rust-lang / rust-lang/rust

Performance analysis on `aws-sdk-ec2`

Open
#127,634 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

C-optimization S-experimental
Dominant language
Rust
Stars
119k
Forks
16.1k
PR metrics pending

Description

I'm doing some performance analysis on the (long) compilation of the crate aws-sdk-ec2, and using this issue to take notes about it.

rustc +nightly --version: rustc 1.81.0-nightly (0c81f94b9 2024-07-10)

cargo +nightly rustc -- -Ztime-passes:

time:   0.000; rss:   47MB ->   48MB (   +0MB)	parse_crate
time:   0.000; rss:   49MB ->   49MB (   +1MB)	setup_global_ctxt
time:   0.000; rss:   51MB ->   51MB (   +0MB)	crate_injection
time:   1.680; rss:   51MB -> 1103MB (+1052MB)	expand_crate
time:   1.680; rss:   51MB -> 1103MB (+1052MB)	macro_expand_crate
time:   0.000; rss: 1103MB -> 1104MB (   +0MB)	maybe_building_test_harness
time:   0.052; rss: 1104MB -> 1103MB (   +0MB)	AST_validation
time:   0.007; rss: 1103MB -> 1106MB (   +3MB)	finalize_imports
time:   0.033; rss: 1106MB -> 1106MB (   +0MB)	compute_effective_visibilities
time:   0.034; rss: 1106MB -> 1108MB (   +2MB)	finalize_macro_resolutions
time:   0.969; rss: 1108MB -> 1366MB ( +257MB)	late_resolve_crate
time:   0.048; rss: 1366MB -> 1397MB (  +32MB)	resolve_check_unused
time:   0.064; rss: 1397MB -> 1397MB (   +0MB)	resolve_postprocess
time:   1.157; rss: 1103MB -> 1397MB ( +294MB)	resolve_crate
time:   0.043; rss: 1398MB -> 1398MB (   +0MB)	write_dep_info
time:   0.044; rss: 1398MB -> 1399MB (   +1MB)	complete_gated_feature_checking
time:   0.103; rss: 2151MB -> 2145MB (   -6MB)	drop_ast
time:   3.029; rss: 1398MB -> 1991MB ( +593MB)	looking_for_derive_registrar
time:   3.577; rss: 1398MB -> 2034MB ( +636MB)	misc_checking_1
time:   9.258; rss: 2034MB -> 2694MB ( +660MB)	coherence_checking
time:  18.136; rss: 2034MB -> 3353MB (+1319MB)	type_check_crate
time:  14.660; rss: 3353MB -> 4533MB (+1181MB)	MIR_borrow_checking
time:   4.442; rss: 4533MB -> 4807MB ( +273MB)	MIR_effect_checking
time:   0.421; rss: 4843MB -> 4845MB (   +2MB)	module_lints
time:   0.421; rss: 4843MB -> 4845MB (   +2MB)	lint_checking
time:   0.453; rss: 4845MB -> 4845MB (   +0MB)	privacy_checking_modules
time:   0.135; rss: 4845MB -> 4849MB (   +4MB)	check_lint_expectations
time:   1.725; rss: 4811MB -> 4849MB (  +38MB)	misc_checking_3
time:   0.044; rss: 5120MB -> 5120MB (   +0MB)	monomorphization_collector_root_collections
time:   6.002; rss: 5120MB -> 5595MB ( +476MB)	monomorphization_collector_graph_walk
time:   2.506; rss: 5589MB -> 5702MB ( +113MB)	partition_and_assert_distinct_symbols
time:  10.513; rss: 4849MB -> 5546MB ( +697MB)	generate_crate_metadata
time:  28.231; rss: 5554MB -> 6196MB ( +642MB)	codegen_to_LLVM_IR
time:  28.440; rss: 5546MB -> 6196MB ( +649MB)	codegen_crate
time:  26.305; rss: 5947MB -> 6209MB ( +261MB)	LLVM_passes
time:   0.865; rss: 6196MB -> 6257MB (  +61MB)	encode_query_results
time:   0.967; rss: 6196MB -> 6167MB (  -28MB)	incr_comp_serialize_result_cache
time:   0.967; rss: 6196MB -> 6167MB (  -28MB)	incr_comp_persist_result_cache
time:   0.968; rss: 6196MB -> 6167MB (  -28MB)	serialize_dep_graph
time:   0.298; rss: 6167MB -> 4211MB (-1957MB)	free_global_ctxt
time:   0.520; rss: 3789MB -> 3893MB ( +104MB)	link_rlib
time:   0.527; rss: 3789MB -> 3893MB ( +104MB)	link_binary
time:   0.544; rss: 3789MB -> 3484MB ( -305MB)	link_crate
time:   0.546; rss: 4211MB -> 3484MB ( -727MB)	link
time:  86.838; rss:   32MB ->  245MB ( +213MB)	total
Optimized version, for reference

