Update to LLVM 20.1.1 #138695

nikic · 2025-03-19T09:47:16Z

rustbot · 2025-03-19T09:47:21Z

rustbot has assigned @cuviper.
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

rustbot · 2025-03-19T09:47:22Z

⚠️ Warning ⚠️

These commits modify submodules.

dianqk · 2025-03-19T12:12:21Z

@bors r+ rollup=never

bors · 2025-03-19T12:12:24Z

📌 Commit 80f8f00 has been approved by dianqk

It is now in the queue for this repository.

bors · 2025-03-19T12:12:25Z

🌲 The tree is currently closed for pull requests below priority 100. This pull request will be tested once the tree is reopened.

jieyouxu · 2025-03-20T03:54:42Z

This fixes two P-high regressions, so I'm bumping its priority for rollup scheduling.
@bors p=5

tgross35 · 2025-03-20T05:47:23Z

I think this should also fix #137873 (via llvm/llvm-project#129546)

bors · 2025-03-20T07:27:48Z

⌛ Testing commit 80f8f00 with merge 87e60a7...

bors · 2025-03-20T13:06:29Z

☀️ Test successful - checks-actions
Approved by: dianqk
Pushing 87e60a7 to master...

github-actions · 2025-03-20T13:08:22Z

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 4e2b096 (parent) -> 87e60a7 (this PR)

Test differences

Show 12 test diffs

[assembly] tests/assembly/riscv-soft-abi-with-float-features.rs#LLVM-PRE-20 (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 19) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 19) (J0)
[assembly] tests/assembly/x86_64-bigint-helpers.rs#llvm-pre-20 (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 19) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 19) (J0)
[codegen] tests/codegen/branch-protection-old-llvm.rs#BKEY (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/branch-protection-old-llvm.rs#BTI (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/branch-protection-old-llvm.rs#LEAF (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/branch-protection-old-llvm.rs#NONE (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/branch-protection-old-llvm.rs#PACRET (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/call-metadata.rs (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[codegen] tests/codegen/integer-cmp.rs#llvm-pre-20 (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 19) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 19) (J0)
[codegen] tests/codegen/vec_pop_push_noop.rs#llvm-pre-19 (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 18) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 18) (J0)
[assembly] tests/assembly/x86_64-cmp.rs#LLVM-PRE-20-OPTIM (stage 2): ignore (ignored when the LLVM version (20.1.0) is newer than majorversion 19) -> ignore (ignored when the LLVM version (20.1.1) is newer than majorversion 19) (J1)
[run-make] tests/run-make/compressed-debuginfo-zstd (stage 2): pass -> ignore (ignored if LLVM wasn't build with zstd for ELF section compression (we want LLVM/LLD to be built with zstd support)) (J2)

Job group index

J0: aarch64-apple, aarch64-gnu, arm-android, armhf-gnu, dist-i586-gnu-i586-i686-musl, i686-gnu-1, i686-gnu-nopt-1, i686-msvc-1, test-various, x86_64-apple-1, x86_64-gnu, x86_64-gnu-nopt, x86_64-gnu-stable, x86_64-mingw-1, x86_64-msvc-1
J1: test-various, x86_64-apple-1, x86_64-gnu, x86_64-gnu-nopt, x86_64-gnu-stable, x86_64-mingw-1, x86_64-msvc-1
J2: x86_64-gnu-nopt

nikic · 2025-03-20T13:39:36Z

@Kobzol Probably shouldn't show the cases where only the ignore message differs? (Or maybe normalize the version).

[run-make] tests/run-make/compressed-debuginfo-zstd (stage 2): pass -> ignore (ignored if LLVM wasn't build with zstd for ELF section compression (we want LLVM/LLD to be built with zstd support)) (J2)

This might be an issue with download-ci-llvm, where we get a different config (with/without zstd) depending on whether we download or build llvm.

Kobzol · 2025-03-20T13:48:37Z

Probably shouldn't show the cases where only the ignore message differs? (Or maybe normalize the version).

I think that seeing the difference in the ignore message is fine, it could be important sometimes (e.g. a test is now ignored for a completely different reason than before).

This might be an issue with download-ci-llvm, where we get a different config (with/without zstd) depending on whether we download or build llvm.

This sounds more worrying, tbh, we should ideally have exactly the same config when building or downloading LLVM 😅

rust-timer · 2025-03-20T14:23:22Z

Finished benchmarking commit (87e60a7): comparison URL.

