[do not merge] Use LLD ICF=safe for compiling rustc

[Kobzol]

Since #99062, which used icf=all, had to be reverted because of linker errors, I want to see if using icf=safe provides any cycle/binary size benefits.

r? @ghost

[Kobzol]

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

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

[bors]

⌛ Trying commit 889afe2 with merge 7d5533de50ff92ab3569671a9d08d30453c9ca2e...

[bors]

☀️ Try build successful - checks-actions
Build commit: 7d5533de50ff92ab3569671a9d08d30453c9ca2e (7d5533de50ff92ab3569671a9d08d30453c9ca2e)

[rust-timer]

Queued 7d5533de50ff92ab3569671a9d08d30453c9ca2e with parent 96c2df8, future comparison URL.

[rust-timer]

Finished benchmarking commit (7d5533de50ff92ab3569671a9d08d30453c9ca2e): comparison url.

Instruction count

• Primary benchmarks: 😿 relevant regressions found
• Secondary benchmarks: mixed results

	mean1	max	count2
Regressions 😿 (primary)	0.3%	0.3%	2
Regressions 😿 (secondary)	2.5%	2.5%	3
Improvements 🎉 (primary)	N/A	N/A	0
Improvements 🎉 (secondary)	-0.4%	-0.4%	1
All 😿🎉 (primary)	0.3%	0.3%	2

Max RSS (memory usage)

Results

• Primary benchmarks: mixed results
• Secondary benchmarks: 😿 relevant regressions found

	mean1	max	count2
Regressions 😿 (primary)	8.5%	8.5%	1
Regressions 😿 (secondary)	2.9%	5.4%	10
Improvements 🎉 (primary)	-1.8%	-2.3%	7
Improvements 🎉 (secondary)	N/A	N/A	0
All 😿🎉 (primary)	-0.5%	8.5%	8

Cycles

Results

• Primary benchmarks: 🎉 relevant improvements found
• Secondary benchmarks: 🎉 relevant improvements found

	mean1	max	count2
Regressions 😿 (primary)	N/A	N/A	0
Regressions 😿 (secondary)	N/A	N/A	0
Improvements 🎉 (primary)	-2.9%	-7.4%	18
Improvements 🎉 (secondary)	-3.5%	-6.1%	44
All 😿🎉 (primary)	-2.9%	-7.4%	18

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

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf +perf-regression

Footnotes

1. the arithmetic mean of the percent change ↩ ↩² ↩³

2. number of relevant changes ↩ ↩² ↩³

[Kobzol]

Interesting! icf=safe doesn't really improve binary size, but the cycle wins are still there. I guess that most of the cycle wins actually come from lld itself, as they were also present when only lld without ICF was used. @matthiaskrgr could you please try if rustc 7d5533de50ff92ab3569671a9d08d30453c9ca2e works for you?

[matthiaskrgr]

./toolchains/7d5533de50ff92ab3569671a9d08d30453c9ca2e/bin/rustc: symbol lookup error: ./toolchains/7d5533de50ff92ab3569671a9d08d30453c9ca2e/bin/rustc: undefined symbol: _ZN3std2rt19lang_start_internal17hc4ce26e998d6538dE

:(

[Kobzol]

@matthiaskrgr Could you please also try 986e50fc1e83d611bf485cf02723d7bd89a093b8? That one is lld only, without ICF.

[matthiaskrgr]

same

/home/matthias/.rustup/toolchains/986e50fc1e83d611bf485cf02723d7bd89a093b8/bin/rustc: symbol lookup error: /home/matthias/.rustup/toolchains/986e50fc1e83d611bf485cf02723d7bd89a093b8/bin/../lib/librustc_driver-8fd17062f1c0f9c7.so: undefined symbol: _ZN9hashbrown3raw11Fallibility17capacity_overflow17hb2863c36cff2e6c5E

[Kobzol]

Then it seems like a lld issue. @lqd Have you met similar lld errors while compiling Rust programs with lld? Maybe the problem is that we use the "stock" lld instead of the rust-lld? I'm not sure how (if) they differ though.

[Mark-Simulacrum]

Closing, since we figured out what was wrong here. (LD_LIBRARY_PATH set in environment).