[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[beta] backports

- Emit specific message for `time<0.3.35` inference failure rust-lang#129343
- Use a reduced recursion limit in the MIR inliner's cycle breaker rust-lang#129714
- rustdoc: do not run doctests with invalid langstrings rust-lang#128838

r? cuviper

[do not merge] test rust-lang#129714 revert on bootstrap times

It seems we're currently experiencing some instability with the benchmarking results. rust-lang#129714 had weird results, somehow failing to measure bootstrap and other benchmarks. Since then bootstrap measurements [have also increased](https://perf.rust-lang.org/bootstrap.html). Let's see if a revert does anything to bootstrap times at the very least, to see if the PR is maybe involved.

r? ghost

[do not merge] test rust-lang#129714 revert on bootstrap times

It seems we're currently experiencing some instability with the benchmarking results. rust-lang#129714 had weird results, somehow failing to measure bootstrap and other benchmarks. Since then bootstrap measurements [have also increased](https://perf.rust-lang.org/bootstrap.html). Let's see if a revert does anything to bootstrap times at the very least, to see if the PR is maybe involved.

r? ghost

…etrochenkov

Make the inliner recursion depth hack stronger

We previously tried to introduce a more gentle version of this hack in github.com//pull/129714, but that wasn't enough because the problem was re-reported in #131960 (comment). So this increases the magic number from 2 to 8. Surely nobody will complain that they have a type whose depth grows more than 8 times faster than its call graph depth.

…elmann

add regression test

closes rust-lang#128887

the errors in that issue are due to two separate issues:
- MIR inlining causing the trait solver to hit the recursion limit (partially fixed in rust-lang#129714)
- using subtyping in method selection for paths (fixed in rust-lang#129073)

We moved any remaining issues due to MIR inlining into rust-lang#131960, but keeping rust-lang#128887 open as well seems unhelpful and confusing.

Rollup merge of #146918 - lcnr:add-regression-test, r=jdonszelmann

add regression test

closes #128887

the errors in that issue are due to two separate issues:
- MIR inlining causing the trait solver to hit the recursion limit (partially fixed in #129714)
- using subtyping in method selection for paths (fixed in #129073)

We moved any remaining issues due to MIR inlining into #131960, but keeping #128887 open as well seems unhelpful and confusing.