GVN: Only propagate borrows from SSA locals #150485

dianqk · 2025-12-29T14:11:18Z

Fixes #141313. This is a more principled fix than #147886.

Using a reference that is not a borrowing of an SSA local at a new location may be UB.

The PR has two major changes.
The first one, if we cannot guard to the reference point to an SSA local in visit_assign, we have to rewrite the value to opaque. This avoids unifying the following dereferences that are new references:

let b: &T = *a;
// ... a is allowed to be modified. c and b have different borrowing lifetime.
let c: &T = *a;

This still allows unifying non-reference dereferences:

let a: &T = ...;
let b: T = *a;
// ... a is NOT allowed to be modified.
let c: T = *a;

The second one, when introducing a new dereference at a new location, is that the reference must point to an SSA local or be an immutable argument.

r? @cjgillot

dianqk · 2026-01-01T23:22:18Z

@bors try @rust-timer queue

[DRAFT] GVN: Only propagate borrows from SSA-locals

rust-bors · 2026-01-02T01:47:59Z

☀️ Try build successful (CI)
Build commit: f23e5de (f23e5de34a174f8bd281d69038b68bf6332b1af6, parent: fcd630976c460c819c4bbcaf107d0c94501205d8)

rust-timer · 2026-01-02T02:41:15Z

Finished benchmarking commit (f23e5de): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.3%	[0.1%, 0.9%]	20
Regressions ❌ (secondary)	0.4%	[0.1%, 2.0%]	27
Improvements ✅ (primary)	-0.6%	[-1.2%, -0.2%]	4
Improvements ✅ (secondary)	-0.2%	[-0.4%, -0.1%]	7
All ❌✅ (primary)	0.1%	[-1.2%, 0.9%]	24

Max RSS (memory usage)

Results (primary -1.1%, secondary -4.7%)

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

	mean	range	count
Regressions ❌ (primary)	2.4%	[1.9%, 2.8%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-3.4%	[-6.8%, -0.6%]	3
Improvements ✅ (secondary)	-4.7%	[-6.5%, -3.7%]	3
All ❌✅ (primary)	-1.1%	[-6.8%, 2.8%]	5

Cycles

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

Binary size

Results (primary 0.1%, secondary 0.0%)

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

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.0%, 6.9%]	11
Regressions ❌ (secondary)	0.1%	[0.0%, 0.2%]	14
Improvements ✅ (primary)	-0.1%	[-1.0%, -0.0%]	27
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.0%]	5
All ❌✅ (primary)	0.1%	[-1.0%, 6.9%]	38

Bootstrap: 475.254s -> 472.456s (-0.59%)
Artifact size: 390.85 MiB -> 390.82 MiB (-0.01%)

dianqk · 2026-01-02T14:21:44Z

@bors try parent=fcd630976c460c819c4bbcaf107d0c94501205d8 @rust-timer queue

[DRAFT] GVN: Only propagate borrows from SSA-locals

rust-bors · 2026-01-02T16:45:18Z

☀️ Try build successful (CI)
Build commit: 2183340 (218334063dc5d70b7f9d69202bf22750167c6cbf, parent: fcd630976c460c819c4bbcaf107d0c94501205d8)

rust-timer · 2026-01-02T19:12:15Z

Finished benchmarking commit (2183340): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.3%	[0.1%, 0.9%]	17
Regressions ❌ (secondary)	0.3%	[0.1%, 0.8%]	23
Improvements ✅ (primary)	-0.6%	[-1.2%, -0.2%]	4
Improvements ✅ (secondary)	-0.2%	[-0.4%, -0.1%]	8
All ❌✅ (primary)	0.1%	[-1.2%, 0.9%]	21

Max RSS (memory usage)

Results (primary -1.9%)

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

	mean	range	count
Regressions ❌ (primary)	1.7%	[1.7%, 1.7%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-3.0%	[-4.6%, -0.6%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-1.9%	[-4.6%, 1.7%]	4

Cycles

Results (secondary -2.3%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.3%	[-2.3%, -2.3%]	1
All ❌✅ (primary)	-	-	0

Binary size

Results (primary 0.2%, secondary 0.1%)

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

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.0%, 6.9%]	11
Regressions ❌ (secondary)	0.1%	[0.0%, 0.2%]	14
Improvements ✅ (primary)	-0.2%	[-0.9%, -0.0%]	11
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.0%]	5
All ❌✅ (primary)	0.2%	[-0.9%, 6.9%]	22

Bootstrap: 475.254s -> 472.783s (-0.52%)
Artifact size: 390.85 MiB -> 390.84 MiB (-0.00%)

cjgillot · 2026-01-18T14:49:45Z

compiler/rustc_mir_transform/src/gvn.rs

-            self.values.forget(deref);
-        }
-    }
-


Glad to see this gone.

compiler/rustc_mir_transform/src/gvn.rs

cjgillot · 2026-01-18T14:54:56Z

compiler/rustc_mir_transform/src/gvn.rs

+                }
+            }
+        }
+


Should this go in try_as_place?

