Let CTFE to handle partially uninitialized unions without marking the entire value as uninitialized.

[oli-obk]

follow up to #94411

To fix #69488 and by extension fix #94371, we should stop treating types like MaybeUninit<usize> as something that the Scalar type in the interpreter engine can represent. So we add a new field to abi::Primitive that records whether the primitive is nested in a union

cc @RalfJung

r? @ghost

[oli-obk]

@bors r+ @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[bors]

📌 Commit 66a1d3d02ea0c3473163301fcd1d31b2fbd0fba6 has been approved by oli-obk

[oli-obk]

@bors r- oops

[oli-obk]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[bors]

⌛ Trying commit efa7539866423e0fd1ca420730ece30204dd8b74 with merge d8d9e85406dc30f48aa6825ae48eb8f7c366efea...

[oli-obk]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[RalfJung]

Bors didn't seem to react?
@bors try

[bors]

⌛ Trying commit 837ac8d340aa2ce177277ddd8e0c1e2f7ffa1f5c with merge 124a62c1537d6f500769f854b7e6d193479561a9...

[RalfJung]

So from what I can tell this is entirely unconnected to the logic in #93670? We might not want to emit noundef everywhere quite yet as this PR would, but not relating these two at all also seems odd.

[oli-obk]

First I want to see if it is feasible to go this route, then we can talk to the llvm people. I do believe we should be able to use it at some point, but it may very well be badly handled by llvm for now.

The linked PR works on types, where this PR encodes the information in the layout. So there may be unintended consequences. For now we should probably only use it in miri.

Maybe we should also flip the logic of the Boolean and search for a better name (!noundef checks are the worst, triple negation)

[bors]

☀️ Try build successful - checks-actions
Build commit: 124a62c1537d6f500769f854b7e6d193479561a9 (124a62c1537d6f500769f854b7e6d193479561a9)

[oli-obk]

@bors r=nagisa,erikdesjardin

[bors]

📌 Commit d57b755 has been approved by nagisa,erikdesjardin

[bors]

⌛ Testing commit d57b755 with merge f262ca1...

[bors]

☀️ Test successful - checks-actions
Approved by: nagisa,erikdesjardin
Pushing f262ca1 to master...

[rust-timer]

Finished benchmarking commit (f262ca1): comparison url.

Summary:

• Primary benchmarks: changes not relevant. 3 results were found to be statistically significant but the changes were too small to be relevant.
• Secondary benchmarks: mixed results

	Regressions 😿 (primary)	Regressions 😿 (secondary)	Improvements 🎉 (primary)	Improvements 🎉 (secondary)	All 😿 🎉 (primary)
count1	1	6	2	6	3
mean2	0.5%	0.8%	-0.4%	-1.1%	-0.1%
max	0.5%	1.1%	-0.4%	-1.3%	0.5%

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

Next Steps: If you can justify the regressions found in this perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please open an issue or create a new PR that fixes the regressions, add a comment linking to the newly created issue or PR, and then add the perf-regression-triaged label to this PR.

@rustbot label: +perf-regression

Footnotes

1. number of relevant changes ↩

2. the arithmetic mean of the percent change ↩