Bump FileEncoder buffer size to 64 kB

[Zoxc]

This helps avoid file system overhead on Windows. The improvement are probably reduced a bit on other platforms. Making the buffer size even larger does further improve performance, but that increase memory use further.

Benchmark	Before	After		Before	After		Before	After
	Time	Time	%	Physical Memory	Physical Memory	%	Committed Memory	Committed Memory	%
🟣 clap:check:unchanged	0.3336s	0.3242s	💚 -2.82%	96.97 MiB	97.12 MiB	0.16%	167.87 MiB	167.97 MiB	0.06%
🟣 hyper:check:unchanged	0.1331s	0.1307s	💚 -1.79%	61.92 MiB	62.05 MiB	0.20%	124.66 MiB	124.74 MiB	0.06%
🟣 regex:check:unchanged	0.2485s	0.2399s	💚 -3.45%	78.32 MiB	78.55 MiB	0.29%	145.22 MiB	145.45 MiB	0.15%
🟣 syn:check:unchanged	0.5321s	0.5175s	💚 -2.76%	118.58 MiB	118.77 MiB	0.16%	192.99 MiB	193.17 MiB	0.09%
Total	1.2474s	1.2123s	💚 -2.81%	355.78 MiB	356.49 MiB	0.20%	630.74 MiB	631.32 MiB	0.09%
Summary	1.0000s	0.9729s	💚 -2.71%	1 byte	1.00 bytes	0.20%	1 byte	1.00 bytes	0.09%

[rustbot]

r? @fee1-dead

rustbot has assigned @fee1-dead.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[saethlin]

r? saethlin

@bors try @rust-timer queue

[saethlin]

r=me unless perf indicates this is a regression

[bors]

⌛ Trying commit 74ca129 with merge d6e561e...

[bors]

☀️ Try build successful - checks-actions
Build commit: d6e561e (d6e561e8c338f90cca31c4bb2b674e611584cbdb)

[rust-timer]

Finished benchmarking commit (d6e561e): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

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.

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

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.3%	[0.3%, 0.3%]	2
Improvements ✅ (primary)	-0.3%	[-0.3%, -0.3%]	1
Improvements ✅ (secondary)	-0.3%	[-0.3%, -0.3%]	1
All ❌✅ (primary)	-0.3%	[-0.3%, -0.3%]	1

Max RSS (memory usage)

Results (primary -2.8%, secondary 2.2%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.9%	[1.7%, 5.4%]	8
Improvements ✅ (primary)	-2.8%	[-2.8%, -2.8%]	1
Improvements ✅ (secondary)	-2.4%	[-2.7%, -2.2%]	3
All ❌✅ (primary)	-2.8%	[-2.8%, -2.8%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 783.243s -> 783.248s (0.00%)
Artifact size: 366.21 MiB -> 366.22 MiB (0.00%)

[saethlin]

@bors r+

[bors]

📌 Commit 74ca129 has been approved by saethlin

It is now in the queue for this repository.

[Zoxc]

Maybe MaybeUninit would help with helloworld?

[saethlin]

I don't see a helloworld regression. What are you referring to?

[Zoxc]

There's a suspicious grouping of helloworld instruction regressions, but I don't think FileEncoder is used in the non-incremental builds, so that might be noise.

[saethlin]

Those are very far below the significance threshold. They could as easily be PGO instability, and I don't think it's worth all the time to investigate such tiny changes.

[ChrisDenton]

Is the maximum number of buffers tied to the number of threads or the number of files?

[Zoxc]

The number of files (up to 4), though I intend to add per thread buffers for one of those files.

[bors]

⌛ Testing commit 74ca129 with merge ae06b79...

[bors]

☀️ Test successful - checks-actions
Approved by: saethlin
Pushing ae06b79 to master...

[github-actions]

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing 9ffde4b (parent) -> ae06b79 (this PR)