Done. This introduces a small regression that reborrows cannot be propagated, but I think ReferencePropagation has done this.
And &(*_2) can be copy _2, since reborrows extend the lifetime of borrows. This can be a subsequent PR if needed.

cjgillot · 2026-01-18T14:57:56Z

compiler/rustc_mir_transform/src/gvn.rs

+                | Value::RuntimeChecks(..)
+                | Value::UnaryOp(..) => unreachable!(),
+            }
        }


I don't understand what this does. Why are some values unreachable? A constant and a cast can be a reference, can't it?

They are not marked as unreachable. The new change should be clearer.

…borrow from non-SSA locals

dianqk · 2026-01-20T13:26:40Z

tests/mir-opt/dont_reset_cast_kind_without_updating_operand.test.GVN.32bit.panic-abort.diff

 +         nop;
-+         _10 = copy (*_2);
+          _10 = copy (*_1);


Can be _10 = copy (*_2);.

dianqk · 2026-01-20T13:28:29Z

compiler/rustc_mir_transform/src/gvn.rs

+                Value::Projection(..) => {
+                    value = None;
+                }
+                Value::Opaque(_)


The latest change allows the opaque value to be propagated. For let a: &T = b;, a extends the lifetime of b.

dianqk · 2026-01-20T22:24:46Z

@bors try parent=fe98ddcfcfb6f185dbf4adeaf439d8a756da0273 @rust-timer queue

GVN: Only propagate borrows from SSA locals

rust-bors · 2026-01-21T00:40:52Z

☀️ Try build successful (CI)
Build commit: 9e686b3 (9e686b35643ef3cd283a81a855b9fdcb961addca, parent: fe98ddcfcfb6f185dbf4adeaf439d8a756da0273)

rust-timer · 2026-01-21T01:49:27Z

Finished benchmarking commit (9e686b3): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.4%	[0.2%, 0.8%]	5
Regressions ❌ (secondary)	0.3%	[0.0%, 1.2%]	26
Improvements ✅ (primary)	-0.5%	[-1.0%, -0.2%]	5
Improvements ✅ (secondary)	-0.2%	[-0.4%, -0.0%]	7
All ❌✅ (primary)	-0.0%	[-1.0%, 0.8%]	10

Max RSS (memory usage)

Results (primary -1.1%, secondary 2.8%)

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

	mean	range	count
Regressions ❌ (primary)	2.5%	[2.5%, 2.5%]	1
Regressions ❌ (secondary)	4.0%	[1.6%, 5.6%]	8
Improvements ✅ (primary)	-2.3%	[-2.8%, -1.8%]	3
Improvements ✅ (secondary)	-2.2%	[-3.2%, -1.2%]	2
All ❌✅ (primary)	-1.1%	[-2.8%, 2.5%]	4

Cycles

Results (secondary -1.7%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	5.9%	[4.2%, 7.6%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-5.5%	[-9.5%, -2.1%]	4
All ❌✅ (primary)	-	-	0

Binary size

Results (primary 0.2%, secondary 0.0%)

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

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.0%, 6.9%]	16
Regressions ❌ (secondary)	0.1%	[0.1%, 0.1%]	9
Improvements ✅ (primary)	-0.3%	[-1.0%, -0.0%]	12
Improvements ✅ (secondary)	-0.1%	[-0.2%, -0.0%]	7
All ❌✅ (primary)	0.2%	[-1.0%, 6.9%]	28

Bootstrap: 473.229s -> 473.099s (-0.03%)
Artifact size: 383.52 MiB -> 383.50 MiB (-0.00%)

dianqk · 2026-01-21T13:55:28Z

The new change may introduce more regressions.
@bors try parent=fe98ddcfcfb6f185dbf4adeaf439d8a756da0273 @rust-timer queue

GVN: Only propagate borrows from SSA locals

rust-bors · 2026-01-21T16:12:08Z

☀️ Try build successful (CI)
Build commit: 8eb4b3c (8eb4b3c6281cc5b61af4ae4a0bc2fe2b95546c08, parent: fe98ddcfcfb6f185dbf4adeaf439d8a756da0273)

rust-timer · 2026-01-21T16:52:27Z

Finished benchmarking commit (8eb4b3c): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.4%	[0.2%, 0.8%]	10
Regressions ❌ (secondary)	0.3%	[0.0%, 1.3%]	19
Improvements ✅ (primary)	-0.3%	[-0.4%, -0.2%]	5
Improvements ✅ (secondary)	-0.2%	[-0.4%, -0.0%]	7
All ❌✅ (primary)	0.1%	[-0.4%, 0.8%]	15

Max RSS (memory usage)

Results (primary 3.2%, secondary 2.5%)

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

	mean	range	count
Regressions ❌ (primary)	6.5%	[2.8%, 10.5%]	3
Regressions ❌ (secondary)	3.7%	[1.6%, 5.2%]	8
Improvements ✅ (primary)	-1.9%	[-1.9%, -1.9%]	2
Improvements ✅ (secondary)	-2.3%	[-3.5%, -1.1%]	2
All ❌✅ (primary)	3.2%	[-1.9%, 10.5%]	5