Overall result: ✅ 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
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.2%	[-0.2%, -0.2%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (secondary -2.1%)

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)	-2.1%	[-2.6%, -1.6%]	2
All ❌✅ (primary)	-	-	0

Cycles

Results (primary 3.1%)

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)	3.1%	[3.1%, 3.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.1%	[3.1%, 3.1%]	1

Binary size

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

Bootstrap: 773.468s -> 773.84s (0.05%)
Artifact size: 365.48 MiB -> 365.50 MiB (0.01%)

The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit 5cf417a. [1]: llvm/llvm-project#129394 [2]: rust-lang/rust#138695

The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695

The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit 5cf417a. [1]: llvm/llvm-project#129394 [2]: rust-lang/rust#138695

…<try> Revert "Disable `f16` on Aarch64 without `neon`" The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695 try-job: aarch64-gnu try-job: aarch64-gnu-debug try-job: armhf-gnu try-job: dist-various-1

…r=Mark-Simulacrum Revert "Disable `f16` on Aarch64 without `neon`" The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695 try-job: aarch64-gnu try-job: aarch64-gnu-debug try-job: armhf-gnu try-job: dist-various-1

Rollup merge of rust-lang#139276 - tgross35:enable-f16-without-neon, r=Mark-Simulacrum Revert "Disable `f16` on Aarch64 without `neon`" The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695 try-job: aarch64-gnu try-job: aarch64-gnu-debug try-job: armhf-gnu try-job: dist-various-1

The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695

…r=Mark-Simulacrum Revert "Disable `f16` on Aarch64 without `neon`" The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit c51b229. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695 try-job: aarch64-gnu try-job: aarch64-gnu-debug try-job: armhf-gnu try-job: dist-various-1

The LLVM issue [1] was resolved and the fix was synced to rust-lang/rust in [2]. This reverts commit 5cf417a9e92bb48e4e55756a645826fd167b9f3a. [1]: llvm/llvm-project#129394 [2]: rust-lang#138695

Update to LLVM 20.1.1

80f8f00

rustbot assigned cuviper Mar 19, 2025

rustbot added the S-waiting-on-review label Mar 19, 2025

jieyouxu mentioned this pull request Mar 19, 2025

Use /usr/bin/strip on macOS -> iOS cross as a temporary workaround #138250

Closed

bors added S-waiting-on-bors and removed S-waiting-on-review labels Mar 19, 2025

dianqk added T-compiler S-waiting-on-review and removed S-waiting-on-bors labels Mar 19, 2025

dianqk assigned dianqk and unassigned cuviper Mar 19, 2025

bors added the merged-by-bors label Mar 20, 2025

bors merged commit 87e60a7 into rust-lang:master Mar 20, 2025
7 checks passed

rustbot added this to the 1.87.0 milestone Mar 20, 2025

tgross35 mentioned this pull request Apr 2, 2025

Revert "Disable f16 on AArch64 without the neon feature" rust-lang/compiler-builtins#809

Merged

tgross35 mentioned this pull request Apr 2, 2025

Revert "Disable f16 on Aarch64 without neon" #139276

Merged

workingjubilee added the A-LLVM label May 26, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Update to LLVM 20.1.1 #138695

Update to LLVM 20.1.1 #138695

nikic commented Mar 19, 2025

Uh oh!

rustbot commented Mar 19, 2025

Uh oh!

rustbot commented Mar 19, 2025

Uh oh!

dianqk commented Mar 19, 2025

Uh oh!

bors commented Mar 19, 2025

Uh oh!

bors commented Mar 19, 2025

Uh oh!

jieyouxu commented Mar 20, 2025

Uh oh!

tgross35 commented Mar 20, 2025

Uh oh!

bors commented Mar 20, 2025

Uh oh!

bors commented Mar 20, 2025

Uh oh!

Uh oh!

github-actions bot commented Mar 20, 2025

Uh oh!

nikic commented Mar 20, 2025

Uh oh!

Kobzol commented Mar 20, 2025

Uh oh!

rust-timer commented Mar 20, 2025

Uh oh!

Update to LLVM 20.1.1 #138695

Update to LLVM 20.1.1 #138695

Conversation

nikic commented Mar 19, 2025

Uh oh!

rustbot commented Mar 19, 2025

Uh oh!

rustbot commented Mar 19, 2025

Uh oh!

dianqk commented Mar 19, 2025

Uh oh!

bors commented Mar 19, 2025

Uh oh!

bors commented Mar 19, 2025

Uh oh!

jieyouxu commented Mar 20, 2025

Uh oh!

tgross35 commented Mar 20, 2025

Uh oh!

bors commented Mar 20, 2025

Uh oh!

bors commented Mar 20, 2025

Uh oh!

Uh oh!

github-actions bot commented Mar 20, 2025

Test differences

Uh oh!

nikic commented Mar 20, 2025

Uh oh!

Kobzol commented Mar 20, 2025

Uh oh!

rust-timer commented Mar 20, 2025

Overall result: ✅ improvements - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!