Test differences

Show 33004 test diffs

Stage 1

• errors::verify_builtin_macros_format_positional_after_named_60: [missing] -> pass (J0)
• errors::verify_const_eval_non_const_intrinsic_19: [missing] -> pass (J0)
• errors::verify_metadata_unsupported_abi_i686_38: [missing] -> pass (J0)
• errors::verify_parse_invalid_attr_unsafe_162: [missing] -> pass (J0)
• errors::verify_parse_modifier_lifetime_136: [missing] -> pass (J0)
• errors::verify_passes_repr_align_greater_than_target_max_74: [missing] -> pass (J0)
• errors::verify_passes_rustc_layout_scalar_valid_range_arg_64: [missing] -> pass (J0)
• lints::verify_lint_map_unit_fn_52: [missing] -> pass (J0)
• owned_slice::tests::boxed: [missing] -> pass (J0)
• parser::tests::minimum_depth: [missing] -> pass (J0)
• spec::tests::x86_64_pc_solaris: [missing] -> pass (J0)
• fs::tests::copy_file_does_not_exist: [missing] -> pass (J1)
• hash::sip::bench_long_str: [missing] -> pass (J1)
• iter::bench_peekable_chain_sum: [missing] -> pass (J1)
• os::unix::net::tests::long_path: [missing] -> pass (J1)
• sort::tests::unstable::stability_cell_i32_random_d2: [missing] -> pass (J1)
• str::contains_bang_str::long_lorem_ipsum: [missing] -> pass (J1)
• tests::exact_filter_match: [missing] -> pass (J1)
• alloc::layout_round_up_to_align_edge_cases: [missing] -> pass (J2)
• collections::hash::map::tests::const_with_hasher: [missing] -> pass (J2)
• collections::vec_deque::tests::make_contiguous_big_head: [missing] -> pass (J2)
• fmt::test_once: [missing] -> pass (J2)
• iter::adapters::flatten::test_flatten_one_shot_arrays: [missing] -> pass (J2)
• iter::adapters::step_by::test_iterator_step_by_nth_try_fold: [missing] -> pass (J2)
• lazy::partialeq_impl: [missing] -> pass (J2)
• nonzero::test_match_option_string: [missing] -> pass (J2)
• num::dec2flt::parse::many_digits: [missing] -> pass (J2)
• num::test_try_i32i16: [missing] -> pass (J2)
• num::test_u16f32: [missing] -> pass (J2)
• rc::test_from_owned: [missing] -> pass (J2)
• slice::test_chunks_nth: [missing] -> pass (J2)
• slice::test_slice_from_ptr_range: [missing] -> pass (J2)
• slice::test_swap_remove: [missing] -> pass (J2)
• str::contains_weird_cases: [missing] -> pass (J2)
• str::test_contains: [missing] -> pass (J2)
• string::test_replace_range_unbounded: [missing] -> pass (J2)
• thin_box::align64small: [missing] -> pass (J2)
• time::mul: [missing] -> pass (J2)
• vec::test_splice_inclusive_out_of_bounds: [missing] -> pass (J2)
• sort::tests::stable::correct_string_random: [missing] -> ignore (J3)
• sort::tests::stable::panic_observable_is_less_saw_mixed: [missing] -> ignore (J3)
• sort::tests::unstable::panic_retain_orig_set_cell_i32_saw_mixed: [missing] -> ignore (J3)
• switch_stdout: [missing] -> ignore (J3)
• f16::test_round: [missing] -> pass (J4)

