refactor: replace rustix::process dependency with rlimit

[xtqqczze]

Closes #14914.
Reverts #14664.

[sylvestre]

sorry but i am not going to revert anything until we have a CI running for redox

[oech3]

I think we don't need to revert this since we were not running test(impossible?) for Redox on the CI previously.

[xtqqczze]

I think we should add a check-only job at the very least:

cargo check --all-targets --features feat_os_unix_redox --target x86_64-unknown-redox

We can set continue-on-error: true for now so that any existing errors don’t block CI, while still giving us visibility into Redox build issues.

[sylvestre]

Sounds good

[oech3]

Do we really need to hurry supporting test on redox? Most logic are shared with other platforms.

[codspeed]

Merging this PR will improve performance by 5.1%

⚠️ 115 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• sort_mixed_utf8_locale
• cksum_default
• dd_copy_64k_blocks
• factor_64bit_semiprime
• factor_multiple_u64s[2]
• sort_unique_utf8_locale
• cksum_multiple_files
• cksum_crc32b
• du_deep_tree[(100, 3)]
• dd_copy_partial
• …and 105 more

ℹ️ 1 benchmark has no flame graph

The flamegraph could not be generated because too much time was spent in runtime internals, making the visualization not useful for performance analysis.

• sort_dictionary_order[500000]

⚡ 3 improved benchmarks
✅ 398 untouched benchmarks
⏩ 70 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	factor_thirteen_39_bit_primes	9.5 s	8.8 s	+7.76%
⚡	Simulation	factor_three_39_bit_primes	567 ms	544.1 ms	+4.2%
⚡	Simulation	factor_five_38_bit_primes	1.6 s	1.6 s	+3.4%

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing xtqqczze:GH14664 (2e3fcdd) with main (7654e67)}

Footnotes

1. 70 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[oech3]

If we replaced rustix with rlimit, we need large diff again to remove it after rustix supported rlimit on redox.

[xtqqczze]

Split the uutests changes to #14252.

[xtqqczze]

If we replaced rustix with rlimit, we need large diff again to remove it after rustix supported rlimit on redox.

rustix doesn't seem to be actively maintained. Perhaps we should consider forking it.

[oech3]

Is there any collaborator for rustix repo in ? We should consider asking maintainer to add collaborator from bytecodealliance.

[oech3]

Since their famous project (e.g. wasmtime) depends on rustix, I think they cannot abandon it.