[TEST] Test out Fjall memory allocator

[Zoxc]

This tests out my Rust port of mimalloc, Fjall.

[Zoxc]

@bors try @rust-timer queue

[rust-bors]

💔 Test for 3bde676 failed: CI. Failed job:

• try - dist-x86_64-linux-quick (web logs, enhanced plaintext logs)

[Zoxc]

@bors try @rust-timer queue

[rust-log-analyzer]

The job aarch64-gnu-llvm-21-2 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

Executing "/scripts/stage_2_test_set2.sh"
+ /scripts/stage_2_test_set2.sh
PR_CI_JOB set; skipping tidy
+ '[' 1 == 1 ']'
+ echo 'PR_CI_JOB set; skipping tidy'
+ SKIP_TIDY='--skip tidy'
+ ../x.py --stage 2 test --skip tidy --skip tests --skip coverage-map --skip coverage-run --skip library --skip tidyselftest
##[group]Building bootstrap
    Finished `dev` profile [unoptimized] target(s) in 0.04s
##[endgroup]
downloading https://static.rust-lang.org/dist/2026-04-14/rustfmt-nightly-aarch64-unknown-linux-gnu.tar.xz
---
[TIMING:start] compile::Rustc { target: aarch64-unknown-linux-gnu, build_compiler: Compiler { stage: 0, host: aarch64-unknown-linux-gnu, forced_compiler: false }, crates: [] }
[TIMING:start] compile::StdLink { compiler: Compiler { stage: 0, host: aarch64-unknown-linux-gnu, forced_compiler: false }, target_compiler: Compiler { stage: 0, host: aarch64-unknown-linux-gnu, forced_compiler: false }, target: aarch64-unknown-linux-gnu, crates: [], force_recompile: false }
[TIMING:end] compile::StdLink { compiler: Compiler { stage: 0, host: aarch64-unknown-linux-gnu, forced_compiler: false }, target_compiler: Compiler { stage: 0, host: aarch64-unknown-linux-gnu, forced_compiler: false }, target: aarch64-unknown-linux-gnu, crates: [], force_recompile: false } -- 0.001
##[group]Building stage1 compiler artifacts (stage0 -> stage1, aarch64-unknown-linux-gnu)
    Updating git repository `https://github.com/Zoxc/fjall.git`
    Updating crates.io index
---
   Compiling ctrlc v3.5.1
[RUSTC-TIMING] shlex test:false 0.212
   Compiling rustc-main v0.0.0 (/checkout/compiler/rustc)
[RUSTC-TIMING] pulldown_cmark test:false 2.777
   Compiling fjall v0.1.0 (https://github.com/Zoxc/fjall.git#955dd1c4)
[RUSTC-TIMING] build_script_build test:false 0.196
[RUSTC-TIMING] ctrlc test:false 0.082
[RUSTC-TIMING] rustc_errors test:false 4.698
   Compiling rustc_abi v0.0.0 (/checkout/compiler/rustc_abi)
[RUSTC-TIMING] fjall test:false 0.732
---
   Compiling rustc_driver_impl v0.0.0 (/checkout/compiler/rustc_driver_impl)
[RUSTC-TIMING] jiff test:false 19.741
   Compiling rustc-main v0.0.0 (/checkout/compiler/rustc)
[RUSTC-TIMING] build_script_build test:false 0.164
   Compiling fjall v0.1.0 (https://github.com/Zoxc/fjall.git#955dd1c4)
[RUSTC-TIMING] fjall test:false 2.063
[RUSTC-TIMING] rustc_interface test:false 49.776
[RUSTC-TIMING] rustc_driver_impl test:false 28.608
[RUSTC-TIMING] rustc_query_impl test:false 116.548
[RUSTC-TIMING] rustc_driver test:false 9.268
[RUSTC-TIMING] rustc_main test:false 0.450
---
double free or corruption (out)
error: test failed, to rerun pass `-p proc-macro-srv-cli --test bidirectional_postcard`

Caused by:
  process didn't exit successfully: `/checkout/obj/build/aarch64-unknown-linux-gnu/stage2-tools/aarch64-unknown-linux-gnu/release/deps/bidirectional_postcard-7fde1f13f25085a6 '--skip=tests::smoke_test_real_sysroot_cargo' --skip=check_code_formatting -Z unstable-options --format json` (signal: 6, SIGABRT: process abort signal)
Bootstrap failed while executing `--stage 2 test --skip tidy --skip tests --skip coverage-map --skip coverage-run --skip library --skip tidyselftest`
Build completed unsuccessfully in 0:53:00
  local time: Mon Apr 20 07:12:19 UTC 2026
  network time: Mon, 20 Apr 2026 07:12:19 GMT
##[error]Process completed with exit code 1.

[rust-bors]

☀️ Try build successful (CI)
Build commit: 8221642 (82216424993ffcbd61fdcb5ea91da2bf464dd694, parent: e22c616e4e87914135c1db261a03e0437255335e)

[rust-timer]

Finished benchmarking commit (8221642): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

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

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.4%	[0.0%, 0.7%]	7
Improvements ✅ (primary)	-3.6%	[-10.8%, -1.3%]	269
Improvements ✅ (secondary)	-4.5%	[-11.5%, -0.1%]	297
All ❌✅ (primary)	-3.6%	[-10.8%, -1.3%]	269

Max RSS (memory usage)

Results (primary 6.3%, secondary 4.3%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	6.3%	[1.1%, 24.2%]	244
Regressions ❌ (secondary)	6.5%	[0.7%, 62.0%]	223
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-8.0%	[-23.1%, -0.7%]	40
All ❌✅ (primary)	6.3%	[1.1%, 24.2%]	244

Cycles

Results (primary 1.5%, secondary -3.2%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	2.9%	[1.4%, 4.6%]	45
Regressions ❌ (secondary)	4.3%	[2.0%, 14.3%]	13
Improvements ✅ (primary)	-3.4%	[-5.6%, -2.1%]	12
Improvements ✅ (secondary)	-5.8%	[-13.9%, -1.1%]	38
All ❌✅ (primary)	1.5%	[-5.6%, 4.6%]	57

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 507.073s -> 504.881s (-0.43%)
Artifact size: 394.31 MiB -> 396.80 MiB (0.63%)

[bjorn3]

Instructions are very positive. Wall time (and task clock) is rather negative.

[lqd]

Here's a mimalloc run from a year ago https://perf.rust-lang.org/compare.html?start=6c8347b9588a302afebb81a1ae6daa64ec37abd0&end=2ae4c6dae2c8206758e06df89d1da08c5eb52f11&stat=instructions%3Au for comparison

[Zoxc]

We should give mimalloc v3 a run too. A pre-lease version was slower vs. v2 on Windows, but that may have changed or it could be platform specific.