Cycles

Results (primary -3.0%, secondary -2.1%)

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

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	5.0%	[3.8%, 6.2%]	2
Improvements ✅ (primary)	-3.0%	[-3.0%, -3.0%]	1
Improvements ✅ (secondary)	-5.7%	[-9.6%, -2.3%]	4
All ❌✅ (primary)	-3.0%	[-3.0%, -3.0%]	1

Binary size

Results (primary 0.1%, secondary -0.0%)

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

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.0%, 6.9%]	18
Regressions ❌ (secondary)	0.1%	[0.1%, 0.1%]	9
Improvements ✅ (primary)	-0.1%	[-1.5%, -0.0%]	35
Improvements ✅ (secondary)	-0.1%	[-0.2%, -0.0%]	10
All ❌✅ (primary)	0.1%	[-1.5%, 6.9%]	53

Bootstrap: 473.229s -> 471.231s (-0.42%)
Artifact size: 383.52 MiB -> 383.43 MiB (-0.03%)

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Dec 29, 2025

dianqk mentioned this pull request Dec 29, 2025

GVN: Use the borrows only if they are always live #147886

Closed

This comment has been minimized.

Sign in to view

dianqk force-pushed the gvn-ssa-borrow branch from af2bcaf to 45cd528 Compare January 1, 2026 14:38

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Jan 1, 2026

Auto merge of #150485 - dianqk:gvn-ssa-borrow, r=<try>

f23e5de

[DRAFT] GVN: Only propagate borrows from SSA-locals

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 1, 2026

This comment has been minimized.

Sign in to view

rustbot added perf-regression Performance regression. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels Jan 2, 2026

dianqk force-pushed the gvn-ssa-borrow branch from 45cd528 to c090eb9 Compare January 2, 2026 12:34

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Jan 2, 2026

Auto merge of #150485 - dianqk:gvn-ssa-borrow, r=<try>

2183340

[DRAFT] GVN: Only propagate borrows from SSA-locals

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 2, 2026

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 2, 2026

dianqk mentioned this pull request Jan 3, 2026

GVN: Avoid merging borrows from the dereferenced value #150270

Closed

dianqk added 2 commits January 17, 2026 21:17

Add miscompiled test cases for GVN

b8d08bb

GVN: Remove invalidate_derefs

e312df7

dianqk force-pushed the gvn-ssa-borrow branch from c090eb9 to e9cd151 Compare January 17, 2026 14:05

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Jan 18, 2026

dianqk force-pushed the gvn-ssa-borrow branch from c6dd1f7 to d3e9dc0 Compare January 18, 2026 14:12

dianqk changed the title ~~[DRAFT] GVN: Only propagate borrows from SSA-locals~~ GVN: Only propagate borrows from SSA-locals Jan 18, 2026

dianqk changed the title ~~GVN: Only propagate borrows from SSA-locals~~ [DRAFT] GVN: Only propagate borrows from SSA locals Jan 18, 2026

dianqk changed the title ~~[DRAFT] GVN: Only propagate borrows from SSA locals~~ GVN: Only propagate borrows from SSA locals Jan 18, 2026

cjgillot reviewed Jan 18, 2026

View reviewed changes

GVN: Do not introduce new dereferences or unify dereferences if they …

30561d9

…borrow from non-SSA locals

dianqk force-pushed the gvn-ssa-borrow branch from d3e9dc0 to 30561d9 Compare January 20, 2026 13:17

dianqk commented Jan 20, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

rust-bors bot pushed a commit that referenced this pull request Jan 20, 2026

Auto merge of #150485 - dianqk:gvn-ssa-borrow, r=<try>

9e686b3

GVN: Only propagate borrows from SSA locals

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 20, 2026

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 21, 2026

dianqk force-pushed the gvn-ssa-borrow branch from a5c2c72 to 724eb83 Compare January 21, 2026 13:54

This comment has been minimized.

Sign in to view

rust-bors bot pushed a commit that referenced this pull request Jan 21, 2026

Auto merge of #150485 - dianqk:gvn-ssa-borrow, r=<try>

8eb4b3c

GVN: Only propagate borrows from SSA locals

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 21, 2026

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jan 21, 2026

dianqk force-pushed the gvn-ssa-borrow branch from 724eb83 to 30561d9 Compare January 22, 2026 00:29

Uh oh!

GVN: Only propagate borrows from SSA locals #150485

Are you sure you want to change the base?

GVN: Only propagate borrows from SSA locals #150485

Uh oh!

Conversation

dianqk commented Dec 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

dianqk commented Jan 1, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Jan 2, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jan 2, 2026

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

dianqk commented Jan 2, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Jan 2, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jan 2, 2026

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

This comment has been minimized.

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dianqk commented Jan 20, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Jan 21, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jan 21, 2026

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

dianqk commented Jan 21, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Jan 21, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jan 21, 2026

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

dianqk commented Dec 29, 2025 •

edited

Loading