Stage 2

• analyze_source_file::tests::non_narrow_long: pass -> [missing] (J0)
• coverage::counters::union_find::tests::transitive: pass -> [missing] (J0)
• errors::verify_builtin_macros_cfg_accessible_has_args_15: pass -> [missing] (J0)
• errors::verify_codegen_ssa_extract_bundled_libs_write_file_74: pass -> [missing] (J0)
• errors::verify_mir_build_literal_in_range_out_of_bounds_6: pass -> [missing] (J0)
• errors::verify_mir_build_nan_pattern_21: pass -> [missing] (J0)
• errors::verify_parse_expect_label_found_ident_138: pass -> [missing] (J0)
• errors::verify_parse_unexpected_token_after_struct_name_found_keyword_81: pass -> [missing] (J0)
• errors::verify_passes_should_be_applied_to_struct_enum_15: pass -> [missing] (J0)
• errors::verify_session_unsupported_crate_type_for_target_42: pass -> [missing] (J0)
• html::tests::href_relative_parts_child_module: pass -> [missing] (J0)
• html::url_parts_builder::tests::push_front_empty: pass -> [missing] (J0)
• lints::verify_lint_invalid_asm_label_format_arg_102: pass -> [missing] (J0)
• lints::verify_lint_invalid_null_arguments_46: pass -> [missing] (J0)
• spec::tests::x86_64_pc_cygwin: pass -> [missing] (J0)
• ascii::long::case08_fake_simd_u64: pass -> [missing] (J1)
• btree::set::clone_100_and_drain_all: pass -> [missing] (J1)
• btree::set::clone_100_and_remove_all: pass -> [missing] (J1)
• hash::sip::bench_bytes_4: pass -> [missing] (J1)
• slice::brute_force_rotate_test_0: pass -> [missing] (J1)
• sort::tests::stable::self_cmp_string_ascending: pass -> [missing] (J1)
• sort::tests::unstable::correct_u64_random: pass -> [missing] (J1)
• sort::tests::unstable::violate_ord_retain_orig_set_string_random_s95: pass -> [missing] (J1)
• stats::tests::test_norm10medium: pass -> [missing] (J1)
• tests::run_include_ignored_option: pass -> [missing] (J1)
• vec::bench_next_chunk: pass -> [missing] (J1)
• collections::btree::map::tests::test_into_iter_drop_leak_height_1: pass -> [missing] (J2)
• collections::hash::map::tests::test_index_nonexistent: pass -> [missing] (J2)
• f128::test_num_f128: pass -> [missing] (J2)
• f64::test_nan: pass -> [missing] (J2)
• iter::adapters::array_chunks::test_iterator_array_chunks_infer: pass -> [missing] (J2)
• iter::adapters::step_by::test_iterator_step_by_nth_try_rfold: pass -> [missing] (J2)
• iter::adapters::take::test_iterator_take: pass -> [missing] (J2)
• net::ip_addr::ipv6_from_octets: pass -> [missing] (J2)
• net::socket_addr::is_v6: pass -> [missing] (J2)
• num::i8::test_signum: pass -> [missing] (J2)
• num::test_try_i64i16: pass -> [missing] (J2)
• num::test_try_isizeu8: pass -> [missing] (J2)
• num::u128::test_parse_bytes: pass -> [missing] (J2)
• num::wrapping::test_wrapping_u64: pass -> [missing] (J2)
• rc::test_unique_rc_drops_contents: pass -> [missing] (J2)
• reentrant_lock::trylock_works: pass -> [missing] (J2)
• single_line_formatting: pass -> [missing] (J2)
• slice::memchr::no_match_empty: pass -> [missing] (J2)
• slice::test_chunks_exact_iterator_0: pass -> [missing] (J2)
• slice::test_copy_from_slice_dst_shorter: pass -> [missing] (J2)
• slice::test_from_fn: pass -> [missing] (J2)
• slice::test_len_divzero: pass -> [missing] (J2)
• slice::test_rchunks_count: pass -> [missing] (J2)
• sort::tests::stable::correct_i32_random_s95: pass -> [missing] (J2)
• thin_box::align2big: pass -> [missing] (J2)
• vec::test_drain_range: pass -> [missing] (J2)
• sort::tests::stable::deterministic_i32_descending: ignore -> [missing] (J3)
• sort::tests::unstable::deterministic_string_descending: ignore -> [missing] (J3)
• sort::tests::unstable::self_cmp_cell_i32_random: ignore -> [missing] (J3)
• sort::tests::unstable::self_cmp_i32_random: ignore -> [missing] (J3)

(and 16404 additional test diffs)

Additionally, 16500 doctest diffs were found. These are ignored, as they are noisy.

Job group index

• J0: aarch64-apple, x86_64-apple-1
• J1: aarch64-apple, test-various, x86_64-apple-1
• J2: aarch64-apple, test-various, x86_64-apple-1, x86_64-gnu-aux
• J3: x86_64-gnu-aux
• J4: aarch64-apple

Job duration changes

1. x86_64-apple-2: 1271.5s -> 4189.9s (229.5%)
2. test-various: 2272.4s -> 4162.4s (83.2%)
3. x86_64-gnu-aux: 4289.5s -> 6079.2s (41.7%)
4. aarch64-apple: 2877.3s -> 3553.3s (23.5%)
5. dist-x86_64-apple: 10760.2s -> 8345.0s (-22.4%)
6. x86_64-apple-1: 7095.9s -> 7841.3s (10.5%)
7. dist-apple-various: 7312.4s -> 6670.8s (-8.8%)
8. aarch64-gnu-debug: 4310.8s -> 4079.7s (-5.4%)
9. i686-gnu-nopt-1: 8587.7s -> 8167.8s (-4.9%)
10. dist-aarch64-linux: 5233.4s -> 5488.1s (4.9%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

[rust-timer]

Finished benchmarking commit (ae06b79): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

@rustbot label: -perf-regression

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.4%	[0.4%, 0.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.3%, -0.3%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary -1.0%, secondary 1.2%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.5%	[2.2%, 5.6%]	7
Improvements ✅ (primary)	-1.0%	[-1.0%, -1.0%]	1
Improvements ✅ (secondary)	-4.0%	[-8.4%, -1.3%]	3
All ❌✅ (primary)	-1.0%	[-1.0%, -1.0%]	1

Cycles

Results (secondary -2.6%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.6%	[-2.6%, -2.6%]	1
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 777.191s -> 777.2s (0.00%)
Artifact size: 365.50 MiB -> 365.49 MiB (-0.00%)