependencies]
pyo3 = "0.26.0"

[profile.dev.package."*"]
opt-level = 3

[apiraino]

Assigning priority (discussion on Zulip).

@rustbot label -I-prioritize +P-high