cargo +nightly rustc -- -C opt-level=3 -Ztime-passes:

time:   0.000; rss:   47MB ->   47MB (   +0MB)	parse_crate
time:   0.000; rss:   48MB ->   49MB (   +1MB)	setup_global_ctxt
time:   0.000; rss:   51MB ->   51MB (   +0MB)	crate_injection
time:   1.788; rss:   51MB -> 1104MB (+1052MB)	expand_crate
time:   1.788; rss:   51MB -> 1104MB (+1053MB)	macro_expand_crate
time:   0.000; rss: 1104MB -> 1104MB (   +0MB)	maybe_building_test_harness
time:   0.048; rss: 1104MB -> 1104MB (   +0MB)	AST_validation
time:   0.006; rss: 1104MB -> 1106MB (   +2MB)	finalize_imports
time:   0.027; rss: 1106MB -> 1106MB (   +0MB)	compute_effective_visibilities
time:   0.034; rss: 1106MB -> 1108MB (   +2MB)	finalize_macro_resolutions
time:   0.975; rss: 1108MB -> 1366MB ( +258MB)	late_resolve_crate
time:   0.041; rss: 1366MB -> 1398MB (  +32MB)	resolve_check_unused
time:   0.059; rss: 1398MB -> 1398MB (   +0MB)	resolve_postprocess
time:   1.142; rss: 1104MB -> 1398MB ( +294MB)	resolve_crate
time:   0.032; rss: 1400MB -> 1399MB (   -1MB)	write_dep_info
time:   0.030; rss: 1399MB -> 1399MB (   +1MB)	complete_gated_feature_checking
time:   0.103; rss: 2153MB -> 2148MB (   -6MB)	drop_ast
time:   3.136; rss: 1399MB -> 1994MB ( +595MB)	looking_for_derive_registrar
time:   3.674; rss: 1399MB -> 2036MB ( +638MB)	misc_checking_1
time:   9.398; rss: 2036MB -> 2696MB ( +660MB)	coherence_checking
time:  18.398; rss: 2036MB -> 3354MB (+1318MB)	type_check_crate
time:  14.451; rss: 3354MB -> 4534MB (+1180MB)	MIR_borrow_checking
time:   4.738; rss: 4534MB -> 4803MB ( +269MB)	MIR_effect_checking
time:   0.421; rss: 4835MB -> 4840MB (   +4MB)	module_lints
time:   0.421; rss: 4835MB -> 4840MB (   +4MB)	lint_checking
time:   0.454; rss: 4840MB -> 4840MB (   +0MB)	privacy_checking_modules
time:   0.136; rss: 4840MB -> 4844MB (   +4MB)	check_lint_expectations
time:   1.728; rss: 4807MB -> 4844MB (  +37MB)	misc_checking_3
time:   2.511; rss: 4844MB -> 5017MB ( +173MB)	generate_crate_metadata
time:   0.049; rss: 5017MB -> 5017MB (   +0MB)	monomorphization_collector_root_collections
time:   8.685; rss: 5017MB -> 5424MB ( +406MB)	monomorphization_collector_graph_walk
time:   4.531; rss: 5417MB -> 5539MB ( +122MB)	partition_and_assert_distinct_symbols
time:  53.323; rss: 5467MB -> 7019MB (+1553MB)	codegen_to_LLVM_IR
time:  66.963; rss: 5017MB -> 7019MB (+2002MB)	codegen_crate
time:   1.174; rss: 7019MB -> 7138MB ( +119MB)	encode_query_results
time:   1.298; rss: 7019MB -> 7071MB (  +52MB)	incr_comp_serialize_result_cache
time:   1.298; rss: 7019MB -> 7071MB (  +52MB)	incr_comp_persist_result_cache
time:   1.298; rss: 7019MB -> 7071MB (  +52MB)	serialize_dep_graph
time:   0.339; rss: 7071MB -> 5014MB (-2057MB)	free_global_ctxt
time: 214.199; rss: 5916MB -> 4364MB (-1552MB)	LLVM_passes
time:   0.001; rss: 1415MB -> 1352MB (  -63MB)	join_worker_thread
time: 161.500; rss: 5014MB -> 1352MB (-3662MB)	finish_ongoing_codegen
time:   0.569; rss:  931MB -> 1054MB ( +123MB)	link_rlib
time:   0.576; rss:  931MB -> 1054MB ( +123MB)	link_binary
time:   0.580; rss:  931MB ->  935MB (   +4MB)	link_crate
time: 162.082; rss: 5014MB ->  935MB (-4079MB)	link
time: 279.744; rss:   32MB ->  365MB ( +333MB)	total

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the reported cargo +nightly rustc -- -Ztime-passes commands against the aws-sdk-ec2 crate and compare the unoptimized and optimized timings. The issue names no source files or tests, and it does not define a requested change or completion criterion beyond recording performance data.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
compilers, performance
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.